Peo- ple began to wonder how they perceived the world, thus
questioning
per- ception in the mode of second-order observation, but only to create paint- ings that imitate nature not only with regard to its "what" but to its "how.
Niklas Luhmann - Art of the Social System
It liberates art, to a large extent, from the demands of consensus.
The sameness of the object sub- stitutes for the conformity of opinion.
Without losing contact with the artist's formal decisions, the beholder can arrive at judgments, valuations, or experiences that radically diverge from the artist's.
One stays with the forms created by the artist while perceiving things other than what he in- tended to express.
Likewise, when producing for other observers, the artist
(This is why the
Observation of the First and of the Second Order 75
need not surrender to their gaze or become dependent on them. The artist knows that his decisions remain his own; he can go about his work au- thentically and leave it to the observer to form his own judgment.
In this way, judgment is released from the constraints of consensus,
while its relation to the object is preserved. This is worth emphasizing be-
cause it flies in the face of widely held notions concerning the conditions
of social communication. Once the old European idea of a natural corre-
spondence between art and nature became obsolete, one began to count
on consensus. This shift is evident in the social contract doctrines of the
seventeenth and eighteenth centuries, and even more so in Enlightenment
premises of a public circulation of ideas and rationally disciplined control
42
of opinions.
stantial consensus (of opinions) leads to complications that surface in cer- tain aberrations in Enlightenment thematizations of art. On the one hand, the debate about good and bad taste failed to come up with the desired criteria. Instead, it drove home the point that all so-called objective crite- ria have the effect of social discrimination; that is to say, those who expe- rience differently find themselves in bad company. On the other hand, the entire realm of art is degraded because it is contaminated with sensory ex- perience and compromised by its dependence on inferior types of cogni- tion. In a state of turmoil, society decides to search for consensus, general- izing as transcendental a prioris or as new mythologies to be expected from the future only those symbols capable of binding each and every subject. In the meantime, one puts up with ideologies.
Today the realization that communication is coordinated by objects
43
The one-sided understanding of society as a kind of sub-
rather than justifications is gaining ground.
Dissent in the realm of jus-
tifications is tolerable so long as adjustments mediated by objects are at
work. This situation implies that bodies can be treated like phenomena,
without understanding the biochemistry of their lives, the neurophysiol-
ogy of their brain processes, or the conscious states actualized at any given
44
moment.
for consensus and ensure that object orientation will have its proper place. This has at least one significant advantage. It releases further communica- tion from its fixation on any given topic and leaves it to communication to decide whether or not opinions are at stake, and if so, how serious and binding they are meant to be.
All of this considerably affects our understanding of art as a form for second-order observations. Art permits a kind of playful relationship to
Limited resources prevent society from overestimating its need
j6 Observation of the First and of the Second Order
questions of reasonable consensus or dissent. It avoids degrading or ex- cluding those who think differently. And it does so in such a way that doubts about whether or not one communicates about the same thing never arise. This is not to say that art does not place high (and exclusive) demands on an observation that seeks to be adequate. The measure for adequacy is not a consensus determined by a "shared symbolic system" (Parsons), but resides instead in the question of whether the viewer can follow the directives for adequate observation embedded in the work's own formal decisions.
VI
Traditionally, the theory of art and literature did not describe the rela-
tionship between artist and observer (author and reader) in terms of an
observational relation. Instead, it assumed a causal relation, an effectua-
tion of effects. Accordingly, the artist was believed to be interested in elic-
iting a certain impression in the beholder, an endeavor at which he could
be more or less successful. The modern critique of this theoretical con-
stellation led to the discovery of the observer of art and, in literary theory,
to the demand that the texts be understood from the perspective of the
45
reader.
it fails to provide a sufficiently adequate theory of art (of the artwork, the text). We must assume that the author of an artwork adapts to the be- holder in the same way as an observer anticipates another observer, and that the artwork must not only mediate between diverging observational modes as they arise but also needs to generate such diverging perspectives to begin with. This is why the demise of the causal theory of art calls for a theory of second-order observation.
Many endeavors point in this direction. The widely popular "symbol-
ism" in twentieth-century literature, for example, can be taken to imply
that every interpretation, including the author's own, imposes limita-
46
tions.
petence. In a series of significant phenomenological investigations, Ro- man Ingarden has called attention to the "blanks" embedded in literary works. Such blanks both assume and require an independent "concretiza-
47
tion" on the part of the reader. An observer can perceive only schemati-
cally; he cannot simultaneously see the front and the back of the same ob- ject. But he can check his speculations against reality and find out if the
While this shift is a plausible reaction to the causal theory of art,
This may be a calculated effect aimed at the author's own incom-
Observation of the First and of the Second Order JJ
rear of a red sphere is as round, smooth, and red, and so forth as its front. In art, this kind of reality check is not possible. The viewer needs to con- jure up the necessary completion (and which one would that be? ). The creator of the work or the author of the text can know that. But is he also in a position to control, direct, mislead (as in the mystery novel), deliber- ately obstruct, or confuse the observations generated by his work? Ingar- den notes--without pursuing the question--that the author may have a
48
stake in inviting the reader to a "grotesque dance of impossibilities," then focuses only on the limits of what is aesthetically acceptable.
but
49
Umberto Eco's Opera aperta takes a step further in this direction. Eco
considers an intentional and deliberate need for supplementation built into the work itself. The observer is called upon to participate in the artis- tic process. The performers not only supplement the work but engage in its composition (a feature already present in the structure of the corn- media dell'arte and its lazzi [jests]). Finally, the spectators step onto the stage, or the actors into the audience, to give the play a deliberately spon- taneous twist. Literary works, too, increasingly expect the reader to en- gage on his own in the production of meaning (in a manner that differs from case to case). Eco's prime example is Finnegan'sWake. We still find the most daring experiments of this sort in literature or in works that re- quire performance. But the visual arts follow closely with works whose meaning or even their status as artworks reveal themselves, if at all, only at a second look. And this seems to be just what the artist has in mind. He revels in saying farewell to the idea of a passive consumption of art and takes delight in the prospect that the beholder will have to do some work on his own.
Calculating this effect amounts to observing observers, which was un- necessary when observing an artwork simply meant supplementing inde- terminacies. Today, this is no longer a matter of supplying accidentals but of cooperating at the level of second-order observation. And the observer too must know, must be able to recognize the choices conceded to him as well as the boundaries he cannot overstep without rejecting the work as a work of art.
We leave our presentation on so abstract a level because it must account for every artistic genre. We could supply concrete examples from painting or lyric poetry, from ballet or drama. For the time being, however, we only wish to show that, and in what ways, art participates in a specifically mod- ern type of operation, how it constitutes itself at the level of second-order
jS Observation of the First and of the Second Order
observation as an autopoietic, operatively closed subsystem of society that decides what does and what does not concern art.
VII
The time has come to explicate the nexus between "second-order obser- vation" and "operative closure" with reference to our example, the functional system of art. To do so, we shall draw on the concept of communication.
As noted earlier, what is at stake in this system is not the mere fact that one can speak and write about art. Works of art, just like everything else, iare potential topics for communication, but this does not qualify them as something out of the ordinary. Nor does it follow that the functional sys- tem of art can differentiate itself as a social system that consists entirely of communications. Rather, works of art themselves are the medium of com-
munication, insofar as they contain directives that different observers fol- jlow more or less closely They are designed exclusivelyfor thatpurpose. Both the artists and their audiences participate in communication only as ob- servers, and the abstract concept of observation, related to distinction and indication and encompassing action and experience, permits us to formu- late what they have in common when they participate in communication.
50
Following Gotthard Gunther,
kction and experience resides in the application of the distinction between
self-reference and hetero-reference, or, from the viewpoint of the system, in the distinction between system and environment. From the cognitive perspective of the observer, experience appears to be determined by the environment. Against this determination, one posits one's own distinc- tions--true/false, pleasure/pain, agreeable/disagreeable--as if one could thereby cancel its effects. If, on the other hand, we are dealing with par- ticipation through action, then the system determines the environment. The system establishes a difference, and assuming the unity of the will, this difference exists in the environment (which does not preclude judg- ing this operation from a cognitive viewpoint as a success or failure). Both perspectives (which may appear in complex combinations), presuppose an
51
we presume that the distinction between
observer who draws distinctions and can distinguish their locations. both cases, a distinction is posited against what would otherwise be expe- rienced as determined in order to ensure the continuation of the system's autopoiesis and the perpetual oscillation between the perspective of cog- nition and the perspective of the will. Since both positions mark observer
In
Observation of the First and of the Second Order 79
standpoints, one might say that art observes itself from alternating per- spectives. The forms that constitute the work's structure are initially fix- ated by an observer for other observers. Like texts, forms abstract from the physical and mental aspects of whoever produces or observes them. Like writing, they assume a material expression that overcomes the temporal distance between subsequent observations. Today, we can find art forms that deliberately focus on a singular event or are performed in front of a random group of spectators--so as to reduce the work's social and tem- poral components to a minimum. But even if the performers staged the work entirely for their own sake, it would still be an art that experiments with its own boundaries, and it would still be communication addressed to an audience, albeit an audience tending toward zero. To produce ob- servability is to communicate order within a formal arrangement that doesn't come about spontaneously. The harlequin may dance in the dark--but his dance is still communication, a communication that sabo- tages its own perfection only to convince itself that it owes its existence to itself alone, not to die gaze of an observer. To top this triumph, one would have to observe what others would observe if they were not excluded from observation.
The other is always anticipated as observer. The audience, too, is bound by communication. They attribute the work of art to an artist. They don't confuse the work with nature. They are aware of themselves as (anony- mous) addressees of a communication and take die artwork as a minimal guarantee for the sameness of their experience. They assume that this is intentional, that something was to be shown to them. And this suffices for communication to realize itself in the observation of a difference between
52
information and utterance.
Understanding system formation via communication requires exclud-
ing the material embodiment of artworks from the system. Bodies belong to the system's environment--although they are connected to communi-
53
cation through structural couplings. What counts is their objecthood. The system knows only one operator: communication. Communication is reproduced by communications and not by operations consisting of marble, colors, dancing bodies, or sounds. One can speak of autopoietic, operatively closed systems only when the system's elements are produced and reproduced by the network of die system's elements and no prefabri-
54
cated, heterogeneous parts are used widiin the system. Like any other so-
cial system, the art system is closed on the operative basis of communica-
8 o Observation of the First and of the Second Order
tion--otherwise it would not be a system but something randomly se-
lected and "thrown together" by an observer. Materials of any kind are
merely resources that are used by communication according to its own
measure of meaning, even when they display their idiosyncratic material-
ity (for example, as raw materials). The social autonomy of the art system
;
those of society at large.
