The problems presented by the materiality of words have often been discussed, at least since Mallarme and frequendy with
reference
to him.
Niklas Luhmann - Art of the Social System
Ornaments organize space and time directly and supply these media with redundancy and variety.
They presuppose a self-defined space that is closed off, as it were, from within.
The same is true when time be- comes ornament (in dance, or in the ups and downs of tension in narra- tive).
From the viewpoint of the creative process, such an enclosure must first be created in the form of a partial space prepared specifically for this purpose (such as the facade of a building or the surface of a container), or
29
as a slice of time with a self-determined beginning and end.
tional art, by contrast, begins by projecting an imaginary space or time in order to gain a free hand in employing this self-created medium for pur- poses of ornamentation and representation. Since the early Renaissance, European art has preferred this option, relegating ornament to the func- tion of decoration, of placing accents and emphasizing the essential. A closer look, however, reveals that the ornamental is always the work's in- frastructure, even in deliberately representational art forms, for the fol- lowing reason: if one employs space and time (how else could an artwork appear? ), then these media must be organized, whatever might subse- quently be represented in them.
IV
No matter how we place the accents, no matter how much a work's fig- urative or ornamental aspects may capture our attention at first, we must assume that the forms that constitute the work by virtue of their distin- guishing force diverge, depending on which medium of perception or in-
30
tuition they deploy. There is no commensurability between painting and
music, or between sculpture and dance, or between the lyric and the novel--which is not to say that there can be no "onomatopoeia" in music or that dancers cannot become sculptures. These formal differences are conditioned, not by the choice of form (if they were, then they could be avoided), but by the medium that underlies each form and that, by virtue of its loose coupling, makes tight couplings possible. Perceptual media do not display a spectrum as broad as that of artistic genres. Painting and
Representa-
n6 Medium andForm
sculpture, theater and dance, depend on light--a visual medium--whereas the lyric, like narrative (the epic, the novel), relies on language as a medium of fixation for intuition. But the manner in which art shapes and makes use of perceptual media differs in each genre. To radicalize our previous question: How can one speak of the unity of art, or of the unity of an artis- tic medium in view of such diversity?
In anticipating this question, we have taken great care to elaborate the distinction between medium and form. We can start from there. Percep- tual media are media of the psyche in Heider's sense, rather than of a so- cial or communicative nature. No one can share the perception of others, although one can perceive that others perceive or that others perceive one's perceiving. We shall make the controversial claim that the same is true in the realm of intuition, that is, in the realm of imagined perception. It is true when language is used for the purpose of stimulating intuition
(rather than communicating information) and thus holds also for art- works created in language, especially the novel. No one knows what an- other experiences intuitively when reading about Odysseus and how he had himself tied to the mast, or about Siebenkas, who searches for his own
31
grave only to find the woman he will eventually marry.
how others imagine Robinson's surprise when Friday appears, or how Napoleon (in War and Peace) is struck by the events of the Russian cam- paign. One tends to speak of "fictional" literature in these cases. Whatever that means, the fictional medium is primarily constituted by the private nature of an intuition that requires no "continuation of communication" and therefore no special effort on the part of consciousness and memory, but instead allows them to operate freely.
How, then, is art nonetheless possible as communication? And what, in this case, would be the medium of communication?
The key to this question may well reside in the intentionally created observational relationships we analyzed in the previous chapter. Once
? someone (no matter who) recognizes, from the manner of presentation, \ an arrangement that is produced for an observer, a social medium has \ come into existence--whether or not this aspect is communicated in the
work of art. Literary texts in particular often distinguish themselves by self-referential clues of this sort. (Incorporating the text's production into the text, addressing the reader, attacking the reviewers in the manner of Jean Paul are still rather crude stylistic means, aimed at differentiating the text at the level of an observation of observations. ) In the wake of these
No one knows
Medium and Form
117
developments, it becomes possible to establish an "artificial" form that si- multaneously serves as a medium for forms within the form--such as the space contained in a painting, the potential movement of a sculpture ar- rested in time, or the realm of possible events in which a narrative estab- lishes sequences that present in tightly coupled forms--in this and no other way--what might well turn out otherwise. Or consider the decep- tive maneuvers of baroque architecture, which are enjoyable only when seen through, or dance, which does not take the direction of its move- ment from ordinary life but instead presents it so that it appears, from moment to moment, as if it had been selected only for the sake of the dance.
Although perceptual media and artistic genres differ greatly with regard to their concrete materialization, they share a common ground in the manner in which they construct novel medium/form relations that are in- tended to be observed and are intelligible only when this is understood. The unity of art resides in that it creates for the sake of observation and observes for the sake of being observed, and the medium of art consists in the freedom to create medium/form relations.
