No More Learning

In systems theory, one can distinguish between operations (elements), structures, and the system, that is, one can discern a difference between sys- tem and environment. This distinction facilitates an appropriate attribu- tion of evolutionary mechanisms. One can speak of variation only where unexpected (new! ) operations occur. In these cases, selection concerns the structural value of an innovation: the innovation is either accepted as something worth repeating, or it is isolated as a singular occurrence and re- jected. Stability might be jeopardized in both cases, because new structures need to be integrated, and discarded innovations must be remembered or
48
perhaps become an object of regret. The sheer quantity of operations al-
lows trivial variations to occur on a gigantic scale, variations that, under
normal circumstances, vanish as soon as they take place. Occasionally their
structural value is recognized. In this case, selection becomes an issue.
When this happens, variation can endanger the system, exposing it to a
persistent pressure of irritation and forcing it to adapt internally to its own
49
problems.
This theoretical schema presupposes a system of sufficient complexity.
Evolutionary mechanisms cannot be thought of in isolation, unless one
can assume a "loose coupling" of multiple simultaneous operations, which
under normal circumstances ensures that variations are immediately elim-
inated; otherwise the pressure variation exerts upon structures would be
50
too high.
and tolerate structural change--in the sense of the older cybernetics, it must be organized in an "ultrastable" manner. Last but not least, evolution is possible only if the system can maintain the stability of prior and sub- sequent states and if it can distinguish between operations and structures, that is, between variations and selections. All of this precludes considering interactive systems among persons as capable of evolution, suggesting in- stead that the social system is the primary bearer of sociocultural evolu-
Apart from that, an evolving system must be able to localize
2 2 6 Evolution
tion. This raises the question--the only one of interest to us here-- whether one can speak of evolution in conjunction with social subsystems --specifically in conjunction with the art system.
Unlike the domain of evolutionary epistemology or the theory of sci- ence, in the domain of art hardly any preparatory work has been done for such analyses. In die past, evolutionary theories of social subdomains have typically been developed where, according to the self-understanding of the domain in question, problems of rationality have come to the fore: in sci- ence, for example, on the occasion of the transcendental-theoretical revo- lution and as a result of the current constructivist revolution; in the econ- omy because of doubts about whether the model of perfect competition can serve as a valid orientation; in law in view of the obsolescence of nat- ural law and the necessity of coming up with other (not just value-related) explanations for the selection of current law. It is evident that theories of evolution are also subject to evolution and that they tend to be advanced when doubts about rationality cannot be overcome in any other way. Art, however, has always thrived on the imagination, so that a typical occasion for evolutionary models of explanation never arose. Social-theoretical mod- els might conceivably be inadequate for applying the theory of evolution in die realm of art. Be this as it may, the nexus between systems theory and the theory of evolution outlined above could be an occasion to attempt an application of this sort with new theoretical tools.
IV
If one wants to apply the theoretical approach outlined above to art, one must first determine (just as in systems theory) the operation that provides the point of onset for variations. This must be the operation that supports whatever happens in art, which must not be confused with other opera- tions--otherwise one might end up with an evolution that has nothing to do with the system of art. Within the systems-theoretical framework we presented earlier, we can define this operation in only one way, namely, in terms of an observation that is focused on art. This notion covers both the production of art and the encounter with artworks. Formally, it indicates a specific way of choosing distinctions for the purpose of using one (but not the other) side as the starting point for further operations. The art- specific nature of such distinctions is evident in the realization that they are not placed haphazardly, but are positioned in relation to an emerging
Evolution 2 2 7
or existing work of art that demands, rewards, or disapproves of certain in- dications (and distinctions).
The evolution of a separate, art-specific domain within society is occa-
sioned by the fact that the artwork demands decisions concerning what
fits (is beautiful) or does not fit (is ugly), for which there is no external ori-
51
entation. We called the binary form of this unlikely occurrence "coding," and we shall use this concept to indicate the "take off" of a special kind of evolution. We can locate its beginning--which, relatively speaking, is without presuppositions--in an ornamental staggering of distinctions that exploit given conditions (for example, in pottery) in order to unfold a life of its own that is at first harmless, insignificant, indeed playful, and certainly dispensable. But this early stage already displays the features that later characterize art. A habitual pattern cries out, so to speak, for varia- tion. A small alteration yields consequences; it requires further elaboration and supplementation, or else it must be eliminated as inappropriate--and this happens repeatedly in numerous attempts that might succeed or fail, establish a tradition or perish. One form seizes the next, the side produced along with it needs to be filled, distinctions must be established or return back into themselves--and all of this is driven by an internal dynamic that propels the execution of these operations without much considera- tion for the object. Of course, the material must be receptive to such a dy- namic, and it must accommodate the purpose for which one wants to use the material. But the ornament decides for itself what fits and what does not fit. It creates an imaginary space that is stabilized by external factors without being determined by them. All of this can happen as a kind of "preadaptive advance"; there is no need to presuppose a differentiated sys- tem of art or specialized roles for artists and connoisseurs.