Communication by means of art, like communication about art, was
customary long before the art system organized itself on the basis of com-
munication. Several attempts were necessary to gain autonomy. The first
efforts to systematize second-order observation might conceivably be
traced to ancient Greece, where they were facilitated by writing, by a high
degree of diversity within structures and semantics, and by the privatiza-
56
tion of religion.
development. It would be difficult to account for the evolutionary emer- gence of autopoietic closure in art--or in any other domain--if there had been no prior experience with suitable components of meaning, here works of art. For autopoietic systems to emerge, the ground must be ready. But the stratified societies of the Old World were far from realizing a fully differentiated art system. Art had to please, and whom it should please was no matter of indifference. Not until modernity--we can date its beginning in the Renaissance--did the art system begin to set its own standards for recruiting observers, and the heyday of the arts in the Mid- dle Ages most likely facilitated this change. For an artist who worked in the service of God, it was only a small step to present himself as directly inspired by God. We will elaborate these issues in the following chapters.
VIII
Looking at a painting, listening to a piece of music, or simply identify- ing a work of art (as opposed to another object) from a first-order observer position does not yet imply a capacity for judging the work. The naked eye does not recognize artistic quality. But if this is true, how do we ac- count for the possibility for qualitative judgment?
The standard answer to this question invokes the role of experience, ed- ucation, or socialization in dealing with art. In the seventeenth and eigh- teenth centuries, the problem was solved via the idea of (cultivated) taste. Taste, while not innate, can be acquired in the course of a class-specific so-
rests on its ability to define and use resources in ways that differ from
55
The role of the chorus in Greek drama highlights this
Observation of the First and of the Second Order 81
cialization and then judges intuitively. The notion of good/bad taste was a first attempt to introduce the recipient or consumer into the theory of art and to problematize, on this basis, the criteria according to which the fine arts ought to be judged. This gave rise to the trend (which did not yield results until the mid-eighteenth century) of subsuming all the arts under a unified concept. This solution may have worked at first, but it begged the question of how taste is acquired to begin with, and how one can recognize its lack when it is not yet fully developed. Staring at the work of art for a prolonged period of time is of no help in this matter. Rather, the observer must assume the presence of qualitative differences that can be mastered in principle, even if they are now beyond his reach. A temporal horizon of further observations is projected into the work-- the possibility for observing with more precision, for using further dis- tinctions, for dissolving identities in dissimilarities--in short, the possi- bility for learning. Since the future is unknown, the evidence for such prospects relies on observing observers; one must observe that, and in what ways, others arrive at cultivated judgments. This temporal dimen- sion refers to the social dimension, not necessarily to the artist but to a generalized observational competence that can be activated in the en- counter with art.
These considerations suggest that a differentiating awareness of quality
emerges, along with a fully differentiated art system, at the level of second-
order observation. (This can be verified historically. ) Prior to the emer-
gence of this system, art depended on catching the beholder's eye (no mat-
ter by what means). By the seventeenth century at the latest, works that all
too deliberately called attention to themselves became suspect. It might be
necessary to impress the crowd, but the connoisseur preferred simpler and
less pompous means. According to French classicism, the artist had to
avoid baroque overkill, strive for natural expression, and reduce the work
to clear and essential forms. In the early eighteenth century, the upper
classes were still expected to distinguish themselves by acquiring good
57
judgment.
mirer in terms of direct interaction, which meant that the artist had a right to be judged competently and critically. As the system evolved, the de- mands upon the experts kept growing. Professional art criticism turned into a business and became the target of criticism. The growing dissatis- faction with arrogant "connoisseurs" and experts reflected changes in the relationship between artist and audience, which was increasingly mediated
One still thought of the relationship between artist and ad-
8 2 Observation of the First and of the Second Order
by the art market and generated a need for certain kinds of expert knowl-
edge that were no longer considered as appurtenances of the privileged
58
classes. Whether or not there were explicable qualitative criteria that
could be employed in the manner of cognitive rules became more and
59
more questionable.
doubtedly emerged and was subject to variation within the context of second-order observation.
Once even the experts no longer provide security, the problem of crite- ria resurfaces in a more radical form. Since the seventeenth century, the originality or authenticity of an artwork has generally been considered the condition of its aesthetic value. But when impostors become experts, sur- passing even the true experts, one can no longer trust one's trained eyes. Nelson Goodman has raised the question of how to preserve authenticity as a criterion for art when the experts (in our sense, #//second-order ob- servers) have failed and recourse to extra-aesthetic (for example, chemical)
60
criteria becomes inevitable.
ever. Goodman begins with the content of the criterion and then points to the future: one cannot claim that one will never be in a position to dis- tinguish the authentic work from a copy (no matter how perfect a copy). In contrast, we draw again on the concept of second-order observation. If it is true that there cannot be two authentic instances of the same object, then one assumes that a criterion capable of distinguishing the two can be found--even if one has no idea who will discover it and when. One pos- tulates an unidentified observer, whom one would have to observe in or- der to arrive at an answer to the question. The problem does not pose it- self unless art shifts to the mode of second-order self-observation.
The restoration of art belongs to the same context. If the aging of art- works, to the point of decay, is considered part of their authenticity, any restoration (even if based on credible theories about the work's original appearance) becomes problematic. Several potentially contradictory crite- ria are at work in such attempts, but one thing is certain: restoration be- comes a problem that can no longer be disposed of once the primary form of observation in art shifts to the level of second-order observation.
IX
In what follows, we will address a special problem that belongs to the realm of second-order observation but displays different logical structures.
But the ability to differentiate between qualities un-
His answer is not entirely satisfying, how-
Observation of the First and of the Second Order 83
This problem concerns the observation of what other observers cannot ob- 61
serve. In operative terms, there is nothing extraordinary about this. Neg- ative versions of, say, neurophysiological processes, thought, or commu- nication are no different from the general form of such processes. Executing a negative version does not require a specialized brain, con- sciousness, or language. But to an observer, it makes a difference whether another observer affirms or negates a proposition. We are led into regions of improbable observations when second-order observation deliberately chooses a negative version and distinguishes itself from the observing ob- servation by this very negativity, in other words, when it wants to observe what another observation does not observe, or, to radicalize the issue, when an observation specializes in observing what another observer is in- capable of observing. In this case, it is not enough to observe observations as particular phenomena* Rather, the second-order observer must focus on the instruments of observation, must observe the distinctions used in the other's observations in order to determine what these distinctions, considered as conditions of possibility for observations, exclude. In this case, the other's mode of observation is observed as a unity, as a form that
enables the observation of something by the exclusion of something else. What is excluded from observation is, first of all, the unity of the distinc- tion that underlies the observation in the form of "this and nothing else. " This is not merely a matter of positional advantages/disadvantages that could be altered by a certain shift of perspective or temporal progression. What is at stake here is rather the exclusion implied in the necessity of basing one's observation on a (any! ) distinction.
The abstractness of these introductory remarks is meant to call atten- tion to the extreme improbability of observational forms that focus on la- tencies. When thought was still based on the idea of perfecting one's na- ture, not being able to see was simply registered as an imperfection, as steresis, or corruptio, as the loss of a basic faculty. After all, we always al- ready find ourselves in the position of an observer and can take this fact for granted. Gradually, the negative version of habitually performed oper- ations turns into a figure of reflection. Blindness--not seeing--becomes the condition of possibility for seeing (and replaces other transcendental categories).
One and a half centuries of "ideology critique" and a hundred years of "psychoanalysis" have failed to incorporate this possibility into our com- mon epistemology or even to consider it as a potential expansion of that
84 Observation of the First and of the Second Order
62
epistemology.
The debate about truth claims in the "sociology of knowl-
edge" shows that the ambition to do so ruptures the cosmos of truth in
ways that can no longer be bridged--neither by "Spirit" in the Hegelian
sense nor through the construction of "metalevels" common in logic and
63
linguistics.
meet observational requirements that replicate the problem. This is why we need to think of this form of latency observation (as we shall abbreviate it here) in terms of a technique of distancing by a second-order observer that explodes the unity of the world or displaces it into unobservability. The question is: What kind of social order can afford and tolerate such a move?
Against this evolutionary theoretical background, we can see that soci- ety apparently used art to play with the possibility of second-order obser- vation before art was adapted to the realms of religion and knowledge, where it would have more serious consequences. The traditionally antago- nistic relationship between art and philosophy can help us understand this process. Philosophy is concerned with the nature and essence of things, whereas art is content with appearances. So long as it remains focused on imitation, there is no need for art to penetrate to the essence of things; it can gain (from the philosopher's viewpoint, superficial) access to nature simply by observing and replicating observations. Second-order observa- tion is at first tested in the fictional realm, and only when this practice yields sufficient evidence so that analogies to everyday experience and ac- tions suggest themselves can one begin to dissolve the unity of the Great Being, of the visible universe, of the cosmos resting securely in its forms, and reduce this unity to relative conditions of observation. This has not yet been done in an appropriately radical manner. Kant's version of tran- scendental theory still assumes that consciousness (that is, every single consciousness) has access through reflection to the conditions of possibil- ity for knowledge. Einstein's relativity theory still assumes the calculability of observational differences that can be traced to differences in speed/ accelerations between observers. It takes a radical constructivism to dis- solve these last residues of certainty. But how could one know that this is a possible, indeed, an inevitable, condition of knowing the conditions of possibility for knowledge?
In the realms of architecture and sculpture, the study of observational perspective began as early as antiquity. Art objects were designed to elicit a certain impression. Occasionally, forms required deformation for the sake of a certain optical impression, that is, they had to deviate from nature.
Any attempt in this direction must rely on distinctions and
Observation of the First and of the Second Order 8 j
The pathbreaking discovery, however, remained dependent upon individ-
64
ual objects.
servation along with new requirements that demanded the integration of an aggregate of objects via a vanishing point or unified perspective with-
65
out giving up the freedom of shaping individual objects. The painter was
now in a position to organize space (with the help of a mathematical grid) in such a way that the viewer got to see the world like God, as if from the outside. By reducing the canvas to a single space, he created an enor-
66
mously intensified combination of variety and redundancy.
The reconstruction of perspective captured an imperceptible condition of habitual seeing but did not yet position itself against the presupposed visibility of the world. One had always been able to see, even before the discovery of perspective. Perspective renders the observer visible, precisely at the point where he cannot see himself. But it assigns him a single, cor- rect position--which makes observing him unnecessary. Other than that, the introduction of perspective remains restricted to the realm of painting, which necessitates the exclusion of many previously possible choices-- such as the depiction of situations belonging to different temporal frames or of multiple appearances of a figure in one and the same painting.