The possibility for combining forms and thereby tightening the work, as it were, from within, suggests another parallel between individual gen- res. We recall that forms are always two-sided. No matter what they indi- cate and fixate in the work of art, they always simultaneously present an- other side that must cooperate in rendering visible what is determined by the work. This is true if the artwork itself is meant to be recognized as a specific object (and nothing else). It holds for every detail that constitutes the work in collaboration with others.
The unmarked space is the indispensable other side, a reference to pos- sibilities that, for their part, point to an infinity that cannot be contained
32
in one place.
unmarked space into a marked space, and it creates a boundary by cross- ing that boundary. Spencer Brown speaks of "drawing a distinction. " At die same time, a difference between medium and form comes into being, an enclosed, specially prepared marked space, in which the artwork fol- lows the pull of its own distinctions and determines its own forms.
The determination of one side does not entirely leave open what can happen on the other side--this accounts for the specificity of individual art forms. While it does not determine the other side, it renders it nonar- bitrary. What can happen there must "fit" or it will cause dissonance, a
The first step in the making of an artwork leads from the
n8 Medium and Form
flaw, or disruption (which can, of course, be intentional, in which case it requires a balancing fit on its part). The determination of one side, as we just stated, does not determine the other. But it facilitates decisions and the observation of decisions concerning what can happen or has been estab- lished on the other side by the artist. If an indication is to become a work of art, then the other side must remain accessible--which assumes that one can demarcate another unmarked space in relation to this other side.
Whenever an object is intended to be a work of art, the indication does not refer merely to itself (to this and no other object). It also refers to the crossing of the boundary that divides the form into two halves and in- structs the observer to search for and fix what has not yet been decided. This holds for the artist as well as for the observer of art, and ultimately for any observation that depends on time. The indication, we might say, is used as meaning. Crossing the boundary never leads into the unmarked space, never into the world as such, but always executes an indication, a new indication. As we know, indications can only be executed as distinc- tions. They specify only one side of (another) distinction that, for its part,
33
has its own other side. This is why an artwork cannot reject the world.
do so, it would first have to indicate and distinguish the world, that is to say, it would have to execute operations that can occur only in the world.
In this sense, the artwork forces both the artist and the observer to ad-
vance from one form to the next in order to return eventually to the form
from which they began, which is now the other side of another form.
Forms play with forms, but the play remains formal. It never arrives at
"matter," it never serves as a sign for something else. H Each formal deter-
mination functions simultaneously as an irritation that leaves room for
subsequent decisions, and advancing from one form to the next is an ex-
periment that either succeeds or fails. ? This is why, as we shall elaborate be-
low, a "code" emerges in art, a continuously maintained binary orientation
concerning the "fit" or "lack of fit" of forms. This is why every artwork
contains "information" in Gregory Bateson's sense--differences that make 35
a difference. And all of this holds for any kind of art!
Accordingly, for all genres the medium of art is the sum total of possible
ways of crossing form boundaries (distinctions) from within toward the outside and of discovering fitting indications on the other side that stimu- late further crossings by virtue of their own boundaries. The medium of art is present in every artwork, yet it is invisible, since it operates only on the other side--the one not indicated--as a kind of attractor for further
To
Medium and Form 119
observations. The process of discovery transforms the medium into form. Or else one fails. In working together, form and medium generate what characterizes successful artworks, namely, improbable evidence.
Does this always have to be another distinction with another side? One may well imagine an artwork that has two components but only one form (or, more accurately, a precise congruence of two inverse forms that over- lap one another). In this case, one focuses on the side that is the other side of the other side and vice versa. The formal asymmetry necessary for ob- servation is canceled in symmetry. One can only oscillate between die two sides. Any further dynamic is blocked. We have, in other words, the pre- cise image of a logical paradox, a kind of time trap that compels the ob- server to move back and forth between its poles in the shortest possible in- stance. This is not meaningless, nor is a logical paradox meaningless. But the meaning of such a figure--of its form--must be sought in the clue that allows for unfolding this paradox and reintroduces asymmetry into the form.
Paradoxes present the world in the form of a self-blocking observation. It is certainly possible to stage works of art as paradoxes, but only in order to demonstrate what doesn't work, only to symbolize the unobservability of the world. Works consisting of only two components are not yet gen- uine artworks. But to the extent that they carry along the other side of the
36
form as an empty form, as mere exteriority, they are works of art after all. This is particularly true where paradox becomes the work's theme--as in Escher's etchings, in Magritte's work, or in certain forms of sixteenth- and
37
seventeenth-century lyric poetry, especially that of John Donne.
cases, the paradoxical oscillation of truth is introduced deliberately--not to represent the world, but to invite the viewer or reader to search for an innovative exit that remains undetermined in the work and about which
38
even the artist himself may have his doubts.
may work together to accomplish this effect, paradox remains the super- form, the frame suggestive of what remains unsaid in the work and what is only marked as unsaid.