We argued earlier that even highly developed art forms can be traced to a kind of "inner ornament," if one pays attention to the connections be-
52
tween its distinctions.
gin with a sense for ornamentation, because ornamentation does not pre- suppose a distinct artistic realm, even though it is possible in such a realm--as if it were a matter of holding in reserve an as yet unknown fu-
53
ture. "Ritual is more than an ornamentation of time," writes Jan Assmann --but it is also just that. Art can start out from its internal ornamental structures and thus get a taste of what lies ahead. The ornament is a pos- session, which art can develop further by ever more bold distinctions and an ever more expanding imagination. From this starting point, self-assured,
The evolution of an imaginary space of art can be-
228 Evolution
art can establish relations to the world and copy familiar or desirable fea-
tures into itself. From within the ornament, which still dominates the
work, human or animal bodies emerge; or poetry creates texts, in which
sound and rhythm function as ornament. Works themselves become free
to refer to all kinds of meanings. Even when this freedom is restricted, de-
cisions remain; even when adhering to classical models, one must pay at-
tention to what is fitting when representing a Dying Gaul. Occasions for
reconstructive invention arise more frequently when die material--the
techniques or frames--is altered, and one must either determine what
kinds of formal combinations are still feasible or else experiment with new
possibilities. Such occasions arise in conjunction with the transformation of
the mural into painting on canvas, or in the relationship between painting,
mosaic, and tapestry; they arise when music that accompanies dance is dis-
engaged from the movement of the body, or when music is played with a
different set of instruments: when one stops using wood to create sculp-
tures, then abandons rock and clay for the sake of granite or marble, then
finally returns to wood; when large sculptures are replicated on a minuscule
scale in ivory; when one considers the relationship between woodblock and
lithograph or between pencil and chalk drawings. Examples could be mul-
54
tiplied,
clear, however, that the struggle with media that impose different kinds of constraints draws attention to the formal correlations that can be realized within these media.
This kind of trial already constitutes an observation specific to art, both with regard to the production of a work and to the appreciation of the work as art. The entire process begins to orient itself recursively, generat- ing a demand for criteria and a need for structure, which stimulate an evo- lution capable of preserving striking occurrences for trie sake of repetition or deviation.
Observation in this sense is the smallest unit in the artistic process. Even when the observational schema is employed repeatedly, the observ- ing operation remains a singularity that vanishes spontaneously and al- ways occurs for the first and last time. This operation focuses on a certain posture in dance (or in sculpture, as in the Laocoon), on a single color that has a certain place and intensity in a painting, on how a certain action in a given narrative moves the plot along or clarifies the motives established by the plot. {Every time a work of art is produced or understood, innu- merable observing operations are necessary. As is typical in evolutionary
but supplying evidence for such innovative thrusts is difficult. It is
Evolution 2 2 9
variations, we are dealing with a massive occurrence of trivial processes that, under normal circumstances, would be of no consequence! At this point, a kind of miniselection already takes place, as well as a test for sta- bility, which resembles the mechanisms at play in the mutations of or- ganic evolution. This raises the question of whether the decisions and opinions that have been established about a given work of art can be sus- tained in the course of further observation, or whether they have to be sacrificed or corrected.
The trivialization of operations that are sensitive to variation shows clearly that this process cannot yet be called evolutionary selection. If structural change is to yield evolutionary consequences, then it must start from a different level. In general, evolutionary selection presupposes that the adaptive relationship between system and environment is preserved in the course of variations by virtue of the system's autopoiesis (this makes selection possible and constrains it at the same time). But it does not tell us anything about the manner in which selection operates. So far as rela- tionships between meanings are concerned, the problem of selection ap- pears to reside in the reusability of the points of view that guide selection, that is, in an identification that simultaneously varies and confirms these points of view. Such identifications require that operations are observed not only as a series of situation-dependent chance events but also as the realization of a program. The differentiation of evolutionary variation and
55
selection rests on the observational level of (self-)programming.
level of observation constitutes itself only when artworks impress the be- holder as successful--whether one prefers the "novelty" of such works or whether they are produced only for the sake of deviation. At first, it might have always been a matter of imitating successful artworks that subse- quently served as models for creating variations on a given theme. There is more than one Pieth, and what is later diagnosed as a change in style might have established itself in this manner. Certain trends emerge and realize themselves in multiple variants--for example, the trend toward re- alism in portraits. One further complicates the construction of ornaments that repeat simple basic patterns and therefore react differently to varia- tions. Another example is increasing freedom in the posture of sculptures, which, when they are skillfully crafted, serve as proof of precisely this skill. So far as music is concerned, one could mention the formal impulses that result from the introduction of new instruments or from the fixation of music in musical notation.
This
2 3 0 Evolution
Unlike other, more rigidly programmed functional systems, in the evo- lution of the art system one cannot presuppose the existence of selection criteria in the way one can assume a profit motive in the economy, a crite- rion of methodological correctness in science, or the distinction equality/ inequality in current legal practice. If artworks constitute their own pro- grams, then they can convince only after the fact. Successful art can be ob- served in terms of criteria only in retrospect, and the question is always whether to imitate or to improve the work, or whether the innovation is based on rejecting all previous criteria. In an extreme sense, this is true of "modern" art, especially when it acts capriciously enough to explode the boundaries of the tolerable and pulls the rug out from underneath all pre- viously valid criteria. Doing so requires a memory that allows the art sys- tem to construct and reconstruct its evolution as if it followed an intelligi- ble order. Seen in such a way, it is no accident that the suspension of previous frame conditions and the emergence of an academic art history occur at the same time and that both demarcate an era by virtue of their operations and their observations.
That types are formed in retrospect has been observed in the art system for quite some time, under such catchwords as maniera, make, style. At first, such types were considered as a means of distinguishing and classify- ing styles and of assigning them to appropriate topics; then they served to recognize changes in styles; and finally, since Winckelmann, they have served as a means of art-historical analysis. We can therefore refer to "style" as the formal level where the evolutionary selection of structure takes place. One must keep in mind, however, that the concept of style is by no
56
means unequivocal;
and is a result of evolution (which is precisely what gives us the license for theoretical abstraction). This leads to the hypothesis we suggested earlier, namely, that the transition to modern art motivated the search for, and the discovery of, an alternative to the freedom of stylistic choice, which resides in the expansion or even dissolution of frame conditions (such as tonality in music or object orientation in painting) that, up to this point, facili- tated the emergence of specific styles and their variations. It looks as if evo- lution motivated the system to introduce concepts that call attention to the difference in level between operation and structure (or variation and selection); apparently such concepts established the boundaries that sub- sequently provoked their transgression.