Peo- ple began to wonder how they perceived the world, thus questioning per- ception in the mode of second-order observation, but only to create paint- ings that imitate nature not only with regard to its "what" but to its "how. " One began to inquire into the latent conditions of seeing, only to let them vanish again in the painting, only to accommodate the vision made possi- ble by art to the nature of seeing. The procedure still rested on the old Eu- ropean, quasi-normative understanding of nature. Reconstructing a uni- fied perspective was meant to realize what nature wants the human eye to see so that the failure of imitation, imperfection, and corruption could be avoided or so that they could be made visible and corrected with reference to the intended perspective. In this sense, unified perspective was and re- mained a technical (artistic) invention, a scaffold for mounting experi-
ences of seeing and painting.
Observing observations was not the genuine aim; it was merely a pre- condition for acquiring appropriate artistic tools. To study these tools and work with them was to confront in full the contingency of appearances. The mastery of perspective permitted experimentation with the difference between reality and appearance, to the point where it became acceptable to present distorted objects, so long as the deviation from standard expec-
In the early Renaissance, painting adopted second-order ob-
8 6 Observation of the First and of the Second Order 67
tations called attention to something specific. The art of perspective borders on the rhetorical, literary, or poetic technique of creating para- doxes. One exploits the malleability of impressions and the corresponding contingency of the object world for the sake of new artistic liberties and representational goals, which the artist must now determine all by him- self. Once this happens, the accent can shift to the side of a constructed appearance. "Perspective no longer appears as a science of reality. It is a
68
technique of hallucinations. "
continues to be the theme of art--not the observation of the act of ob- serving as such. The main concern is with technical instruction and schematic reproduction (for example, by means of a pyramid whose top serves as the vanishing point), rather than with observation, let alone with
69
the observational mode of other observers.
In addition, perspective allows observational relations to be integrated
into the unity of the pictorial space; it displays what the depicted figures, due to their spatial position in the painting, can see and what they cannot see. The unity of space guaranteed by perspective renders the represented figures observable as observers. The painting's unity is thus no longer guaranteed solely by the composition but also by the observational rela- tionships displayed within the painting. The frame does not cease to func- tion as the boundary of the composition, but the painting's internal ob- servational relationships, together witli its vanishing-point perspective, reveal the world as transcending the frame and as the true object of repre- sentation. In this way, even the invisible can be drawn into the painting and rendered visible. To be sure, relationships of observation and nonob- servation were represented prior to the invention of unified perspective, but exclusively in context-bound forms that were intelligible only with reference to a shared body of knowledge (for example, Susanna in her bath). Perspective transforms observational relationships of this kind into a universal possibility capable of integrating new constellations.
In the seventeenth century, the possibilities of painting on this side-stage of latency observation appear to have been exhausted. They do not pene- trate deeply enough into the world of individual motives. Modern society demands that social relationships be motivated, and the need to be guided by motives creates suspicion. Accordingly, the leading role in the develop- ment of latency observations shifts to theater and literature, especially to the novel. The figures that populate narrative are no longer legendary he- roes of an accepted (biblical or Greco-Roman) history. They are blatantly
The world of natural and artificial objects
Observation of the First and of the Second Order 8 7
presented as invented personae. They must be ordinary individuals drawn from real life. (What sense does it make to invent heroes? ) The interest shifts from moral, exemplary perfection or cosmic fate to complex motiva- tional structures that appear in a different light depending on whether they are the object of self-observation or are observed from the outside.
There is room for doubt whether the first variants of this technique--
the attempts to describe behavior as a consequence of reading (for exam-
ple, the critique of women's reading of novels in Don Quixote)--can al-
ready be subsumed under the function of observing latent motives, even
though the reader is invited to witnesses what the novel's heroes do not
experience themselves. These novels still foreground a transitional syn-
drome--the problem of the consequences of print--that plays a role in
other realms as well, for example, in the reactions to the public revelation
of tricks and "secrets" of the state. The depiction of a powerful profit mo-
tive that offends morality {Moll Flanders) or violates the parental determi-
nation of one's profession {Robinson Crusoe) partakes in the problematic
of second-order observation only to a limited extent. The extensive de-
bate, after 1678, about whether the behavior of the Princess de Cleves--
her confession and ultimate renunciation of love--was commendable
foregrounds the fissures in contemporary morality but does not necessar-
ily reveal latent motives. With Richardson's Pamela, at the latest, it be-
comes evident that the novel shows the reader--whether intentionally or
not--how to land a marriage without acknowledging sexual or social mo-
70
tives.
Ever since, the woman who wants to have it her way in arranging
her marriage has been considered either a mixture of innocence and rafi-
nesse or someone who intuitively follows unconscious motives in a man-
71 ner that is transparent to the reader.
One might ask whether literature or painting has been more influenced
by commercial considerations. Be this as it may, the notion that literature
ought to be "interesting" gains prominence in the course of the eighteenth
72
century.
conventional sense, by pointing to something absent. Moreover, it insists that its aesthetic accomplishment be judged not by its surprise effect but
73
in terms of its artistic means.
servations for the reader much more effectively, and they are "interesting" when they succeed. They invite the reader to observe his own observations and to discover previously unacknowledged idiosyncrasies, prejudices, and limitations.
Painting, too, wants to surprise and catch the eye, but in a more
Literary works can stage second-order ob-
8 8 Observation of the First and of the Second Order
What makes an artwork beautiful--not only pleasing upon first sight but interesting--is precisely the suggestion that second-order observation is at stake. The lines of this discussion are determined less by the distinction between artistic talent and its effect upon the audience than by the removal of obstacles that stand in the way of sympathy and interest--obstacles hav- ing to do with the formalities of a rule-based poetics, with a prescribed so- cial order that determines the relevance of persons and actions, and, above all, with moral obligations. The novel closely follows the theater in estab- lishing a level of second-order observation, and unrelated criteria of literary
74
quality are difficult to detect.
be unstable, something that could be demonstrated empirically by investi- gating which works were translated and which ones were not.
We shall not elaborate this point--for example, with reference to the practice, transparent only to the reader, of deriving highly individualistic
75
motives from an imitation of competitors.
Certainly, the intention of
The attitude toward such criteria appears to
letting the reader speculate about latent motives necessitates structurally
more complex characters who have not been flattened by the strictures of
76
morality.
generalizing the topos of latent motives. In the novel, this process leads to an internalization of signs. One can read the narrative as pointing to what it leaves unsaid but nonetheless belongs to it. In this way, the habit of ex- ploring latent motives can be trained and finally passed on into common knowledge for those versed in psychological and psychiatric research. De- spite its interest in mirrors and doubles, romanticism still believed in the
77
The complex narrative structure of the novel paves the way for
"idea" that "holds together the doubles of our knowledge. "
twentieth century, sociology and social psychology dropped this assump- tion. The individual exists as a fragmentary self that acquires a repre- sentable identity solely under pressure from the expectations of others. The excess of consciousness is repressed into the "unconscious," so that the self becomes habitual and no longer needs to experience consistency
78
as an external necessity. Whatever one may think of such theories, what
matters here is that their plausibility no longer depends on their literary, fictional presentation. Instead, they can be "verified" by common scien- tific procedures or in the successes of therapeutic practice.
In this cultural-historical situation, art begins to discover a new topic, the topic of "authenticity. " To the extent to which the observation of art- works as second-order observation becomes normalized, oppositional trends begin to emerge. At bottom, all of these trends point to the problem
In the early
Observation of the First and of the Second Order 89
of authenticity. We shall mention only one example, the cult of the sub-
lime in the late seventeenth and eighteenth centuries, triggered "just in
time" by the rediscovery of Pseudo-Longinus, the apotheosis of the sub-
lime rebels, on the one hand, against the "pompous" style that glorified the
powers of the social order and, on the other hand, against an aesthetics
79
based on rules.
rendered obsolete by second-order observation, especially by the emerging critical tendency to observe what other observers cannot observe. The fas- cination with the sublime cannot be explained in terms of its role as a sub- stitute, since it rejects rules. Rather, the cult of the sublime reacts to the loss of authenticity that occurs (or must be feared) when second-order obser- vation establishes itself and artistic production and art criticism shift to that level. No matter how Boileau might have read Longinus's text, what fascinates him is its autological, self-referential structure. "In speaking of
80
the sublime, one becomes sublime oneself. " In corresponding to itself,
the sublime is spontaneous and nondeliberate. This is why the sublime de-
81
fies definition.
But it can be demonstrated! The style of die Old Testa-
In both respects, the protest concerns positions already
ment, the ruin as a product of incessant decay, death, and what it leaves be-
hind, the cemetery--none of these phenomena is produced for the sake of
observation. One looks for remnants of a more archaic authenticity and
draws from them inspiration for authentic art. What cannot be observed is
integrated into art by representing self-transcending boundary phenom-
ena, that is, by presenting the unpresentable. Representing the renuncia-
tion of representation can once again claim credibility--or so one hopes--
just as seducers in the French novel, in somewhat different ways, try to be
sincere in their insincerity. As a solution, however, the sublime, or das Er-
habene, cannot convince for long, since it, too, is eventually proclaimed to
be a style and becomes subject to observation. For the romantics, such pro-
ductions have only a single function, namely, to emphasize the incredible,
to suggest that there is something worth suggesting. For August Wilhelm
Schlegel, the sublime is nothing more than a refined laxative for intellec-
82
tual constipation.
"sweet horror" that propels the baroness to sleep with her maid in the same
83
Others ridicule, while shuddering at the thought, the
Once the sublime takes shape, it displays a new side, from which
room.
it can be observed as both fashionable and ridiculous.
Generally speaking, second-order observation transforms latencies into contingencies, followed by the tendency to replace "what" questions by "how" questions. The result is a gradual dissolution of the constraints and
90 Observation of the First and of the Second Order
obstacles that used to secure the continuity between past and future. Con- tingencies exceed their frame conditions. But when everything can be done differently, the self-referential conclusion eventually suggests itself that one might as well continue as before--provided one conforms to the new demand for authenticity.
Contrary to all appearances, authenticity--both as a problem and as a topic--is thematized at the level of second-order observation. The ques- tion is now: How can one maintain an immediate relationship to the world while being aware of being observed as an observer, or even while knowing that one produces for the sake of being observed? How, in other words, can one abstract from a system that is fully established at the level of second-order observation and return to the paradise of first-order ob- servation? Usually this happens when the artist allows himself to be capti- vated by his own work while observing its emergence. But this begs the question of how one can demonstrate, or make observable, that one is not irritated, influenced, or manipulated by the fact of being observed.
Perhaps this problem is merely one of the forms in which art reflects, for its own sake and for the sake of other functional systems, on what modern society has rendered impossible.