Let us return to the ordinary. What remains open as the other side of a form is generally determined by an indication that actualizes another dis- tinction for which the same is true. This process keeps going until a system of references closes itself off, in which nothing remains undetermined. Oc- casionally, things may go wrong in the process, and the remaining discord must be covered up or minimized. This is a fact of life. Our concern here
While several aesthetic forms
In these
IZO Medium and Form
is not art criticism but the principle of form, the process of observing (the making and observing) of artworks.
It should be clear by now that this analysis precludes comprehending an
artwork in terms of the relation between a whole and its parts. Dividing a
work and judging die relationship between parts misses its internal nexus.
Nor are we concerned with the primacy of the whole in relation to its
39
parts.
bution to the work consists in what they are not, what they make available for further elaboration. The artwork closes itself off by reusing what is al- ready determined in the work as the odier side of further distinctions. The result is a unique, circular accumulation of meaning, which often escapes one's first view (or is grasped only "intuitively"). Eventually, it may turn out that a determination plays a role in several distinctions at once, that it is multifunctional and therefore not interchangeable. This creates an over- all impression of necessity--the work is what it is, even though it is made, individual, and contingent, rather than necessary in an ontological sense. The work of art, one might say, manages to overcome its own contingency.
Several more or less standardized, genre-specific formal models are available for this process. The basic form for generating forms from odier
40
forms is the (misleadingly so called) ornament.
on the problem of broken symmetry, on the problem of form. At stake here is the projection of asymmetries that still exhibit traces of the sym- metries from which diey emerged. Ornaments are recursions tliat keep go- ing by recalling previous and anticipating further forms. They display the
41
unity of redundancy and variety. Transitions are effaced; at least they are
not emphasized as breaks, since each place in an ornament is at the same time the place of another place. This includes the repetition of forms at other points in space or time, whereby their positional difference suggests nonidentity within the ornament's overall identity. Continuation is the principle that integrates what is excluded at first, that defines the excluded as a motif, reiterates it, or connects other motifs. This clearly shows that art is neither a sign for something else nor the mere form of the material. The ornament generates its own imaginary space by continuously trans- forming formal boundaries into transitions that have more than one meaning. It prevents the work from falling apart into isolated figures, on which one can focus or from which one can turn away. The ornament, in other words, holds the artwork together, precisely because it does not par- take in its figurative division. One can call this "mere decoration" only in
If one wants to isolate parts, then one discovers that their contri-
All ornaments are based
Medium and Form 121
social situations in which the differentiation of art is already under way, whereas decorative ornaments also adorn utilitarian objects, jewelry, sa- cred objects, or "crafts," so that mere decoration must be distinguished from art. In such a situation, the ornamental structure can be granted only
42
a subordinate role.
supplements his observation by referring to the masking function of dec-
43
oration:
tinct and enables faster comprehension; on the other hand, it suppresses contradictory, confusing information. The ornament accomplishes all this with a luxurious excess that tends to become an end in itself.
The basic idea is still one of "hierarchical opposition," of a contrast be- tween balanced proportion and mere adornment. One expects art to keep decoration under control. This places the burden on the dominant side of the distinction, on what is supposedly balanced proportion or is later called symbolic meaning. But the inability to resolve the problem of hier- archy eventually "deconstructs" the distinction itself. The ornament, ini- tially meant to fulfill a secondary function, takes on the burden of sup- plying meaning. If one wants to observe artworks as art with reference to their play of forms, then one must pay attention to their ornaments.
Only then can one return to the question of how the work is made and which secondary meanings serve the ornament while receiving from the ornament the electrical charge tliat accounts for their artistic quality. Painting, too, eventually pushes its ornaments to the margins or into the background--which needs to be filled anyway--in order to foreground its figures. Assisted by unified perspective, painting develops the back- ground into an open space--a landscape, for example--only to discover the need to compensate for the ornaments function by filling its imagi- nary space with nonarbitrary objects until, in the end, even the landscape
44
is no longer necessary. While the ornament is marginalized as mere dec-
oration that can also adorn nonartistic objects, a functional equivalent of the ornament emerges from within artworks themselves--an inner "line
45
of beauty"
more distinctly curved and of much greater condensation than any nat- ural line. Even as ornamental adornment is excluded from art, the orna- ment is reborn within the work. Similarly, in poetry sound and rhythm are more easily replaced by meanings, which requires that the play with distinctions be reproduced as a formal nexus among narrative elements. By indicating an action, narrative, for example, can serve two distinctions
Gombrich speaks of an "explanatory division" and
on the one hand, decoration renders the information more dis-
that joins what has been separated figuratively, a line that is
122 Medium and Form
The au-
The intense use of these possibilities leads to the modern novel of the
47
individual. "Flat" heroes become "round,"
parent, the strength of motives is demonstrated (typically those preferred by the author, such as the profit motive in Robinson Crusoe or Moll Flan- ders). At the same time, narrative can induce changes in a character; it can motivate learning, religious conversion, or regret; and it can recommend its result to the reader as a laudable attitude. This development allows the novel to free itself from moral guidelines, and the reader is confronted
48
at once: it characterizes the agent and moves the plot forward.
thor arranges the events, that is, he shapes the medium in such a way that the reader can infer the changing mental state of the hero. The fictionality of the arrangement hides behind the contingency ofevents and actions that must serve as the starting point for the reader who follows the narrative.
with live models and experiences that could be his own.
mal combination of character and plot by means of actions--a combina- tion that informs both of these distinctions--has established itself and guides the reader's expectations, it becomes possible to write narratives that still present themselves as novels but break with this combination. These narratives seek to distinguish themselves by precluding any infer- ences concerning the character and motives of the hero or by refusing to move the plot along by means of actions. This development begins as early as Flaubert's L'Mucation sentimentale (1869).