In sum, these developments brought about what Darwin sought to ex-
the concept has been subject to historical change
Evolution 231
plain: a variety of species. Evolution does not guarantee survival; as a mat- ter of fact, most species in life and in art have vanished or are about to vanish. We are not dealing with essences, whether secured by nature or by a cosmos of essences. But evolution remains problematic, and so does the question of how such a proliferation of species is possible to begin with.
In the evolution of artistic genres, the development of types bifurcates in the wake of the differentiation of perceptual media for seeing and hear- ing and along with the differentiation of space and time. Any further de- velopment becomes a matter of additional bifurcations (text-art, painting, sculpture) or of combining seeing and hearing (film, theater). Under these frame conditions, a differentiation of genres occurs, which is culturally and historically important but unstable. Among these genres, the diversity of textual arts is the most impressive--displaying a wide spectrum from the epic to the epigram, from the novel to the short story, from the metric differentiation of the lyric to theme-based narrative genres (such as biog- raphy, the historical novel, science fiction, the mystery novel, and so forth). This differentiation of types is not to be understood as a "fight for life" be- tween the epic and the ode (or as a struggle for attention). The principle of competition is supplemented by the insight into the advantages--sug-
57
gested and facilitated by specific "frames" --of "insulating" innovations,
so that they do not immediately transform the entire art system.
The consequences of the separation of variation and selection and their effects are crucial for the differentiation of an art system and for the sta- bility of such a system. From the perspective of the art system, the inter- nal differentiations that establish themselves in this process no longer cor- respond to those one finds in the social environment of this system: they have nothing to do with the separation between the state apparatus and political parties, let alone with the internal differentiation of the party spectrum itself; nor do they correspond to the differentiation of banking houses and savings banks, grade schools and high schools, or to the inter- nal differentiation of faculties, not to speak of the mega-differentiations of religion, politics, the economy, education, and so on. Any one-to-one cor- respondence between system and environment (of the kind one observes
58
in tribal societies that practice a totem symbolism, for example ) is inter-
rupted. The art system decouples itself from its social environment. To be sure, the social environment does supply certain divisions in the form of neurophysiologically integrated orders that become distinguishable in die form of media of perception. While these "natural" boundaries anticipate
2 3 2 Evolution
the evolution of art, it is easy to see that they present no obstacle to a fur- ther differentiation of types, neither in the realm of seeing nor in the realm of hearing. Perhaps the differences among these media of percep- tion provide an indispensable impulse for such a differentiation.
At any rate, the "mismatch" between the system and its social environ-
ment isolates the art system from the evolution of society in general. This
is not to say that the evolution of society is without significance for the
evolution of art. On the contrary! It certainly is, but only for the internal
evolution of art. For better or for worse, art exploits the evolutionary
transformation that leads from a stratified to a functionally differentiated
59
society. But it meets this transition halfway by virtue of its internal evo-
lution. The noncorrespondence between these two types of differentiation
forces art to develop criteria for its own affairs. In the shadow of the Aris-
totelian tradition, one continued to speak of imitation well into the eigh-
teenth century, and the beginnings of a modern philosophical "aesthetics"
were motivated by the search for a common notion of beauty in nature
60
Within this framework, Hutcheson already suggested a notion
and art.
of absolute beauty, which, he believed, grounds all other types of compar-
61
ative or relative (imitative) beauty.
principle of beauty show that this is not a matter of differentiating be- tween Whigs and Tories; nor does it concern the practices of accounting in firms or of determining a focus for research in the new sciences that were about to develop into disciplines.
Starting in antiquity, guiding concepts such as harmony, balanced pro- portion, or the notion of a unity that shines through multiplicity served to
62
reconcile a sense of beauty with religion.
concepts guaranteed stability. The cosmos, understood as nature or as cre- ation, gathered a multiplicity (which can be distinguished! ) into a unity:
63
rerum dissimilium convenientia [the agreement of things dissimilar]. artistic achievements of the Renaissance inherited this notion of beauty but put it to the test, both in texts and in view of what could be repre- sented at all. On the one hand, there were no direct links to an envi- ronment ordered along the lines of politics, religion, or households. If art was appreciated, then it was appreciated as art. On the other hand, the trust in one's own critical judgment was strengthened through experiences in the workshop, comparison with other artworks, and texts that addressed issues related to art. After the notion of a general mathematical-musical- architectural world harmony was sacrificed in the sixteenth century (be-
The efforts to determine a universal
In evolutionary terms, such
The
Evolution
233
cause musical proportions could not be rendered in architecture
to create its own concept of nature and aim at "another nature. " conceded that much, then the principle of imitation could survive as a topic for quite some time; but it could no longer guarantee stability in the sense that beautiful forms could be readily repeated and reproduced.
Discussions based on criteria specific to art then began to take place. Art
mobilized, as we have pointed out, a memory of its own to orient itself in
its own history. The initial impulse was to consider everything according
to the schema of rank, as if discussing criteria were a matter of imitating
social hierarchy. One debated the primacy of individual artists and genres
and, above all, the hierarchical relationship between the ancients and the
66
moderns.
network of rules--especially in texts concerning poetics--from which one violently sought to free oneself. In the sixteenth century, the discussion of criteria was still oriented toward educational tasks. In the seventeenth cen- tury, the propagation of "beautiful appearance" in the form of art over- lapped with the science des mceurs, with the theory of political (= public) conduct, and with the doctrine of passionate love, even though there was no exact correspondence between these realms. Hutcheson still sought a unifying principle capable of comprehending the beautiful, the true, and the good, of uniting the beauty of nature and art, and of reconciling sci-
67
entific theorems and moral principles.
in these function-specific realms--such as the increasing orientation of politics toward the state and enhanced intimacy in love relationships-- such notions were dismissed one after the other. What remained was the problem of how to define criteria, a problem framed as an inquiry into the nature of beauty--that is, in a manner that did not yet distinguish be- tween coding and programming. At least officially, reflection on the art system was cast as the problem of defining beauty. But how could one come to terms with this problem, if experience showed that further differ- entiation required the generalization of symbols that nonetheless claimed
68 to represent the unity of the system?