X
The question of how the world can observe itself is not new, nor can it
84
be traced exclusively to Wittgenstein.
world order and the erosion of the observation of God as world observer, the questions arose: "Who else? " and "What else? " At this moment the subject began to announce itself, occasionally under the pseudonym "Spirit. " Here, ever since romanticism, art has found its niche. One re- jected alternative options for the self-observation of the world, above all, those provided by physics. "Suppose we think of nature as a self-conscious being," writes August Wilhelm Schlegel, "what would it consider the
85
greatest imposition? To study itself in terms of experimental physics. " He meant that nature finds its way "blindly. " In the twentieth century, one can no longer follow this. Instead, the world's self-observation has fallen primarily into the domain of physics, which must take into account its physically functioning tools, including living physicists, so that the world can observe itself in a manner that is irritating (and therefore in need of reflection). In this situation, can "poetry" still compete? Now that
With the retreat of the religious
Observation of the First and of the Second Order 91
such an epistemological insight is everywhere, how can poetry stand the pressure of reflection it is bound to apply to itself?
In mathematics, physics, biology, and sociology alike, the form of re- flection adapts to the radicality with which the problem is framed. A prob- lem of second-order observation is always at stake--how to observe how the world observes itself, how a marked space emerges from the unmarked space, how something becomes invisible when something else becomes visible. The generality of these questions allows one to determine more precisely what art can contribute to solving this paradox of the invisibi- lization that accompanies making something visible.
The shift to a level of second-order observation radically alters what is presupposed as the world. The first-order observer finds his objects amidst other objects and events. He can assume that his observations are linked to other objects and events and together constitute a world. To him, the world is a universitas rerum. Since he cannot see everything, he imagines invisible things. This leads to the development of symbols that represent the invisible in the visible world. Among other things, art can take over such symbolizing functions.
The second-order observer, by contrast, observes the distinctions that first-order observers (including himself) employ to emphasize and indicate something. This operation renders the world invisible. First, the world it- self cannot be observed. The act of observing, which constitutes itself in the move from an unmarked to a marked space, does not make the un- marked space disappear. (It is not clear how this could happen without a prior marking of that space. ) Rather, observation preserves that space as a necessary component of its capacity to distinguish. The unmarked space remains the other side of the form. Second, the distinguishing operation produces a two-sided form that cannot be observed as a unity (unless one employs yet another distinction) and thus remains invisible in the opera- tion. In this twofold sense, the notion of a final unity--of an "ultimate re- ality" that cannot assume a form because it has no other side--is displaced into the unobservable. With regard to the world, the distinction between inside and outside does not apply. Nor does it make sense to say that the world has an inside but no outside. The inside/outside distinction is a "pri-
86
mary distinction" that must be introduced into the world.
of the world is retained to indicate reality in its entirety, then it is that which--to a second-order observer--remains invisible in the movements of observation (his own and those of others).
If the concept
9 2 Observation of the First and of the Second Order
From this perspective, art can no longer be understood as an imitation pf something that presumably exists along with and outside of art, al- though both artworks and artists can be found and indicated in the world along with other observers. To the extent that imitation is still possible, it now imitates the world's invisibility, a nature that can no longer be appre- hended as a whole and must therefore be represented by emphasizing its
87
curves, its "lines of beauty. "
Art activates distinctions that operate in a
"connexionist" manner, thus hiding the unity of the distinction that guides
the observation. Regarding texts (and with a slightly different slant) one
might speak with Kristeva of a "zone of multiple marks and intervals whose
non-centered inscription makes possible in practice a multi-valence with-
88
out unity. "
Theology initially investigated such issues in conjunction with the no-
tion of God. Inspired by the idea of God as observer, theology began to observe this observer, even though it was forced to concede that an ob- server who creates and sustains the world by virtue of his observation ex-
89
cludes nothing and hence cannot assume an observable form.
nalizing this paradox and by incorporating the notion of observing the unobservable into the idea of God, one sought to shield the conventional notion of the world as universitas rerum from infection by logical para- doxes. To the extent, however, that modern society imbued all of its func- tional systems with second-order observation and itself ceased to provide a stable counterbalance, the concept of the world had to be altered. The world was now conceived, along the lines of a Husserlian metaphor, as an unreachable horizon that retreats further with each operation, without ever holding out the prospect of an outside.
This epochal turning point results in a shift in the "eigenvalues" that gain stability in the recursive operations of observing observations. With
90
regard to the world, they assume the modality of contingency. Whatever
exists or is made in the world could be otherwise. At least so far as con- cerns the world, the counterconcepts of necessity and impossibility are dropped; henceforth they apply only to temporally or regionally limited affairs. The world no longer owes its stability to a scaffold of essential forms that separate the necessary from the impossible. All forms, espe- cially the forms of art, must persist against the challenge that they could be different. They convince by evoking alternative possibilities while neu- tralizing any preference for forms not chosen.
However, the contingency of forms by no means prevents us from estab-
By exter-
Observation of the First and of the Second Order 93
lishing what is so in everyday life. To the first-order observer, the world re- mains exactly what it used to be. And any second-order observer is always also a first-order observer to the extent that he must focus on the observer he wants to observe. Systems theory, too, needs to establish a system refer- ence from which it can observe how a given system observes itself and the world. Not everything is other than what it is, nor does the unobservabil- ity of the world imply that we can no longer find our way from one place to the next because there is nothing "in between. " But understanding the specificity of modern society and of modern art requires taking into ac- count that these systems establish their advanced structures recursively at the level of second-order observation and that they have become so adapted to this situation that it is difficult to imagine how society could continue to operate if it were to regress entirely to a level of first-order observation.
This confirms once again that in the modern world neither consensus nor authenticity can be taken for granted or presumed to be attainable. Neither the unobservable world nor the paradox of form can secure these conditions. It means further that individuals cannot participate "authen- tically" in matters of consensus and that consensus cannot be justified simply by pointing out that individuals consent without force (that is, au- thentically). Such losses must be accepted by a society that carries out its most important operations at the level of second-order observation. The notion of the individual has long since adapted to that situation.
Individuals are self-observers. They become individuals by observing their own observations. Today, they are no longer defined by birth, by so- cial origin, or by characteristics that distinguish them from all other indi- viduals. Whether baptized or not, they are no longer "souls" in the sense of indivisible substances that guarantee an eternal life. One might argue with Simmel, Mead, or Sartre that they acquire their identity through the gaze of others, but only on condition that they observe that they are being observed.
Participating in art (which is neither necessary nor impossible) provides an opportunity for individuals to observe themselves as observers and to experience themselves as individuals. Since in art their experience is medi- ated by the perception of improbable things and events, the chances for self-observation are greater than in verbal communication. It does not mat- ter whether one acts or experiences "uniquely" in the sense of employing forms that occur only to oneself and that are inaccessible to anyone else. How could one, if there is no way to prove it? The point of self-observation
9 4 Observation of the First and of the Second Order
is not to cultivate self-reference at the expense of hetero-reference. It is merely a question of attributing visual perceptions to a perceiver, thus cre- ating an awareness of contingency that relies neither on necessities nor on impossibilities. By no means is the individual free to interpret at will. We learn from participating in art that any attempt at arbitrary interpretation is thwarted, and in what ways. This is why we can remain content with ob- serving ourselves as observers despite the fact that there is no ultimate cer- tainty of the One, the True, and the Good.
XI
The old European tradition explained the nature of society (domestic
or political) by appealing to the nature of man. From the very beginning,
however, the notion of a common nature contained a time bomb, built
into the necessity of distinguishing human nature from other creatures.
The rift between human and nonhuman nature continued to deepen--
partly because of religious concerns about "souls" and their salvation,
pardy because of the growing demands an increasingly complex society
placed on human resources. In the transition to the modern age, special-
ized human faculties such as reason and cunning were sharply empha-
91
sized. As a result, the natural foundation of society began to erode, and
its unity had to be reconstituted on the basis of reasonable motives-- hence the escape into a social contract that engages subjects and no longer relies on objects to stabilize society. Even German Idealism, despite its im- portance for aesthetics, never managed to develop a theory of observation that would acknowledge the dependence of observation on distinctions. Distinctions began to multiply, but they were always taken to be a pre- liminary step in the inquiry into the ultimate unity or ground proclaimed
92
under the name of the Idea or the Ideal.
The trajectory of this line of
questioning ends in a rejection of ontological metaphysics and the hu-
manist tradition--in the postulate of an "exemplary Being" or the reduc-
tion of society to the anonymity of the "they," as in Heidegger's Being and
93
Time, which preserves the traces of the tradition it rejects.
In the course of modernization, previous forms of grounding society in nature lost their plausibility. A normative concept of social unity displaced nature--by transforming natural law into a law of reason, by introducing the doctrine of the social contract, or by advancing the notion, shared by sociologists such as Durkheim and Parsons, that the unity and persistence
Observation of the First and of the Second Order 9 5
of society depend on a moral or, at any rate, a value-driven consensus. To- day, this notion still blocks recognition of the unity of a global system or world society, apparently because of a prevailing need for security, partic- ularly in modern society. Appeals to solidarity seek to compensate for dra- matically increasing inequalities that are still interpreted in stratificatory terms and experienced as unjust. In the face of the inevitable insecurity and volatility of crucial structures, one holds on to basic expectations, even though they are frustrated in the particular instance. The obligatory form of the normative brings this about, although it can promise nothing but counterfactual validity.
By contrast, many domains lack normative determination, especially when seen from the viewpoint of the individual. Consider love or money. Norms cannot prescribe or prohibit whether or not we love and whom. The economy would collapse (or lose its unique rationality) if rules pre- scribed how we should spend our money. Some normative constraints cer- tainly do exist in these domains. As one can learn from specific cases or from the movies, love is no excuse for espionage, and there are countless legal constraints on business transactions. But the core of these symboli- cally generalized media eludes normative regulation in much the same way as the interior of the home once did.
This basic fact refutes any theory that would establish the structure of society in the normative domain--in a tacitly assumed social contract or in a moral consensus. No one denies that expectations need to be pro- tected against disappointment. This is indispensable, as are many other things. Such protection is above all the function of law, and without law there is no society. But neither the unity nor the reproduction (auto- poiesis) of society can be reduced to that function.
All of this concerns the function of second-order observation--hence this lengthy excursus. Second-order observation takes the place of the su- pervisory authorities that a normative theory of the social system would consider indispensable and name as such. The second-order observer may be a guardian but is not necessarily so. He is not adequately described-- following the tradition of the past two centuries--as a critic who knows better. Rather, by reducing and increasing the complexity that is available to communication, his function is to arrive at a level compatible with the autopoiesis of the social system.