The internal ornament serves the artworks self-description; it beautifies
49
because it is beautiful.
it can bind. On the one hand, the form combination that is selected indi- vidualizes the artwork and qualifies it as a distinct object. This allows for technical reproduction, which neither affects the work's recognizability nor "damages" its individual form, but instead makes it more accessible. Technical reproduction, in turn, generates a new form: the distinction be- tween original and copy. On the other hand, observing how the work is
50 made yields an observation of a more general type often called "style. " At
the level of stylistic forms, the art system is able to evolve; it can replace form combinations that have already been tested or derive new forms from the rejection of what has become all too familiar. Moreover, the sys^ tem can turn rejection itself into a form that is intelligible only to those aware of what used to be customary and what, accordingly, is the expec-
51
tation to be disappointed.
itself the avant-garde has taken this backward-looking manner to an ex-
their motives become trans-
It absorbs as much variety as possible, as much as
The movement that, strangely enough, calls
46
Once the for-
Medium and Form 123
treme--like oarsmen, who face the direction they are coming from and have the goal of their journey behind their backs.
We will not elaborate on these reflections. Instead, we would note that situations of this sort--in any artistic genre--presuppose the artwork as a specific combination of forms, no matter how this presupposition is then attacked in the search for new forms. This protest is possible only when (and so long as! ) a loosely coupled medium is available for the creation of possible forms. If the previous reflections hold, then this medium may well reside in the need for another, yet undefined side on the other side of the indications that make up an artwork. This need would explain why an artwork, in closing itself off and in determining its open side by means of other distinctions, coagulates the medium in such a way that a tightly coupled form emerges, though the medium always reproduces itself at the same time--that is, the question of what lies on the other side of the form is posed anew in each instance.
V
We began with the medium of perception, mentioning the medium of
language only in passing. But in order to demonstrate the unity of art and
the formal similarity of all artistic genres, we must pay attention to the
verbal arts as well, especially to poetry. Here, we are dealing with words as
medium, as a loosely coupled set of elements. Words serve not only as el-
ements to be coupled but also as a means for the artwork's self-descrip-
tion, a means of bringing about the unity of the description and the de-
52
scribed.
formal combination of words that calls upon words to display an unusual meaning--whether or not it relies on the structure of grammatically cor- rect sentences. This can happen only if distinctions are placed in words. There is no need to paraphrase these distinctions; nor can they be para- phrased, for to do so would create the possibility for rejection--just what art wants to avoid.
Ordinary language uses the same words in multiple contexts and there- fore relies on wearing down their meanings and on phrases as an aid to understanding. It seeks to establish unambiguous denotations and ac- complishes this goal via naming and the construction of abstract objects, conceptual correlates, or ideas. Poetic language operates in reverse--with or without the aid of phrases. It reflects on the usage of language--as if
Accordingly, the creation of poetic form consists in a special
124 Medium and Form
language were one material among others that one finds in the world.
54
53
Poetry has no use for denotations. Instead, it relies on connotations, ploying words as a medium in which the connotations that select one an- other assume form. Rather than forcing the diffuse referential richness of words into the most unambiguous relation to facts in the world, poetry posits this richness (including the resonance of what remains unsaid) over against the facts. Just as atoms alter their internal electronics when com- bining into molecules, poetry modifies the meaning of words. It may gen- erate striking new nuances or create estrangement, but it can also surprise by restoring the original meaning of everyday words. Overgeneralized forms are pried open and reconstructed. Poetry contextualizes such forms so that they become structurally determined and can no longer easily be recognized. Like any use of form, this strategy aims at what it excludes. Other words can pick up what remains unsaid, but they can also serve as a reminder that many important things must be left unsaid. The other side of the form always plays along--as a boundary that can either be crossed by further guided observations or else is fixed, again and again, as the same boundary, as the unmarked space of different words.