We can assume further that the experience of a criterion-dependent se- lection of art affects the perception of artworks as well. If it is obvious that a work follows injunctions--that is, if rules and works are observed sepa- rately, yet simultaneously--the results no longer satisfy. They appear mo- notonous and uninteresting. Works produced in the classical style are no longer appreciated. Apart from the postulate of originality, the eighteenth
The result, in the seventeenth century, was a tightly woven
Because of internal developments
64
), art had
65
If one
234 Evolution
century demanded that works fulfill die additional requirement of being
"sublime," "interesting," "bizarre," "gothic," "picturesque. " Such formu-
las sought to explode the previous norms of "decorum" or "bienseance" de 69
rigueur. If one could no longer rely on generally accepted and stable cri- teria, one could at least agree on the desire for variation. Then one would be ready to concede that works of art speak to the "lower" senses of the upper classes.
In the second half of the eighteenth century, the concept of style was historicized, along with many other traditional concepts. The emergence of historical thinking uprooted the querelle des anciens et modernes--which still relied on universal comparative criteria--and the issues of that debate were displaced by analyses of historical correlations in the emergence and transformation of styles, especially in art. Styles were now defined both fac- tually and temporally. They displayed style-immanent criteria--one might say, programs for programming art. Such criteria could no longer be canonized. (Instead, one invented "classicism. ") Style provides its own directives for stylistic deviation, which is always justified when the execu- tion of the artwork succeeds. This evolutionary step destabilized the struc- tural factors that secure selection. A selection that concerns a certain style cannot also guarantee the evolutionary restabilization of the structural change it brings about. At that point the evolutionary functions of selec- tion and restabilization separate. As a result evolution gains a momentum that continually surpasses itself. There are parallels in other functional sys- tems: consider die role of profit in the economy, of passion in love, of a context-bound reason of state as a criterion of politics, and of positivity as a criterion of law. From a social-theoretical perspective, such parallels in- dicate a correlation between functional differentiation and an accelerated evolutionary structural change that affects individual functional systems in different ways, depending on their own criteria of selection. Art criti- cism can no longer appeal to individually correct insights; instead, it must be content--following the romantics--to reflect upon given accomplish- ments and merely collaborate in the creation of art. The experience of the system's internal dynamic forces one to base its stability on autonomy and to ensure that art--by means of "ideas" or by deliberate breaks with tra- dition--remains distinguishable and observable.
In such a situation, functional systems explore new semantic stabilities that are capable of oudasting such fluidities while still allowing one to for- mulate the unity and the point of one's endeavor. Typically, one sought so-
Evolution
*35
70
lutions in values. Heydenreich already questioned the value of purposes. Along with the claim of a unique, art-specific value formulated initially in terms of an "idea," art made its entrance into the nineteenth century. It is a peculiarity of values to be capable of retaining their identity even under conditions of change. They present themselves as a plurality, undisturbed by the presence of other values that might be preferable in certain situa- tions. On the contrary, displacing a disadvantaged value preserves its memory as a consolation. The notion of value indicates the manner in which the system secures its own stability while attempting to incorporate innovations. Schopenhauer believed that the object of aesthetic contem- plation was not the mere objecthood of individual artworks but "the idea that strives to reveal itself in them, that is, the adequate objectification of
71
the will at a certain stage. "
declaring: "These lectures are dedicated to Aesthetics; its object is the vast realm of the beautifuland, more specifically, art, in particular the fine arts, constitute its domain. " For Hegel, the "object" indicates the moment in which the self-reproducing consciousness experiences its own determina- tion. We can rephrase this insight as follows: the object is the system's memory.
In this way, the perspective of stability is indicated as a value. But in the context of a theory of observation and description one wants to know what the value distinguishes itself from. It goes without saying that this cannot be the countervalue of ugliness; after all, not everything that is not art (business, for example, or politics) deserves to be called ugly. The de- bate about criteria thus gives rise to problems within the self-description of the art system, and these problems point to the difference between self- reference and hetero-reference. Problems resulting from the system's need to maintain stability in the face of evolutionary change must be dealt with in the realm of the systems self-description, and this description varies de- pending on how art distinguishes itself from nonart. That topic deserves careful attention, and we therefore postpone it to the following chapter.
V
After all that has been said, the evolution of art is its own accomplish- ment. It cannot be caused by external intervention--neither the sponta- neous creativity of individual artists, nor a kind of "natural selection" by
72
Hegel still began his lectures on aesthetics by
the social environment, as Darwinian theories would have to assume.
236 Evolution
Nor can evolution be explained, as it used to be, by appealing to origins or beginnings. The theory of evolution is designed in a circular rather than a linear manner, because variation presupposes a prior state that, as a result of evolution, is stable enough to absorb variation and perhaps even evaluate it. As our previous analyses have shown, the separation between the levels of variation and selection is a result of evolution. Evolution
73
brings forth its own conditions and hence itself evolves. Recourse to an
origin in order to account for evolutionary trends becomes obsolete, in-
74
deed, becomes suspect.