Second-order observation has a toxic quality. It alters one's immediate contact with the world, eroding the mode of first-order observation, which
9 6 Observation of the First and of the Second Order
it nonetheless retains. It plants the seeds of suspicion within the life-world (in the Husserlian sense) without being able to leave that world.
(This is why the
Observation of the First and of the Second Order 75
need not surrender to their gaze or become dependent on them. The artist knows that his decisions remain his own; he can go about his work au- thentically and leave it to the observer to form his own judgment.
In this way, judgment is released from the constraints of consensus,
while its relation to the object is preserved. This is worth emphasizing be-
cause it flies in the face of widely held notions concerning the conditions
of social communication. Once the old European idea of a natural corre-
spondence between art and nature became obsolete, one began to count
on consensus. This shift is evident in the social contract doctrines of the
seventeenth and eighteenth centuries, and even more so in Enlightenment
premises of a public circulation of ideas and rationally disciplined control
42
of opinions.
stantial consensus (of opinions) leads to complications that surface in cer- tain aberrations in Enlightenment thematizations of art. On the one hand, the debate about good and bad taste failed to come up with the desired criteria. Instead, it drove home the point that all so-called objective crite- ria have the effect of social discrimination; that is to say, those who expe- rience differently find themselves in bad company. On the other hand, the entire realm of art is degraded because it is contaminated with sensory ex- perience and compromised by its dependence on inferior types of cogni- tion. In a state of turmoil, society decides to search for consensus, general- izing as transcendental a prioris or as new mythologies to be expected from the future only those symbols capable of binding each and every subject. In the meantime, one puts up with ideologies.
Today the realization that communication is coordinated by objects
43
The one-sided understanding of society as a kind of sub-
rather than justifications is gaining ground.
Dissent in the realm of jus-
tifications is tolerable so long as adjustments mediated by objects are at
work. This situation implies that bodies can be treated like phenomena,
without understanding the biochemistry of their lives, the neurophysiol-
ogy of their brain processes, or the conscious states actualized at any given
44
moment.
for consensus and ensure that object orientation will have its proper place. This has at least one significant advantage. It releases further communica- tion from its fixation on any given topic and leaves it to communication to decide whether or not opinions are at stake, and if so, how serious and binding they are meant to be.
All of this considerably affects our understanding of art as a form for second-order observations. Art permits a kind of playful relationship to
Limited resources prevent society from overestimating its need
j6 Observation of the First and of the Second Order
questions of reasonable consensus or dissent. It avoids degrading or ex- cluding those who think differently. And it does so in such a way that doubts about whether or not one communicates about the same thing never arise. This is not to say that art does not place high (and exclusive) demands on an observation that seeks to be adequate. The measure for adequacy is not a consensus determined by a "shared symbolic system" (Parsons), but resides instead in the question of whether the viewer can follow the directives for adequate observation embedded in the work's own formal decisions.
VI
Traditionally, the theory of art and literature did not describe the rela-
tionship between artist and observer (author and reader) in terms of an
observational relation. Instead, it assumed a causal relation, an effectua-
tion of effects. Accordingly, the artist was believed to be interested in elic-
iting a certain impression in the beholder, an endeavor at which he could
be more or less successful. The modern critique of this theoretical con-
stellation led to the discovery of the observer of art and, in literary theory,
to the demand that the texts be understood from the perspective of the
45
reader.
it fails to provide a sufficiently adequate theory of art (of the artwork, the text). We must assume that the author of an artwork adapts to the be- holder in the same way as an observer anticipates another observer, and that the artwork must not only mediate between diverging observational modes as they arise but also needs to generate such diverging perspectives to begin with. This is why the demise of the causal theory of art calls for a theory of second-order observation.
Many endeavors point in this direction. The widely popular "symbol-
ism" in twentieth-century literature, for example, can be taken to imply
that every interpretation, including the author's own, imposes limita-
46
tions.
petence. In a series of significant phenomenological investigations, Ro- man Ingarden has called attention to the "blanks" embedded in literary works. Such blanks both assume and require an independent "concretiza-
47
tion" on the part of the reader. An observer can perceive only schemati-
cally; he cannot simultaneously see the front and the back of the same ob- ject. But he can check his speculations against reality and find out if the
While this shift is a plausible reaction to the causal theory of art,
This may be a calculated effect aimed at the author's own incom-
Observation of the First and of the Second Order JJ
rear of a red sphere is as round, smooth, and red, and so forth as its front. In art, this kind of reality check is not possible. The viewer needs to con- jure up the necessary completion (and which one would that be? ). The creator of the work or the author of the text can know that. But is he also in a position to control, direct, mislead (as in the mystery novel), deliber- ately obstruct, or confuse the observations generated by his work? Ingar- den notes--without pursuing the question--that the author may have a
48
stake in inviting the reader to a "grotesque dance of impossibilities," then focuses only on the limits of what is aesthetically acceptable.
but
49
Umberto Eco's Opera aperta takes a step further in this direction. Eco
considers an intentional and deliberate need for supplementation built into the work itself. The observer is called upon to participate in the artis- tic process. The performers not only supplement the work but engage in its composition (a feature already present in the structure of the corn- media dell'arte and its lazzi [jests]). Finally, the spectators step onto the stage, or the actors into the audience, to give the play a deliberately spon- taneous twist. Literary works, too, increasingly expect the reader to en- gage on his own in the production of meaning (in a manner that differs from case to case). Eco's prime example is Finnegan'sWake. We still find the most daring experiments of this sort in literature or in works that re- quire performance. But the visual arts follow closely with works whose meaning or even their status as artworks reveal themselves, if at all, only at a second look. And this seems to be just what the artist has in mind. He revels in saying farewell to the idea of a passive consumption of art and takes delight in the prospect that the beholder will have to do some work on his own.
Calculating this effect amounts to observing observers, which was un- necessary when observing an artwork simply meant supplementing inde- terminacies. Today, this is no longer a matter of supplying accidentals but of cooperating at the level of second-order observation. And the observer too must know, must be able to recognize the choices conceded to him as well as the boundaries he cannot overstep without rejecting the work as a work of art.
We leave our presentation on so abstract a level because it must account for every artistic genre. We could supply concrete examples from painting or lyric poetry, from ballet or drama. For the time being, however, we only wish to show that, and in what ways, art participates in a specifically mod- ern type of operation, how it constitutes itself at the level of second-order
jS Observation of the First and of the Second Order
observation as an autopoietic, operatively closed subsystem of society that decides what does and what does not concern art.
VII
The time has come to explicate the nexus between "second-order obser- vation" and "operative closure" with reference to our example, the functional system of art. To do so, we shall draw on the concept of communication.
As noted earlier, what is at stake in this system is not the mere fact that one can speak and write about art. Works of art, just like everything else, iare potential topics for communication, but this does not qualify them as something out of the ordinary. Nor does it follow that the functional sys- tem of art can differentiate itself as a social system that consists entirely of communications. Rather, works of art themselves are the medium of com-
munication, insofar as they contain directives that different observers fol- jlow more or less closely They are designed exclusivelyfor thatpurpose. Both the artists and their audiences participate in communication only as ob- servers, and the abstract concept of observation, related to distinction and indication and encompassing action and experience, permits us to formu- late what they have in common when they participate in communication.
50
Following Gotthard Gunther,
kction and experience resides in the application of the distinction between
self-reference and hetero-reference, or, from the viewpoint of the system, in the distinction between system and environment. From the cognitive perspective of the observer, experience appears to be determined by the environment. Against this determination, one posits one's own distinc- tions--true/false, pleasure/pain, agreeable/disagreeable--as if one could thereby cancel its effects. If, on the other hand, we are dealing with par- ticipation through action, then the system determines the environment. The system establishes a difference, and assuming the unity of the will, this difference exists in the environment (which does not preclude judg- ing this operation from a cognitive viewpoint as a success or failure). Both perspectives (which may appear in complex combinations), presuppose an
51
we presume that the distinction between
observer who draws distinctions and can distinguish their locations. both cases, a distinction is posited against what would otherwise be expe- rienced as determined in order to ensure the continuation of the system's autopoiesis and the perpetual oscillation between the perspective of cog- nition and the perspective of the will. Since both positions mark observer
In
Observation of the First and of the Second Order 79
standpoints, one might say that art observes itself from alternating per- spectives. The forms that constitute the work's structure are initially fix- ated by an observer for other observers. Like texts, forms abstract from the physical and mental aspects of whoever produces or observes them. Like writing, they assume a material expression that overcomes the temporal distance between subsequent observations. Today, we can find art forms that deliberately focus on a singular event or are performed in front of a random group of spectators--so as to reduce the work's social and tem- poral components to a minimum. But even if the performers staged the work entirely for their own sake, it would still be an art that experiments with its own boundaries, and it would still be communication addressed to an audience, albeit an audience tending toward zero. To produce ob- servability is to communicate order within a formal arrangement that doesn't come about spontaneously. The harlequin may dance in the dark--but his dance is still communication, a communication that sabo- tages its own perfection only to convince itself that it owes its existence to itself alone, not to die gaze of an observer. To top this triumph, one would have to observe what others would observe if they were not excluded from observation.
The other is always anticipated as observer. The audience, too, is bound by communication. They attribute the work of art to an artist. They don't confuse the work with nature. They are aware of themselves as (anony- mous) addressees of a communication and take die artwork as a minimal guarantee for the sameness of their experience. They assume that this is intentional, that something was to be shown to them. And this suffices for communication to realize itself in the observation of a difference between
52
information and utterance.
Understanding system formation via communication requires exclud-
ing the material embodiment of artworks from the system. Bodies belong to the system's environment--although they are connected to communi-
53
cation through structural couplings. What counts is their objecthood. The system knows only one operator: communication. Communication is reproduced by communications and not by operations consisting of marble, colors, dancing bodies, or sounds. One can speak of autopoietic, operatively closed systems only when the system's elements are produced and reproduced by the network of die system's elements and no prefabri-
54
cated, heterogeneous parts are used widiin the system. Like any other so-
cial system, the art system is closed on the operative basis of communica-
8 o Observation of the First and of the Second Order
tion--otherwise it would not be a system but something randomly se-
lected and "thrown together" by an observer. Materials of any kind are
merely resources that are used by communication according to its own
measure of meaning, even when they display their idiosyncratic material-
ity (for example, as raw materials). The social autonomy of the art system
;
those of society at large.