Poetry shifts from a denotative to a connotative use of words because of its need for poetic closure--a closure for which the poetic genre does not choose the form of narrative. Referential meaning would refer the reader into the world, where it would lose itself among the multiple references
55
that make up the meaning of reality. The poem becomes a unity only at
the level of connotation, by exploring the liberties that come with using
56
words exclusively as a medium.
can be secured at this level, forms one must relinquish when moving from one poem to the next. Poetry requires an intensification of memory, that
57
This means that only temporary forms
is, a retroactive reading (if one still wants to call this "reading"
and reader must leave behind the linearity of the text and apprehend its structure in a circular manner; they must be able to take apart the net- work of multiple circles that makes up a text. But such an effort is feasible and can be accomplished only within a single poem.
Connotations start out from the familiar meanings of words, truncating only their hetero-reference so that opposites, for example, can appear as a unity, even though and even because this is not how they appear in the ex- ternal world. This technique requires a suspension of the ordinary referen- tial meaning of words, which catches the reader off guard. Last, but not least, this is possible because of the ornamental quality of verbal constella-
). Author
em-
Medium and Form 125
tions. This quality may reside in the sound of words (nevermore 2nd vast [English in the original] are well-known examples), or it may be an effect of the relation between short and long syllables or of repetitions, echoes, stereotypes, contrasts, anagrams. In Finnegans Wake, the ornamental qual- ity of language, the resonance between sounds and other words, over- whelms the text to an extent that intelligible words communicate only that they do not matter. Rhythms are complicated to the point where they es- cape reading altogether and require recitation to be apprehended. To con- vince, poetry appeals to perception, not to thinking. And the function of the ornament, in poetry as elsewhere, is to intensify redundancy and vari- ety in ways that would otherwise hardly be possible.
Poetry, then, is not just rhymed prose. If one reads poetry as a sequence of propositions about the world and considers the poetic only as beautifi- cation, adornment, or decoration, one does not observe it as a work of art. Nor can one apprehend in this way the formal combination the poet uses
58
to compose his work. Only at the level where symbols,
ings, and rhythms conspire--a level that is difficult to "read"--do poems refer to themselves in the process of creating forms. They generate contex- tual dependencies, ironic references, and paradoxes, all of which refer back to the text that produces these effects. Supported by the text, poetic self-
59
reference may eventually articulate itself explicitly
statement, but as a form within the nexus of forms that constitute the text.
The problems presented by the materiality of words have often been discussed, at least since Mallarme and frequendy with reference to him. It seemed plausible to locate the problem in the relationship between con- sciousness and language, in the poet's access to language, or in the self- sacrifice commanded by the shaping of language. This general insight, which ultimately points to the distinction between psychic and social sys- tems, can be supplemented by the distinction between medium and form. This distinction is a projection of art, here a projection of poetry, a form of its autopoiesis. It is given neither as matter nor as Spirit. It has no on- tological substrate, which makes it tempting to observe how observers handle this distinction.
All of this needs further elaboration. William Empson and Cleanth Brooks brought these problems to the attention of a literary theory that
60
calls itself "critical. "
form, this awareness yielded only the formal analyses of the New Literary Criticism and subsequently led to the critique of what these analyses ig-
But instead of producing a general terminology of
sounds, mean-
--not as a flat, abrupt
Medium and Form
nored.
literature, the promise of a unified theory of artistic kinds went unful- filled. Currently, however, literary theory is sufficiently receptive to inter- disciplinary suggestions that this separation is unlikely to prevail. Thus af- ter this excursion into the special domain of the medium of language, we return to more general analyses.
VI
The medium of art renders the creation of forms at once possible and improbable. The medium always contains other possibilities and makes everything determined appear to be contingent. This improbability is em- phasized when everyday purposes and utilities are bracketed as the guid- ing principles of observation. Artistic form (backed by aesthetic reflection) goes out of its way not to appear useful.
In this way, the artwork directs the beholder's awareness toward the im- probability of its emergence. If attention is drawn to poetic constructions,
then it is only because they do not seem very likely, whereas the likelihood of using other constructions is, on the contrary, very high. "Poetic is that
62
which has not become law," writes Julia Kristeva.
texts, one might add that their improbability must not be based on their informational value, which always implies a certain quality of surprise, but consists in their renunciation of information in the sense of mundane utility.
During the past two centuries, a number of doctrines derived from diis enforced improbability surfaced in poetic reflection--the rejection of a rule-based poetics, for example, or the emphasis on the individuality and originality of an authentic artwork, and eventually the search for an alter- native explanation of the improbable that points to the "genius" of the artists. But these are collateral circumstances, secondary phenomena that accompany efforts to come to terms with the improbable. When focusing on the improbability of form itself, one is primarily concerned with the observer's fascination, his staying-put-with-the-work in a sequence of ob- servations that attempt to decipher it.