In the final analysis, a circular conception of the theory of evolution
serves to reformulate the problem of the probability of the improbable or the problem of stability, which is the beginning and end of evolutionary changes in structure. Eventually, one might ask: How can an autopoietic system come into existence, if it must presuppose itself in all of its opera- tions in order to recognize what does and what does not belong to the system?
Gunther Teubner suggests that we give up thinking of autopoiesis in terms of a rigid either/or and adopt a more gradual version of the concept
75
that would solve this problem (or perhaps only make it more gradual? ). This suggestion, however, gives away the decisive advantages of the con- cept of autopoiesis, for no compelling reason. One can solve the same problem via the concept of "preadaptive advances," which has proven use- ful in the theory of evolution.
Of course, evolution is not possible without presuppositions; it is not creatio ex nihilo. Evolution presupposes a sufficiently prepared world, in which autopoietic systems can close themselves off and operate as if they had existed there before. Numerous examples could be cited--such as the
76
emergence of writing,
or the emergence of money in the form of coins
in the trading houses of Sardinia.
77
Innovations of this sort may or may
not initiate the "take off" of a new branch of sociocultural evolution. For
the art system, there are good (and goodly debatable) reasons for believing
that such a take off--which differentiates the art system from religion,
politics, and the economy and initiates an evolution of irresistible struc-
tural changes--happened only once in world history, namely, in early
78
modern Europe.
The preconditions for this evolution can be specified with accuracy
and situated historically. They reside in the already existent, highly de- veloped artistic skill and literary culture of the artes and in a poetics that
Evolution
237
offers models and allows for imitation and critical appreciation. These conditions established themselves in Europe, especially after, in the late Middle Ages, works of antiquity began to be rediscovered and admired. At first, no uniform concept could cover both the visual arts and paint- ing; nor did one have a sense of art as separate from the outside world. But an admiration of perfection oriented to the work made it possible for the "Renaissance" to assume that art already existed and only needed to be reactualized
Under such conditions, art takes off--epigenetically, indeed, counterin- tuitively and against all declared intentions.
One could just as well have continued to imitate existing models or experiment with new themes in ap- propriate fashion (maniera). In addition, a second factor has to be taken into account. The development of early modern society toward functional differentiation establishes radically new environmental conditions and cre- ates stability conditions of a different kind for the self-differentiating art system. As we indicated in Chapter 4, supporting contexts for art were ini- tially provided by the courts of the new territorial states and later by the emerging art market, both of which allowed art a certain degree of in- difference and willfulness in relation to the environment. Moreover, the splitting off of Protestantism from the Catholic Church undermined the certainty of the established religious world order. The intensification of religious propaganda led to a powerful critique of the internal dynamic of the art system--from the Protestant as well as from the Catholic side-- which, however, could not prevail and merely ended up radicalizing the problem of art-internal criteria. The development of the modern empirical- mathematical sciences relieved art from competition, especially in the edu- cational sector. Science could no longer interfere with art, nor could art in- terfere witli science. Debates about rank subsided. This development culminated around 1800, when art found itself in a societal system where it had to operate without external support, even if environmental conditions such as economic purchasing power or political nonintervention remained as important as before.
One can discuss this briefly sketched development from a number of different perspectives. For systems theory, it concerns the differentiation of the art system. When treating the self-description of the art system, we shall return to the consequences of differentiation for a reflection on the meaning of art. In the context of a theory of evolution, one can show that changes within socially presupposed stability conditions yield possibilities
238 Evolution
of variation and selection that are left to their own internal dynamic and lead to a rapidly accelerating, self-generated structural change.
When its attention was focused inward, the art system had greater op- portunities for variation, and it could expand its own criteria of selection --indeed, make them more "irrational" (if "rationality" means employing criteria that are equally acceptable in a scientific, religious, or political sense). In this way, art could cultivate intuition, imagination, exaggeration, deception, obscurity, and ambivalence and exploit these means to refer back to itself. Artistic endeavors that supported religion or politics were then criticized as "pompous. " What one later calls "baroque" aimed at op- tical illusion, especially in the construction of churches and castles but also in painting and internal architecture, as if ingeniously to escape the by now discredited representational demands of religious and political do- mains of intelligibility by adhering to these demands without fulfilling them. Another way of eluding such demands was by discovering the every- day life of peasants and burghers. At the same time, allegory provided the means to represent ideas in the form of abstract concepts and to personify such concepts. One cultivated paradox in literature with the intent of en-
79
forcing a creative, paralogical search for escape. A multileveled structure
of deception and self-deception became the object of poetry, of the theater,
and of the novel. In relation to modern science (for example, of the sort
advanced by Galileo), art no longer thought of itself as opting for false-
hood (as it did in the sixteenth-century historialpoesia debate). These issues
no longer mattered. The true/false code was "rejected" as a guiding dis-
80
and science, for its part, was no longer interested in registering
tinction,
the representations of the belles lettres and of the arts as "falsehoods. "
The debate about criteria took on a dynamic of its own, apparently be- cause it was no longer affected by external factors. One understood that autonomy, enforced by nonidentity, is a necessity of self-determination. As early as the seventeenth century, this understanding undermined the orientation toward proven recipes and rules. The emerging reflection on art pursued a separatist course that aims at the inexplicable--no so che, je ne sais quoi. Because beauty could not be subsumed under rules or laws, it could claim a domain of its own. It participated in social communica- tion because it was different. Like the sovereignty of the king and of love, the sovereignty of art displayed an inexplicability that was nonetheless not to be understood as arbitrary. One employed analytically charged concepts such as acutezza, cunning, Witz, and so forth to characterize and
Evolution 239
praise art and the artist. Such concepts promised clarification yet refused
to deliver on tJhat promise. All of these trends characterized an autonomy
forced upon art, because its claims to engage itself in a religious, political,
or scientific sense were denied by the internal logic of these functional
81
sectors.