Communication by means of art, like communication about art, was
customary long before the art system organized itself on the basis of com-
munication. Several attempts were necessary to gain autonomy. The first
efforts to systematize second-order observation might conceivably be
traced to ancient Greece, where they were facilitated by writing, by a high
degree of diversity within structures and semantics, and by the privatiza-
56
tion of religion.
development. It would be difficult to account for the evolutionary emer- gence of autopoietic closure in art--or in any other domain--if there had been no prior experience with suitable components of meaning, here works of art. For autopoietic systems to emerge, the ground must be ready. But the stratified societies of the Old World were far from realizing a fully differentiated art system. Art had to please, and whom it should please was no matter of indifference. Not until modernity--we can date its beginning in the Renaissance--did the art system begin to set its own standards for recruiting observers, and the heyday of the arts in the Mid- dle Ages most likely facilitated this change. For an artist who worked in the service of God, it was only a small step to present himself as directly inspired by God. We will elaborate these issues in the following chapters.
VIII
Looking at a painting, listening to a piece of music, or simply identify- ing a work of art (as opposed to another object) from a first-order observer position does not yet imply a capacity for judging the work. The naked eye does not recognize artistic quality. But if this is true, how do we ac- count for the possibility for qualitative judgment?
The standard answer to this question invokes the role of experience, ed- ucation, or socialization in dealing with art. In the seventeenth and eigh- teenth centuries, the problem was solved via the idea of (cultivated) taste. Taste, while not innate, can be acquired in the course of a class-specific so-
rests on its ability to define and use resources in ways that differ from
55
The role of the chorus in Greek drama highlights this
Observation of the First and of the Second Order 81
cialization and then judges intuitively. The notion of good/bad taste was a first attempt to introduce the recipient or consumer into the theory of art and to problematize, on this basis, the criteria according to which the fine arts ought to be judged. This gave rise to the trend (which did not yield results until the mid-eighteenth century) of subsuming all the arts under a unified concept. This solution may have worked at first, but it begged the question of how taste is acquired to begin with, and how one can recognize its lack when it is not yet fully developed. Staring at the work of art for a prolonged period of time is of no help in this matter. Rather, the observer must assume the presence of qualitative differences that can be mastered in principle, even if they are now beyond his reach. A temporal horizon of further observations is projected into the work-- the possibility for observing with more precision, for using further dis- tinctions, for dissolving identities in dissimilarities--in short, the possi- bility for learning. Since the future is unknown, the evidence for such prospects relies on observing observers; one must observe that, and in what ways, others arrive at cultivated judgments. This temporal dimen- sion refers to the social dimension, not necessarily to the artist but to a generalized observational competence that can be activated in the en- counter with art.
These considerations suggest that a differentiating awareness of quality
emerges, along with a fully differentiated art system, at the level of second-
order observation. (This can be verified historically. ) Prior to the emer-
gence of this system, art depended on catching the beholder's eye (no mat-
ter by what means). By the seventeenth century at the latest, works that all
too deliberately called attention to themselves became suspect. It might be
necessary to impress the crowd, but the connoisseur preferred simpler and
less pompous means. According to French classicism, the artist had to
avoid baroque overkill, strive for natural expression, and reduce the work
to clear and essential forms. In the early eighteenth century, the upper
classes were still expected to distinguish themselves by acquiring good
57
judgment.
mirer in terms of direct interaction, which meant that the artist had a right to be judged competently and critically. As the system evolved, the de- mands upon the experts kept growing. Professional art criticism turned into a business and became the target of criticism. The growing dissatis- faction with arrogant "connoisseurs" and experts reflected changes in the relationship between artist and audience, which was increasingly mediated
One still thought of the relationship between artist and ad-
8 2 Observation of the First and of the Second Order
by the art market and generated a need for certain kinds of expert knowl-
edge that were no longer considered as appurtenances of the privileged
58
classes. Whether or not there were explicable qualitative criteria that
could be employed in the manner of cognitive rules became more and
59
more questionable.
doubtedly emerged and was subject to variation within the context of second-order observation.
Once even the experts no longer provide security, the problem of crite- ria resurfaces in a more radical form. Since the seventeenth century, the originality or authenticity of an artwork has generally been considered the condition of its aesthetic value. But when impostors become experts, sur- passing even the true experts, one can no longer trust one's trained eyes. Nelson Goodman has raised the question of how to preserve authenticity as a criterion for art when the experts (in our sense, #//second-order ob- servers) have failed and recourse to extra-aesthetic (for example, chemical)
60
criteria becomes inevitable.
ever. Goodman begins with the content of the criterion and then points to the future: one cannot claim that one will never be in a position to dis- tinguish the authentic work from a copy (no matter how perfect a copy). In contrast, we draw again on the concept of second-order observation. If it is true that there cannot be two authentic instances of the same object, then one assumes that a criterion capable of distinguishing the two can be found--even if one has no idea who will discover it and when. One pos- tulates an unidentified observer, whom one would have to observe in or- der to arrive at an answer to the question. The problem does not pose it- self unless art shifts to the mode of second-order self-observation.
The restoration of art belongs to the same context. If the aging of art- works, to the point of decay, is considered part of their authenticity, any restoration (even if based on credible theories about the work's original appearance) becomes problematic. Several potentially contradictory crite- ria are at work in such attempts, but one thing is certain: restoration be- comes a problem that can no longer be disposed of once the primary form of observation in art shifts to the level of second-order observation.
IX
In what follows, we will address a special problem that belongs to the realm of second-order observation but displays different logical structures.
But the ability to differentiate between qualities un-
His answer is not entirely satisfying, how-
Observation of the First and of the Second Order 83
This problem concerns the observation of what other observers cannot ob- 61
serve. In operative terms, there is nothing extraordinary about this. Neg- ative versions of, say, neurophysiological processes, thought, or commu- nication are no different from the general form of such processes. Executing a negative version does not require a specialized brain, con- sciousness, or language. But to an observer, it makes a difference whether another observer affirms or negates a proposition. We are led into regions of improbable observations when second-order observation deliberately chooses a negative version and distinguishes itself from the observing ob- servation by this very negativity, in other words, when it wants to observe what another observation does not observe, or, to radicalize the issue, when an observation specializes in observing what another observer is in- capable of observing. In this case, it is not enough to observe observations as particular phenomena* Rather, the second-order observer must focus on the instruments of observation, must observe the distinctions used in the other's observations in order to determine what these distinctions, considered as conditions of possibility for observations, exclude. In this case, the other's mode of observation is observed as a unity, as a form that
enables the observation of something by the exclusion of something else. What is excluded from observation is, first of all, the unity of the distinc- tion that underlies the observation in the form of "this and nothing else. " This is not merely a matter of positional advantages/disadvantages that could be altered by a certain shift of perspective or temporal progression. What is at stake here is rather the exclusion implied in the necessity of basing one's observation on a (any! ) distinction.
The abstractness of these introductory remarks is meant to call atten- tion to the extreme improbability of observational forms that focus on la- tencies. When thought was still based on the idea of perfecting one's na- ture, not being able to see was simply registered as an imperfection, as steresis, or corruptio, as the loss of a basic faculty. After all, we always al- ready find ourselves in the position of an observer and can take this fact for granted. Gradually, the negative version of habitually performed oper- ations turns into a figure of reflection. Blindness--not seeing--becomes the condition of possibility for seeing (and replaces other transcendental categories).
One and a half centuries of "ideology critique" and a hundred years of "psychoanalysis" have failed to incorporate this possibility into our com- mon epistemology or even to consider it as a potential expansion of that
84 Observation of the First and of the Second Order
62
epistemology.
The debate about truth claims in the "sociology of knowl-
edge" shows that the ambition to do so ruptures the cosmos of truth in
ways that can no longer be bridged--neither by "Spirit" in the Hegelian
sense nor through the construction of "metalevels" common in logic and
63
linguistics.
meet observational requirements that replicate the problem. This is why we need to think of this form of latency observation (as we shall abbreviate it here) in terms of a technique of distancing by a second-order observer that explodes the unity of the world or displaces it into unobservability. The question is: What kind of social order can afford and tolerate such a move?
Against this evolutionary theoretical background, we can see that soci- ety apparently used art to play with the possibility of second-order obser- vation before art was adapted to the realms of religion and knowledge, where it would have more serious consequences. The traditionally antago- nistic relationship between art and philosophy can help us understand this process. Philosophy is concerned with the nature and essence of things, whereas art is content with appearances. So long as it remains focused on imitation, there is no need for art to penetrate to the essence of things; it can gain (from the philosopher's viewpoint, superficial) access to nature simply by observing and replicating observations. Second-order observa- tion is at first tested in the fictional realm, and only when this practice yields sufficient evidence so that analogies to everyday experience and ac- tions suggest themselves can one begin to dissolve the unity of the Great Being, of the visible universe, of the cosmos resting securely in its forms, and reduce this unity to relative conditions of observation. This has not yet been done in an appropriately radical manner. Kant's version of tran- scendental theory still assumes that consciousness (that is, every single consciousness) has access through reflection to the conditions of possibil- ity for knowledge. Einstein's relativity theory still assumes the calculability of observational differences that can be traced to differences in speed/ accelerations between observers. It takes a radical constructivism to dis- solve these last residues of certainty. But how could one know that this is a possible, indeed, an inevitable, condition of knowing the conditions of possibility for knowledge?
In the realms of architecture and sculpture, the study of observational perspective began as early as antiquity. Art objects were designed to elicit a certain impression. Occasionally, forms required deformation for the sake of a certain optical impression, that is, they had to deviate from nature.
Any attempt in this direction must rely on distinctions and
Observation of the First and of the Second Order 8 j
The pathbreaking discovery, however, remained dependent upon individ-
64
ual objects.
servation along with new requirements that demanded the integration of an aggregate of objects via a vanishing point or unified perspective with-
65
out giving up the freedom of shaping individual objects. The painter was
now in a position to organize space (with the help of a mathematical grid) in such a way that the viewer got to see the world like God, as if from the outside. By reducing the canvas to a single space, he created an enor-
66
mously intensified combination of variety and redundancy.
The reconstruction of perspective captured an imperceptible condition of habitual seeing but did not yet position itself against the presupposed visibility of the world. One had always been able to see, even before the discovery of perspective. Perspective renders the observer visible, precisely at the point where he cannot see himself. But it assigns him a single, cor- rect position--which makes observing him unnecessary. Other than that, the introduction of perspective remains restricted to the realm of painting, which necessitates the exclusion of many previously possible choices-- such as the depiction of situations belonging to different temporal frames or of multiple appearances of a figure in one and the same painting.