One might expect the sequence of medium-form-medium-form for- mations progressively to constrain the medium's possibilities, thus leading to an increase in redundancy. Many potential sculptures come to mind if one considers only the media of space and material. When one represents
126
61
Caught in the conventional division between die visual arts and
Especially for poetic
Medium and Form 127
a mobile living being, the limitations of its body constrain what can be rendered. Lessing's analysis of the Laocobn shows that the artist is not en- tirely free in selecting the moment from which the before and after of movements can be rendered visible. Once we are dealing with the Dying Gaulor with stage productions of Lucia di Lammermoor, very few perfor- mances are conceivable that could give form to precisely this medium. The improbability of the composition must be wrested from these limita- tions. This may result in a "structural drift" within the art system that turns into an end in itself not only the artwork but above all its improba- bility. Eventually, one begins to experiment with the idea of declaring everything a work of art so long as an artistic claim can be asserted and maintained. The work's probability then boils down to the credibility of such a claim.
But artistic credibility is still a relationship between medium and form. The difficulty of creating forms shifts to the difficulty of claiming a work as art and maintaining this claim. Yet the medium remains a medium of art by virtue of its ties to the history of art; it continues to function as a medium that propels the historical machinery of the art system beyond its current state with new and ever more daring forms. The medium might absorb decontextualized historical references, as it does in postmodernism, whereby the improbability resides precisely in this decontextualization, in free selection from a historical reservoir of forms. What used to be bound historically is now up for grabs on the condition that it remain recogniz- able as such. One might equally well continue the project of the avant- garde in an effort to expand, via the production of art, the concept of art itself. In both cases, art turns into the artistic medium insofar as, and so long as, it is capable of making the observer recognize the improbable as improbable. In the end, the observer might even be challenged to com- prehend the incomprehensibility--created especially for him--of an art- work as a reference to an incomprehensible world.
The recognition that every form is a form-in-a-medium dates back to romanticism. A medium suited for artistic forms had to be sought, dis- covered, and eventually constructed by dismanding interpretive aids taken from everyday life. The fairytale-like incredibility of backdrops served this purpose and simultaneously indicated that henceforth only absolute self- reflection--a reflection that includes the observer--could be presupposed
63
as the ultimate medium. Fantastic art,
the events and forms presented can be explained naturally, is subject to
by leaving open whether or not
128 Medium and Form
similar conditions. But the general situation was still conceived in terms of the subject, an authority that underlies itself and everything else. In the meantime, the dynamic of the art system has evolved in ways that can no longer be attributed to a subject. The observer--the self-reference under- lying all distinctions and the distinctions underlying all self-reference-- seems a more appropriate basic figure. From the position of the observer, one can unfold this circle by distinguishing the distinction from the indi- cation of one of its sides, and self-reference from hetero-reference. This distinction specifies the operation of observing as a distinguishing indica- tion and defines the concept of the self-referential system as a system that copies the operatively generated distinction between system and environ- ment into itself and bases its observing operations on the distinction be- tween self-reference and hetero-reference.
Once the "subject" gives way to the observer, we no longer need the counterconcept of an object. Regarding the manner of operation, a wide range of possibilities is now conceivable besides intentional awareness (a state of consciousness). The observer can be a social system, and observa- tion can be communication. The artwork is not necessarily a device that causes the perspectives of producer and observer (or of theories of produc- tion and reception) to oscillate. Still, nothing speaks against starting out from references to psychic systems, either to the artist or the observer. But the emergent unity of the art system and its unique medium cannot be grasped in this manner. The art system is a special system of social com- munication. It has its own self-reference and hetero-references tliat indi- cate forms existing exclusively in a medium unique to art. This medium is the improbability of the combinatory structure of form that art wrests from everyday life and that refers the observer to other observers.
These reflections eventually raise the question of whether an artwork
64
has to be difficult and, if so, why. Like everything else, this proposition
can be questioned today, and there is a tendency to separate art from craftsmanship. Pushed to its extreme, difficulty might ultimately boil down to the problem of how one can work as an artist in a manner that is still recognizable. Pointing to the essence of art--to the idea of art, the rarity of genius, or the like--is of no help in this matter. The question, rather, is whether and for what reasons the mediums potential for creat- ing forms must be limited, and how this limitation is accomplished.
Within a theory of symbolically generalized media, Talcott Parsons as- sumed that each of these media, just like money, requires a real backing
Medium and Form 129
that can be overdrawn by confidence but not expanded at will. Using the
medium below or in excess of its capacities is certainly possible, but it
leads to an inflation or deflation of the medium that jeopardizes its func-
65
tioning.
ing of art, especially of modern art? Obviously, nothing external to the medium can fulfill this function; what backs the medium of art is the works triumph over its own improbability.
This is why the trend toward facilitating the creation of forms and re- ducing forms to simple distinctions cannot be countered by judgments of taste or values. Even tlie concept of art apparendy no longer sets limits to what can count as art. But one can know that the medium/form dynamic requires constraints and that expansive trends lead to inflation. How much inflation the art system can tolerate boils down to an empirical question. Sanctions are evident not in the reaction against violations of the norm but in the loss of interest in the observation of observations.