At the same time, the aristocratic world of stratified society, the world
of households and their political alliances, exhibited corresponding signs of crisis. This crisis has often been discussed, and it is well documented with respect to the loss of political power by the nobility and to financial crises (although in places where they were most severe, as in Spain, such crises hardly affected the nobility). We shall focus only on one aspect of these crises. The individual no longer found adequate support in the old ways of life. One searched for more powerful, individual modes of ex-
82
pression--in a stylized amorous passion, for example --but also in an
elaborate code of honor, in the provocation of the duel, and in view of one's own (inevitably individualized) death. The semantics of the aristoc- racy flourished for the last time; but it did so with involuted means that
83
were out of sync with social reality. This meant that art was burdened
with the search for its own audience--by turning toward the "common" people, which one now sought to impress, or by valuing artistic expertise
84
and critical appreciation.
count the effects of reception. But now specific complementary roles were at stake, which ran parallel to those in other functional realms (for exam- ple, government and subject, plaintiff and court, buyer and seller, lover and beloved, believer and clergy), and which could no longer be integrated via a stratification ofhouseholds.
When we add up all of these changes, it becomes clear that under such conditions art--so far as its mode of operation in producing and evaluat- ing artworks is concerned--began to draw on its own resources and, in so doing, triggered evolutionary sequences unique to art. Reality was robbed of its authority to provide meaning. The rationality continuum that had traditionally joined the nature of action with its natural conditions (as it joined cognition with its object) broke apart. Don Quixote found mean- ing in his actions and an intense and unperturbed experience in reading, but not in the real world, and this experience was doubled, so to speak, and offered to the reader as the meaning of his own reading. The opera- tions that were now executed as observations specific to art could draw their meaning only from art itself. But this meant that they were subject
Since early antiquity, one had taken into ac-
2 4 0 Evolution
to the law of variation, that they could no longer pretend to accomplish perfection, but instead had to present novelty. If this was so, however, then the criteria for selection had to be redefined. The artistic effort to surpass itself could not turn into arbitrariness; it had to satisfy criteria of judgment. Ever since Gracian, one has used "taste" to refer to representa- tions in morality as well as in aesthetics, that is, in a sense that encom-
85
passes both human conduct and works of art.
tanced oneself from rational verification. Taste judged intuitively, directly, and instantaneously. That the judgments of taste were correct could be shown after the fact through reasoning and justification.
The concept of taste joined several distinctions. Initially it rejected the pedantry of applying rules; this was its historical thrust. But it also al- lowed good taste to be distinguished from bad taste, sorting not only
86
judgments but people as well.
sent argument, however, the concept of taste made possible the separation of variation and selection by rupturing--via such notions as natural per- fection or a rule that promises success--the tight coupling between the two, without turning selection into an arbitrary process.
In a parallel development--as if to compensate for the traditionally subjective and indisputable nature of the concept of taste--a notion of classicism emerged in France that offered a history of timeless models to
87
which one could return.
taste propagated during the final third of the seventeenth century and during the first third of the eighteenth century contain a resonance of
88 trust in one's judgments which one finds nowhere else.
Taste exploited the recursive network of anticipations and recapitula- tions without committing one's judgment about individual works to gen- eral and binding perspectives. But the situation in France around the mid- eighteenth century was about to change, precisely because of its ties to classicism. One subsequently spoke of goutorAy to express certain stylistic preferences--for or against color as opposed to drawing, or for or against Boucher. This is what Diderot has in mind when he demands of the critic "all sorts of tastes, a heart that is sensitive to all pleasures, a soul suscepti-
89
ble to an infinite variety of enthusiasms. "
The pseudocriterion of good taste--a criterion without criteria--regis-
tered that the evolution of the art system was already under way and was subject to perpetual structural change. But such change depended above all on the success or failure of individual self-programming artworks.
By doing so, one dis-
Most important in the context of the pre-
Perhaps this explains why French theories of
Evolution 2 4 1
There was no ordering hand above this level (in the sense in which the seventeenth and eighteenth centuries appeal to the "invisible hand," if only to deny organized religion and the absolute state jurisdiction over certain domains). Taste was still vaguely oriented toward criteria of social rank (not everyone has taste); however, rank was no longer defined in terms of birth but in terms of an expertise that the art system itself at-
90
tracted and cultivated.
sented in art. But not until the second half of the eighteenth century did corresponding semantic developments affirm the evolutionary autonomy of art and search for explanations of artistic autonomy in art itself--for example, in the historicization of the concept of style or in a new concept of "culture," both of which described evolution from a kind of bird's-eye perspective, but also in what was offered under the name of "aesthetics" as a reflection theory of the art system. At this level, one could claim stabil- ity in the face of change. What had thus evolved within a few centuries was a diversity of forms which in retrospect could no longer be under- stood as nature or perfection and which today cannot even be considered progress. This development began in early modernity; that much we know, because at that time stylistic orientation was already coupled to the production of art. Variation was motivated not only by the production of works but also by work-transcending structures, which could be experi- enced as contingent--as a maniera one could choose--and which could be hypercorrected (thus appearing to the observer of styles as a kind of mannerism or a symptom of stylistic decay). Style legitimized both con- forming behavior and deviation--precisely because it was a structural condensation of what was going on in art at any given time. Some theo- ries (for example, of landscape painting) preceded the production of cor- responding works, and one demanded of the work, among other things,
91 that it acknowledge the maniera it followed.
Art has been condemned to autonomy, and the problems of self- description that arise from this situation are translated into works of art in ways that (still) elude an intelligible theoretical explication. This is evi- dent, for example, in the poetry of paradox and later in romanticism, and it manifests itself forcefully in the avant-garde of the nineteenth and twen- tieth centuries. We shall return to this point. What interests us here is that this situation does not lead to the collapse of the mechanisms of variation, selection, and restabilization, but rather to a circular network that com- prises these different levels of evolution. Evolution is accelerated as a re-
This meant that society was no longer repre-
2 4 2 Evolution
suit, and the reservoir of forms expands to the point where the only pur- pose of constraints is to be challenged.