Peo- ple began to wonder how they perceived the world, thus questioning per- ception in the mode of second-order observation, but only to create paint- ings that imitate nature not only with regard to its "what" but to its "how. " One began to inquire into the latent conditions of seeing, only to let them vanish again in the painting, only to accommodate the vision made possi- ble by art to the nature of seeing. The procedure still rested on the old Eu- ropean, quasi-normative understanding of nature. Reconstructing a uni- fied perspective was meant to realize what nature wants the human eye to see so that the failure of imitation, imperfection, and corruption could be avoided or so that they could be made visible and corrected with reference to the intended perspective. In this sense, unified perspective was and re- mained a technical (artistic) invention, a scaffold for mounting experi-
ences of seeing and painting.
Observing observations was not the genuine aim; it was merely a pre- condition for acquiring appropriate artistic tools. To study these tools and work with them was to confront in full the contingency of appearances. The mastery of perspective permitted experimentation with the difference between reality and appearance, to the point where it became acceptable to present distorted objects, so long as the deviation from standard expec-
In the early Renaissance, painting adopted second-order ob-
8 6 Observation of the First and of the Second Order 67
tations called attention to something specific. The art of perspective borders on the rhetorical, literary, or poetic technique of creating para- doxes. One exploits the malleability of impressions and the corresponding contingency of the object world for the sake of new artistic liberties and representational goals, which the artist must now determine all by him- self. Once this happens, the accent can shift to the side of a constructed appearance. "Perspective no longer appears as a science of reality. It is a
68
technique of hallucinations. "
continues to be the theme of art--not the observation of the act of ob- serving as such. The main concern is with technical instruction and schematic reproduction (for example, by means of a pyramid whose top serves as the vanishing point), rather than with observation, let alone with
69
the observational mode of other observers.
In addition, perspective allows observational relations to be integrated
into the unity of the pictorial space; it displays what the depicted figures, due to their spatial position in the painting, can see and what they cannot see. The unity of space guaranteed by perspective renders the represented figures observable as observers. The painting's unity is thus no longer guaranteed solely by the composition but also by the observational rela- tionships displayed within the painting. The frame does not cease to func- tion as the boundary of the composition, but the painting's internal ob- servational relationships, together witli its vanishing-point perspective, reveal the world as transcending the frame and as the true object of repre- sentation. In this way, even the invisible can be drawn into the painting and rendered visible. To be sure, relationships of observation and nonob- servation were represented prior to the invention of unified perspective, but exclusively in context-bound forms that were intelligible only with reference to a shared body of knowledge (for example, Susanna in her bath). Perspective transforms observational relationships of this kind into a universal possibility capable of integrating new constellations.
In the seventeenth century, the possibilities of painting on this side-stage of latency observation appear to have been exhausted. They do not pene- trate deeply enough into the world of individual motives. Modern society demands that social relationships be motivated, and the need to be guided by motives creates suspicion. Accordingly, the leading role in the develop- ment of latency observations shifts to theater and literature, especially to the novel. The figures that populate narrative are no longer legendary he- roes of an accepted (biblical or Greco-Roman) history. They are blatantly
The world of natural and artificial objects
Observation of the First and of the Second Order 8 7
presented as invented personae. They must be ordinary individuals drawn from real life. (What sense does it make to invent heroes? ) The interest shifts from moral, exemplary perfection or cosmic fate to complex motiva- tional structures that appear in a different light depending on whether they are the object of self-observation or are observed from the outside.
There is room for doubt whether the first variants of this technique--
the attempts to describe behavior as a consequence of reading (for exam-
ple, the critique of women's reading of novels in Don Quixote)--can al-
ready be subsumed under the function of observing latent motives, even
though the reader is invited to witnesses what the novel's heroes do not
experience themselves. These novels still foreground a transitional syn-
drome--the problem of the consequences of print--that plays a role in
other realms as well, for example, in the reactions to the public revelation
of tricks and "secrets" of the state. The depiction of a powerful profit mo-
tive that offends morality {Moll Flanders) or violates the parental determi-
nation of one's profession {Robinson Crusoe) partakes in the problematic
of second-order observation only to a limited extent. The extensive de-
bate, after 1678, about whether the behavior of the Princess de Cleves--
her confession and ultimate renunciation of love--was commendable
foregrounds the fissures in contemporary morality but does not necessar-
ily reveal latent motives. With Richardson's Pamela, at the latest, it be-
comes evident that the novel shows the reader--whether intentionally or
not--how to land a marriage without acknowledging sexual or social mo-
70
tives.
Ever since, the woman who wants to have it her way in arranging
her marriage has been considered either a mixture of innocence and rafi-
nesse or someone who intuitively follows unconscious motives in a man-
71 ner that is transparent to the reader.
One might ask whether literature or painting has been more influenced
by commercial considerations. Be this as it may, the notion that literature
ought to be "interesting" gains prominence in the course of the eighteenth
72
century.
conventional sense, by pointing to something absent. Moreover, it insists that its aesthetic accomplishment be judged not by its surprise effect but
73
in terms of its artistic means.
servations for the reader much more effectively, and they are "interesting" when they succeed. They invite the reader to observe his own observations and to discover previously unacknowledged idiosyncrasies, prejudices, and limitations.
Painting, too, wants to surprise and catch the eye, but in a more
Literary works can stage second-order ob-
8 8 Observation of the First and of the Second Order
What makes an artwork beautiful--not only pleasing upon first sight but interesting--is precisely the suggestion that second-order observation is at stake. The lines of this discussion are determined less by the distinction between artistic talent and its effect upon the audience than by the removal of obstacles that stand in the way of sympathy and interest--obstacles hav- ing to do with the formalities of a rule-based poetics, with a prescribed so- cial order that determines the relevance of persons and actions, and, above all, with moral obligations. The novel closely follows the theater in estab- lishing a level of second-order observation, and unrelated criteria of literary
74
quality are difficult to detect.
be unstable, something that could be demonstrated empirically by investi- gating which works were translated and which ones were not.
We shall not elaborate this point--for example, with reference to the practice, transparent only to the reader, of deriving highly individualistic
75
motives from an imitation of competitors.
Certainly, the intention of
The attitude toward such criteria appears to
letting the reader speculate about latent motives necessitates structurally
more complex characters who have not been flattened by the strictures of
76
morality.
generalizing the topos of latent motives. In the novel, this process leads to an internalization of signs. One can read the narrative as pointing to what it leaves unsaid but nonetheless belongs to it. In this way, the habit of ex- ploring latent motives can be trained and finally passed on into common knowledge for those versed in psychological and psychiatric research. De- spite its interest in mirrors and doubles, romanticism still believed in the
77
The complex narrative structure of the novel paves the way for
"idea" that "holds together the doubles of our knowledge. "
twentieth century, sociology and social psychology dropped this assump- tion. The individual exists as a fragmentary self that acquires a repre- sentable identity solely under pressure from the expectations of others. The excess of consciousness is repressed into the "unconscious," so that the self becomes habitual and no longer needs to experience consistency
78
as an external necessity. Whatever one may think of such theories, what
matters here is that their plausibility no longer depends on their literary, fictional presentation. Instead, they can be "verified" by common scien- tific procedures or in the successes of therapeutic practice.
In this cultural-historical situation, art begins to discover a new topic, the topic of "authenticity. " To the extent to which the observation of art- works as second-order observation becomes normalized, oppositional trends begin to emerge. At bottom, all of these trends point to the problem
In the early
Observation of the First and of the Second Order 89
of authenticity. We shall mention only one example, the cult of the sub-
lime in the late seventeenth and eighteenth centuries, triggered "just in
time" by the rediscovery of Pseudo-Longinus, the apotheosis of the sub-
lime rebels, on the one hand, against the "pompous" style that glorified the
powers of the social order and, on the other hand, against an aesthetics
79
based on rules.
rendered obsolete by second-order observation, especially by the emerging critical tendency to observe what other observers cannot observe. The fas- cination with the sublime cannot be explained in terms of its role as a sub- stitute, since it rejects rules. Rather, the cult of the sublime reacts to the loss of authenticity that occurs (or must be feared) when second-order obser- vation establishes itself and artistic production and art criticism shift to that level. No matter how Boileau might have read Longinus's text, what fascinates him is its autological, self-referential structure. "In speaking of
80
the sublime, one becomes sublime oneself. " In corresponding to itself,
the sublime is spontaneous and nondeliberate. This is why the sublime de-
81
fies definition.
But it can be demonstrated! The style of die Old Testa-
In both respects, the protest concerns positions already
ment, the ruin as a product of incessant decay, death, and what it leaves be-
hind, the cemetery--none of these phenomena is produced for the sake of
observation. One looks for remnants of a more archaic authenticity and
draws from them inspiration for authentic art. What cannot be observed is
integrated into art by representing self-transcending boundary phenom-
ena, that is, by presenting the unpresentable. Representing the renuncia-
tion of representation can once again claim credibility--or so one hopes--
just as seducers in the French novel, in somewhat different ways, try to be
sincere in their insincerity. As a solution, however, the sublime, or das Er-
habene, cannot convince for long, since it, too, is eventually proclaimed to
be a style and becomes subject to observation. For the romantics, such pro-
ductions have only a single function, namely, to emphasize the incredible,
to suggest that there is something worth suggesting. For August Wilhelm
Schlegel, the sublime is nothing more than a refined laxative for intellec-
82
tual constipation.
"sweet horror" that propels the baroness to sleep with her maid in the same
83
Others ridicule, while shuddering at the thought, the
Once the sublime takes shape, it displays a new side, from which
room.
it can be observed as both fashionable and ridiculous.
Generally speaking, second-order observation transforms latencies into contingencies, followed by the tendency to replace "what" questions by "how" questions. The result is a gradual dissolution of the constraints and
90 Observation of the First and of the Second Order
obstacles that used to secure the continuity between past and future. Con- tingencies exceed their frame conditions. But when everything can be done differently, the self-referential conclusion eventually suggests itself that one might as well continue as before--provided one conforms to the new demand for authenticity.
Contrary to all appearances, authenticity--both as a problem and as a topic--is thematized at the level of second-order observation. The ques- tion is now: How can one maintain an immediate relationship to the world while being aware of being observed as an observer, or even while knowing that one produces for the sake of being observed? How, in other words, can one abstract from a system that is fully established at the level of second-order observation and return to the paradise of first-order ob- servation? Usually this happens when the artist allows himself to be capti- vated by his own work while observing its emergence. But this begs the question of how one can demonstrate, or make observable, that one is not irritated, influenced, or manipulated by the fact of being observed.
Perhaps this problem is merely one of the forms in which art reflects, for its own sake and for the sake of other functional systems, on what modern society has rendered impossible.