VII
As we suggested earlier, the distinction between medium and form is based on a complex relationship to time. On the one hand, medium and form must be actualized simultaneously. On the other hand, the medium reproduces itself only by alternating the forms that an observer employs as distinctions. The stability of the medium rests on the instability of forms that repeatedly realize and dissolve tightly coupled relationships. Media are constant, forms alternate. In the basic medium of meaning, all other media are subject to variation as well, but only to the extent that they are observed as forms in another medium.
The paradoxical "simultaneity" of invariance and variability corre- sponds to the general problem of structuring the autopoietic reproduction of systems. Only actualized elements that assume the form of events (op- erations) can reproduce the system, which requires a recursive recapitula- tion of the past and anticipation of future events; in other words, nonac- tualized events must be actualized as nonactual. The actualization of the nonactual requires (and is made possible by) a selectivity that employs the logic of the distinguishing indication. Selections capable of actualizing the nonactual always function as structures--in the moment of their actual-
66
ization--by virtue of references that transcend the actual.
In art, the work's material substratum guarantees that the observing op-
If we follow Parsons's suggestion, then what would be the back-
130 Medium and Form
erations can be repeated. It ensures that the potential for repetition is per- ceived along with the work and that it actualizes what is momentarily nonactual. At the same time, the work indicates the nonidentity of the repetition. We are aware that we experience the same work (without ques- tioning its sameness) differently each time--for example, as recognizable or familiar, as confirming our attitudes rather than presenting us with as- tonishing information. Redundancy and variation collaborate in their ef- fects. Repetition alters what is repeated--especially when the repeated content is recognized and affirmed as the same. Identity is necessary--but only to allow for the nonidentical reproduction of the observing opera- tion. Observational sequences can build up comfortable redundancies and suppress provoking irritations; they can search for confirmation in one as- pect of the work and find it in another. In the visual arts, the stability of the material secures this process. In texts, writing--and in music, the re- peatability of the production (with or without notation)--does the same. We need not pursue the details of this art-external (material, memorylike) anchorage here; what should be stressed is that it requires a separation of individual artworks. The horizon of reference must be interrupted in or- der to allow for recursion, for the return to the same, and for the struc- turing anticipation of this return. But if this is true, doesn't the art system disintegrate into a disjointed ensemble of individual works?
This question enforces the recurrence of the temporal problem at the level of the systems autopoiesis, a level that transcends the individual art- work. At this level, the temporal paradox of structuring, the paradoxical actuality of the nonactual, recurs at a higher level. It is no surprise that un- folding the paradox once again boils down to a distinction--not between externally secured constants and the fluidity of observation, but between change and conservation in what counts as art.
To observe the changes in a domain common to many artworks, the (historical) concept of style has been available since the last three decades
67
Long before that, the concept of style had been 68
of the eighteenth century.
used to designate ways in which the elements of an artwork are coupled. Rhetoric, following a general trend of hierarchization, proposed a ranking
69
of styles and prescribed styles according to the dignity of their objects. Not until Winckelmann was the concept of style--which concerns factual differences such as "writing," manner, and presentation--anchored in a temporal dimension and claimed to reveal (and cause) historical differ- ences. The distinctions employed by artworks, the "against what? " of their
Medium and Form 131
manner, were subjected to the pressure of innovation. Not only did indi- vidual works have to distinguish themselves, but what did not distinguish them needed to be distinguishable at another level of comparison, as be- ing sanctioned by the concept of style. One expected style to legislate it- self--not to succumb to a prescribed canon, but rather to distinguish itself by deviating from models. A style prolongs the half-life of the publics in- terest in a work of art; one is reminded of similarities in other works and can observe each work anew with reference to similarities and differences. Style respects tradition by deviating from it. Deviation is a specific form of acknowledging relevance; it is not indifference or ignorance. Deviation re- quires knowledge of the subject matter, circumspection, and precision in selecting aspects where deviation matters, and often it is necessary to re- formulate the unity of the preceding style with disregard for what was rel- evant and accessible to this style. The procedure is a typical case of recur- sive reconstruction!
At the same time, there is the opposite trend: to hold on to what is worth preserving, precisely because of deviation. Objects are put into mu- seums or, when this is not possible--as with textual art--they are identi-
70
fied as "timeless" classics. Museums grow out of processes that decide
what is and what is not accepted. Today, even the most recent art can be defined as given (= already dated) by the mere fact of being accepted and displayed in a museum. The decision observes observers, that is, it belongs to the level of second-order observation. Classicism, too, is a construct,
71
created by observers for other observers,
has always been to reverse time: in contrast to other works, classical works
72
improve with time.