The nexus between diversification and acceleration corresponds to pre- cisely the kind of hypothesis an evolutionary theory would propose. So far as we can tell, we currently have no other explanation.
VI
An autonomous art system has emerged as the result of an internal evo- lution within art. The same holds for other functional systems. They all realize operative closure and self-organization, and, in so doing, they in- crease causal dependencies and independence in selective forms that are typical of modern society. Art, however, bears a special trait which it shares only with religion: participation is optional. Inclusion, whether ac- tive or passive, is a matter of individual choice. The low rate of participa- tion in art is astonishing. Only a fraction of the population participates in art, and the idiosyncrasies of modern art often serve as an excuse for stay- ing away from it. Difficulties of observation and understanding abound. On the active side, it becomes increasingly cumbersome for artists to find a niche, a style, a manner of presentation for which they can claim origi- nality. These difficulties seem extravagant and forced, which in turn af- fects people's readiness to participate passively in art.
There has been much lament over this development. But there is an- other side to it, an important advantage. Because the art system looks upon participation/nonparticipation as a matter of individual choice, both options are socially possible. No one is forced--as they were in William James's times--to feign a musical self in order to visit the opera in Boston. One abandons conventions of this sort, which were always class specific. The art system thereby gains the advantage of making its mechanism of inclusion/exclusion largely independent of the inclusions and exclusions of other functional systems. Empirical research has no difficulty noticing that the number of people who go to concerts and visit museums and art exhi- bitions is not a representative selection of the population at large. But this bias cannot be comprehended as a result of a social regulation. Rather, it is a correlate of the evolutionary improbability of equal distribution, which manifests itself differently in mega-events, such as rock concerts, and in classical theater.
In another terminology, one might say: there are only a few, rather
Evolution
243
loose structural couplings between the art system and other systems. As before, a specialized art market couples the art system and the economic system. But in this market, artworks are traded as capital investments or as extremely expensive individual goods. On the production side, access to this market depends on an established reputation, and the market takes an active role in developing such reputations. However, one should not overestimate the irritating effects of the market on the production of art. Precisely the demand for artistic originality prevents the artist from work- ing with an eye to the market.
Compared to other intersystemic relationships--between law and pol- itics, for example, between the health care system and the economy as an employment system, or between the economy and science--the art sys- tem is surprisingly isolated. This might explain why modern art is capable of developing a symbolization of fundamental social problems of modern society that relies neither on an imitation of society's "nature" nor on a cri- tique of its effects.
Art is a "playful" doubling of reality; this is both the result and the con- dition of its evolution. But what is this enigmatic double? How can it be observed? As a unity? As a boundary one can cross without being able to occupy it? As nothing, and thus once again as something that can serve as an indication of the unobservability of the world?
This situation, which has evolved (unplanned), can be described in sev- eral ways, from which society can choose the one it likes best, the one that it finds convincing and that works for communication--whether as a sign for its essence or as a critique, depending on whether society searches for a positive or negative relationship to itself. But if there is a plurality of possible self-descriptions, why not several at once? Perhaps it is the prob- lem of a "postmodern" poly-contexturality of self-descriptions, which so- ciety at first explores in the realm of art.
? 7 Self-Description
I
One of the incalculable effects of Wittgenstein's philosophy was to raise the question of whether a concept of art can be defined. If the notion of play defies definition, then art should remain undefined as well. This view
1
was widely held in the 1960s. It denies only die possibility for a definition
that corresponds to the "essence" of art and holds unequivocally for all ob- servers, however, thus leaving a loophole for the recent theory of operative constructivism, which no longer raises issues of essence or of the consen- sus of all observers but instead leaves the decision of what counts as art to
2
the art system itself. In relation to this system, all other observers assume
the position of second-order observers. Such observers must restrict them- selves to reporting what the art system designates as art. They must leave it to the system to determine its own boundaries. This move burdens the theory of self-describing systems with a momentous inheritance. It must salvage a highly encumbered "firm" that has been dealing in "essences" and "referring signs," for which tliere is no market left.
Moreover, this means that the notion of self-description is not a consti- tutive operation--in the sense that the system needs to know what art is before it can begin to produce art. Self-description--here and in other con- texts as well--is a retrospective operation that requires die prior existence of something it can resort to. This might still leave open the possibility of characterizing self-description as a cognitive apprehension of the "essence" of art, so long as such a terminology is still acceptable. Modern art in par- ticular initially thought of itself as representing a kind of clean, purified
244
Self-Description
245
essence or quest for truth. But from the viewpoint of self-description, the appeal to "essence" or "truth" is only one possibility among others, which disintegrates under the gaze of the second-order observer. All products of self-description must be treated as contingent, even if they resist contin- gency at the semantic level. Most importantly, they must be treated as se- lective choices, incapable of retaining and representing in the system's memory the sum total of what is happening in the system.