X
The question of how the world can observe itself is not new, nor can it
84
be traced exclusively to Wittgenstein.
world order and the erosion of the observation of God as world observer, the questions arose: "Who else? " and "What else? " At this moment the subject began to announce itself, occasionally under the pseudonym "Spirit. " Here, ever since romanticism, art has found its niche. One re- jected alternative options for the self-observation of the world, above all, those provided by physics. "Suppose we think of nature as a self-conscious being," writes August Wilhelm Schlegel, "what would it consider the
85
greatest imposition? To study itself in terms of experimental physics. " He meant that nature finds its way "blindly. " In the twentieth century, one can no longer follow this. Instead, the world's self-observation has fallen primarily into the domain of physics, which must take into account its physically functioning tools, including living physicists, so that the world can observe itself in a manner that is irritating (and therefore in need of reflection). In this situation, can "poetry" still compete? Now that
With the retreat of the religious
Observation of the First and of the Second Order 91
such an epistemological insight is everywhere, how can poetry stand the pressure of reflection it is bound to apply to itself?
In mathematics, physics, biology, and sociology alike, the form of re- flection adapts to the radicality with which the problem is framed. A prob- lem of second-order observation is always at stake--how to observe how the world observes itself, how a marked space emerges from the unmarked space, how something becomes invisible when something else becomes visible. The generality of these questions allows one to determine more precisely what art can contribute to solving this paradox of the invisibi- lization that accompanies making something visible.
The shift to a level of second-order observation radically alters what is presupposed as the world. The first-order observer finds his objects amidst other objects and events. He can assume that his observations are linked to other objects and events and together constitute a world. To him, the world is a universitas rerum. Since he cannot see everything, he imagines invisible things. This leads to the development of symbols that represent the invisible in the visible world. Among other things, art can take over such symbolizing functions.
The second-order observer, by contrast, observes the distinctions that first-order observers (including himself) employ to emphasize and indicate something. This operation renders the world invisible. First, the world it- self cannot be observed. The act of observing, which constitutes itself in the move from an unmarked to a marked space, does not make the un- marked space disappear. (It is not clear how this could happen without a prior marking of that space. ) Rather, observation preserves that space as a necessary component of its capacity to distinguish. The unmarked space remains the other side of the form. Second, the distinguishing operation produces a two-sided form that cannot be observed as a unity (unless one employs yet another distinction) and thus remains invisible in the opera- tion. In this twofold sense, the notion of a final unity--of an "ultimate re- ality" that cannot assume a form because it has no other side--is displaced into the unobservable. With regard to the world, the distinction between inside and outside does not apply. Nor does it make sense to say that the world has an inside but no outside. The inside/outside distinction is a "pri-
86
mary distinction" that must be introduced into the world.
of the world is retained to indicate reality in its entirety, then it is that which--to a second-order observer--remains invisible in the movements of observation (his own and those of others).
If the concept
9 2 Observation of the First and of the Second Order
From this perspective, art can no longer be understood as an imitation pf something that presumably exists along with and outside of art, al- though both artworks and artists can be found and indicated in the world along with other observers. To the extent that imitation is still possible, it now imitates the world's invisibility, a nature that can no longer be appre- hended as a whole and must therefore be represented by emphasizing its
87
curves, its "lines of beauty. "
Art activates distinctions that operate in a
"connexionist" manner, thus hiding the unity of the distinction that guides
the observation. Regarding texts (and with a slightly different slant) one
might speak with Kristeva of a "zone of multiple marks and intervals whose
non-centered inscription makes possible in practice a multi-valence with-
88
out unity. "
Theology initially investigated such issues in conjunction with the no-
tion of God. Inspired by the idea of God as observer, theology began to observe this observer, even though it was forced to concede that an ob- server who creates and sustains the world by virtue of his observation ex-
89
cludes nothing and hence cannot assume an observable form.
nalizing this paradox and by incorporating the notion of observing the unobservable into the idea of God, one sought to shield the conventional notion of the world as universitas rerum from infection by logical para- doxes. To the extent, however, that modern society imbued all of its func- tional systems with second-order observation and itself ceased to provide a stable counterbalance, the concept of the world had to be altered. The world was now conceived, along the lines of a Husserlian metaphor, as an unreachable horizon that retreats further with each operation, without ever holding out the prospect of an outside.
This epochal turning point results in a shift in the "eigenvalues" that gain stability in the recursive operations of observing observations. With
90
regard to the world, they assume the modality of contingency. Whatever
exists or is made in the world could be otherwise. At least so far as con- cerns the world, the counterconcepts of necessity and impossibility are dropped; henceforth they apply only to temporally or regionally limited affairs. The world no longer owes its stability to a scaffold of essential forms that separate the necessary from the impossible. All forms, espe- cially the forms of art, must persist against the challenge that they could be different. They convince by evoking alternative possibilities while neu- tralizing any preference for forms not chosen.
However, the contingency of forms by no means prevents us from estab-
By exter-
Observation of the First and of the Second Order 93
lishing what is so in everyday life. To the first-order observer, the world re- mains exactly what it used to be. And any second-order observer is always also a first-order observer to the extent that he must focus on the observer he wants to observe. Systems theory, too, needs to establish a system refer- ence from which it can observe how a given system observes itself and the world. Not everything is other than what it is, nor does the unobservabil- ity of the world imply that we can no longer find our way from one place to the next because there is nothing "in between. " But understanding the specificity of modern society and of modern art requires taking into ac- count that these systems establish their advanced structures recursively at the level of second-order observation and that they have become so adapted to this situation that it is difficult to imagine how society could continue to operate if it were to regress entirely to a level of first-order observation.
This confirms once again that in the modern world neither consensus nor authenticity can be taken for granted or presumed to be attainable. Neither the unobservable world nor the paradox of form can secure these conditions. It means further that individuals cannot participate "authen- tically" in matters of consensus and that consensus cannot be justified simply by pointing out that individuals consent without force (that is, au- thentically). Such losses must be accepted by a society that carries out its most important operations at the level of second-order observation. The notion of the individual has long since adapted to that situation.
Individuals are self-observers. They become individuals by observing their own observations. Today, they are no longer defined by birth, by so- cial origin, or by characteristics that distinguish them from all other indi- viduals. Whether baptized or not, they are no longer "souls" in the sense of indivisible substances that guarantee an eternal life. One might argue with Simmel, Mead, or Sartre that they acquire their identity through the gaze of others, but only on condition that they observe that they are being observed.
Participating in art (which is neither necessary nor impossible) provides an opportunity for individuals to observe themselves as observers and to experience themselves as individuals. Since in art their experience is medi- ated by the perception of improbable things and events, the chances for self-observation are greater than in verbal communication. It does not mat- ter whether one acts or experiences "uniquely" in the sense of employing forms that occur only to oneself and that are inaccessible to anyone else. How could one, if there is no way to prove it? The point of self-observation
9 4 Observation of the First and of the Second Order
is not to cultivate self-reference at the expense of hetero-reference. It is merely a question of attributing visual perceptions to a perceiver, thus cre- ating an awareness of contingency that relies neither on necessities nor on impossibilities. By no means is the individual free to interpret at will. We learn from participating in art that any attempt at arbitrary interpretation is thwarted, and in what ways. This is why we can remain content with ob- serving ourselves as observers despite the fact that there is no ultimate cer- tainty of the One, the True, and the Good.
XI
The old European tradition explained the nature of society (domestic
or political) by appealing to the nature of man. From the very beginning,
however, the notion of a common nature contained a time bomb, built
into the necessity of distinguishing human nature from other creatures.
The rift between human and nonhuman nature continued to deepen--
partly because of religious concerns about "souls" and their salvation,
pardy because of the growing demands an increasingly complex society
placed on human resources. In the transition to the modern age, special-
ized human faculties such as reason and cunning were sharply empha-
91
sized. As a result, the natural foundation of society began to erode, and
its unity had to be reconstituted on the basis of reasonable motives-- hence the escape into a social contract that engages subjects and no longer relies on objects to stabilize society. Even German Idealism, despite its im- portance for aesthetics, never managed to develop a theory of observation that would acknowledge the dependence of observation on distinctions. Distinctions began to multiply, but they were always taken to be a pre- liminary step in the inquiry into the ultimate unity or ground proclaimed
92
under the name of the Idea or the Ideal.
The trajectory of this line of
questioning ends in a rejection of ontological metaphysics and the hu-
manist tradition--in the postulate of an "exemplary Being" or the reduc-
tion of society to the anonymity of the "they," as in Heidegger's Being and
93
Time, which preserves the traces of the tradition it rejects.
In the course of modernization, previous forms of grounding society in nature lost their plausibility. A normative concept of social unity displaced nature--by transforming natural law into a law of reason, by introducing the doctrine of the social contract, or by advancing the notion, shared by sociologists such as Durkheim and Parsons, that the unity and persistence
Observation of the First and of the Second Order 9 5
of society depend on a moral or, at any rate, a value-driven consensus. To- day, this notion still blocks recognition of the unity of a global system or world society, apparently because of a prevailing need for security, partic- ularly in modern society. Appeals to solidarity seek to compensate for dra- matically increasing inequalities that are still interpreted in stratificatory terms and experienced as unjust. In the face of the inevitable insecurity and volatility of crucial structures, one holds on to basic expectations, even though they are frustrated in the particular instance. The obligatory form of the normative brings this about, although it can promise nothing but counterfactual validity.
By contrast, many domains lack normative determination, especially when seen from the viewpoint of the individual. Consider love or money. Norms cannot prescribe or prohibit whether or not we love and whom. The economy would collapse (or lose its unique rationality) if rules pre- scribed how we should spend our money. Some normative constraints cer- tainly do exist in these domains. As one can learn from specific cases or from the movies, love is no excuse for espionage, and there are countless legal constraints on business transactions. But the core of these symboli- cally generalized media eludes normative regulation in much the same way as the interior of the home once did.
This basic fact refutes any theory that would establish the structure of society in the normative domain--in a tacitly assumed social contract or in a moral consensus. No one denies that expectations need to be pro- tected against disappointment. This is indispensable, as are many other things. Such protection is above all the function of law, and without law there is no society. But neither the unity nor the reproduction (auto- poiesis) of society can be reduced to that function.
All of this concerns the function of second-order observation--hence this lengthy excursus. Second-order observation takes the place of the su- pervisory authorities that a normative theory of the social system would consider indispensable and name as such. The second-order observer may be a guardian but is not necessarily so. He is not adequately described-- following the tradition of the past two centuries--as a critic who knows better. Rather, by reducing and increasing the complexity that is available to communication, his function is to arrive at a level compatible with the autopoiesis of the social system.
Second-order observation has a toxic quality. It alters one's immediate contact with the world, eroding the mode of first-order observation, which
9 6 Observation of the First and of the Second Order
it nonetheless retains. It plants the seeds of suspicion within the life-world (in the Husserlian sense) without being able to leave that world.