This "modalization" of all self-descriptive propositions implies nothing yet about the limits of plausibility to which self-descriptions must adhere. Making concessions to contingency, belatedness, selectivity, or a plurality of self-descriptions does not answer the question of what is accomplished by such self-descriptions. It merely addresses this question to another au- thority--an authority that is presumably motivated by its own interests to keep arbitrariness under control and is in a better position to do so. Whether it does so remains to be seen. At any rate, works of art must be distinguishable as such. Otherwise they are perceived as objects of utility or, more recently, as trash--or they are mistaken for sacred objects, in- structional texts, and so on. As we have shown in detail, society, in order to recognize art, requires a recursive network of observations that makes use of identifiable structures to generate nonidentical reproductions. The artist must anticipate what an observer might observe as a work of art and what kind of additional information one might be able to expect (in terms of theater buildings, art exhibitions, museums, the length of lines in po- etry, and so forth). Even the individual operations one performs when ob- serving a work of art must, via other operations, refer back to themselves. They acquire their distinguishable identity only by a detour through other objects--even when, and precisely when, they are unique. Observations of art occur only in the autopoietic network of the art system. In this sense, one can speak of a basal self-reference at the level of operations that
3
cannot be decomposed further. Without this self-reference, there would
be no art. Art, in other words, is not a "composition" made up of preex- isting "autochtonous" parts that only need to be put together.
Moreover, the institutionalization of art and the establishment of sup- porting information (exhibitions and so on) require that works of art "converse" with one another, that art cite, copy, reject, renew, ironize art --that art is reproduced, no matter how, within a referential nexus that transcends the work. Today this is called "intertextuality," which is an-
4
other way of saying that the art system must have a memory. Memory is
246 Self-Description
presupposed especially when the evolution of artistic communication leads to a situation in which the artwork makes its own laws. We have called this phenomenon the self-programming ofart. Self-programming re- quires the specification of such referential networks to ensure that art is still recognized as art, despite the growing tolerance for idiosyncratic art- works. One can now identify the types of form (stylistic levels, sympho- nies, sonnets) that underlie certain formal constraints. One can identify the styles or "signatures" of a certain artist, or isolate periods in his work in which he draws on his work in a recognizable manner. If one takes this into account, then one can distinguish several levels at which observations of art are self-referentially determined ("level" here does not indicate pri- ority, in the sense that the general would be more important than the par- ticular or vice versa). Under the strenuous conditions of a complexity that increases in the course of evolution, all of these circumstances contribute to the autopoietic self-reproduction of art.
If we speak in the following of the self-description of the art system, then we presuppose these developments. The concept of self-description points in another direction, however. We must presuppose all the operations that produce a difference between art and nonart within the recursive network of these operations. We further assume the basal self-referentiality of ob- servation to be an operation. Without it, there would be nothing to be de- scribed as art. But the type of reflection that goes under the name "self- description" uses a different distinction. It refers to an other different from the system's basal self-reference, namely, to the environment and specifi- cally to the inner-social environment of the autopoietic system of art. The
theory of self-description always already assumes the existence of self- descriptions. Theoretical analysis only reiterates the system's own self-
5
descriptions.
In society, all the interpretations generated by communication come to-
gether. When art becomes visible as a distinct phenomenon, it stimulates descriptions. One wants to determine what art is all about. Since antiq- uity, there has been a literature on art. Recognizing works of art as works of art was understood to be a kind of astonishment that provokes curios- ity, or a surprise that imprints itself in memory. Such notions lack speci- ficity. One might be surprised on other occasions as well; the narratives of the religious system abound with such reports. Moreover, descriptions of this sort are not localized within the art system. They do not engage in the internal affairs of art, not in the manner in which romantic art criticism
Self-Description
247
engaged itself. They are basically philosophical texts, concerned with one
aspect of world descriptions that desire and search for truth. Accordingly,
neither in antiquity nor in the Middle Ages was there any concept for
what we today call the (fine) arts. What catches one's eye at first are par-
ticular differences between media of perception, and also between the vi-
sual arts and texts (poetry). August Wilhelm Schlegel still published his
lectures of 1801--which were conceived as a comprehensive presentation
and certainly belong to the epoch of reflection--in two volumes, under
6 the title Lectures on the Fine Arts and Literature.
It was difficult to separate out what, according to our modern under- standing, does not belong to art. When representational issues moved into the foreground, as they did in the sixteenth and seventeenth centuries, one felt compelled to include in the realm of art the beautiful appearance of good manners and benevolence (bienseance)--everything that was called morality at the time and treated in a science de mceurs. Under such condi- tions, aesthetics and morality became inseparable. The debate about the "beautiful"--whether understood as exemplary perfection, balanced pro- portion, or a refined sharpening of the intellect (acutezza, Witz)--was car- ried on for several centuries. In these terms, however, it was impossible to separate artistic beauty from natural beauty, the good looks of people, the elegance of their behavior, the eloquence of their speeches, or the dissim- ulation of their imperfections.
In retrospect one might ask: What was missing? And what would be the theoretical criterion for a self-description of the art system? Approaching the wealth of materials contained in the pertinent literature in the manner of the "intellectual historian" is not enough. We first need to clarify what we mean by self-description.
An understanding of this phenomenon has been obscured by the con- cept of "culture"--one of the most detrimental concepts ever to be in- vented. While the concept allowed one to distinguish between objective and subjective culture, both referred to an (artificial) state of affairs that was relativized by attributing it to individuals or groups. The invention of culture toward the end of the eighteenth century--of a form of reflection that subsumed under culture everything that was not nature--presup- posed this kind of relativization, which served as a basis for generating his- torical or national comparisons between cultures--an event staged by "ed- ucated Europe," as it was called in those days. Despite its comparative relativization, culture remained an object of essential propositions that
2 4 8 Self-Description
could be either true or false. What we mean by "self-description," by con- trast, refers to the mode of operation by which systems generate their in- ternal identity, whatever the observers of this process might think of it. One can certainly imagine a plurality of simultaneously generated self- descriptions; but the notion of relativity is completely inappropriate in this case. (Similarly, no relativism whatsoever is involved in making the point that some animals have tails and some don't. ) What creates prob- lems is classical bivalent logic, since the description of a self-description projects realities that differ from the ones it describes.
A definition of the concept of self-description can be quickly intro- duced.