from stylistic change, which is of no significance whatever unless there is something against which the preservation of worthy objects is directed, namely, the perpetual historicization of styles. The notion, implicit in the idea of style, "that one is no longer able and will never again be able to work that way" reinforces the conservation of resources that are no longer reproduced, and each loss becomes an "irreplaceable" loss. One needs in-
73
stitutions of mourning, of the "nevermore" [English in the original]. This diagnosis shows that even at this level the paradoxical unity of the distinction between medium and form seeks identifications, plausible dis- tinctions that can be carried on and prove one another. Style as form, the museum as form, classicism as form: all of these forms respond to the same fundamental situation (concealed by forms themselves), namely, that loose
and the intent of this construct
Museums and the classics symbolize an art removed
132 Medium and Form
and tight couplings are reproduced simultaneously in a manner that is in- variant and invisible in the medium and variable and visible in the form. What reacts to this situation is not a supermeaning, a principle of art, an ultimate, convincing idea, but yet another distinction plausible enough to enable convincing identifications. The form of "style" processes the bur- den of innovation and along with it the temporality of all forms while cast- ing a secret glance toward an eternal life beyond its own time. The form of the museum and the form of classicism live off a work's ability to outlast changing styles and find therein its meaning.
Although art collections have been around for a long time and there have always been preferred authors and composers, the museum and clas- sicism as forms of conservation presuppose an art system that operates at the level of second-order observation. It is therefore no accident that these forms, along with a historicized concept of style, appear in the final dec- ades of the eighteenth century--at a time when the differentiation of the art system reaches the level of second-order observation, when it estab- lishes itself and begins to solve its problems at this level. Now one begins to inquire into the unity of the arts regardless of the different media of perception in which they realize their primary forms. Only now is art, no matter what kind, defined in temporal and historical terms. This period also introduces a reflexive concept of culture: it situates culture within the context of historical and regional ("national") comparisons for the pur- pose of self-evaluation. Once the game of observation is played at this level, it finds rules and opportunities for self-affirmation, which, for the time being, provide sufficient orientation. An "analytical" terminology ca- pable of more rigorous analyses is nowhere in sight. At any rate, such a
terminology would only reveal the paradox that informs any operation with distinctions.
? 4 The Function of Art and the Differentiation of the Art System
I
One of the few constants in the century-old academic history of sociol- ogy is the assumption that modern society is characterized by a certain de-
1
gree of social differentiation and by some unique form of differentiation. Historical development, it turns out, is subject not only to differentiation
2
but also to de-differentiation.
ory and the theory of evolution have altered the manner in which the the- orem of differentiation is justified and how it is conceptually framed. To- day, one no longer works with analogies based on the paradigm of the division of labor, which was believed to occur spontaneously whenever there was an opportunity, simply because of its yields or its productive ra- tionality. It is doubtful that more differentiation, or differentiation at the level of the division of labor, is generally desirable. (Adam Smith already pointed out its disadvantages. ) Currendy, an overall critical, more skepti- cal, and doubtful attitude prevails. This does not change the fact that the theorem of differentiation posits a crucial accent, if not the main criterion for distinguishing modern society from its predecessors. However, if dif- ferentiation in its specifically modern form turns out to be not as benefi- cial as was previously assumed, then one needs to revise one's judgment of modern society. Many indications point toward this.
A first step--which hardly improves our relation to the tradition--is to describe modern society as a functionally differentiated system. Generally speaking, this means that the orientation toward specific functions (or problems) of the social system catalyzes the formation of subsystems that
Moreover, the elaboration of systems the-
m
134 The Function of Art and the Differentiation of the Art System
dominate the face of society. If one wants to investigate the consequences of this development and wants to know how differentiation affects the subdomains of social communication (in this case art), then one must fo- cus one's conceptual apparatus more accurately. Most importantly, one must clarify how functions can serve as evolutionary "attractors," and in
3
precisely what sense subsystems constitute systems in their own right. A description of the art system that takes these background assump- tions for granted and analyzes the form of social differentiation in general systems-theoretical terms yields consequences that will accompany us from now on. Today, systems theory is a highly developed, albeit contro- versial, analytical instrument. It requires theoretical decisions that do not directly concern art. (This, of course, holds for other--for example, semi- ological--analyses of art as well. ) In conjunction with the thesis that soci- ety is a functionally differentiated system and is in this form historically unique, a systems-theoretical orientation has further consequences. It means that the different functional systems are treated in many respects as comparable. The terminology we introduced earlier demonstrates this in remarkable detail. Issues such as system formation and system boundaries, function, medium and forms, operative closure, autopoiesis, first- and second-order observation, and coding and programming can be investi- gated with regard to any functional system. As these investigations take shape and yield answers, a theory of society emerges that does not depend on discovering a unified meaning behind society--for example, by deriv- ing societies from the nature of man, from a founding contract, or from an ultimate moral consensus. Such propositions may be treated as part of the theory's subject matter, as different forms of self-description available to the system of society. What ultimately characterizes society, however,
4
manifests itself in the comparability of its subsystems.
In a domain such as art (just as for law, science, politics, and so on), we
discover not unique traits of art but features that can be found, mutatis mutandis, in other functional systems as well--for example, the shift to a mode of second-order observation. Art participates in society by differen- tiating itself as a system, which subjects art to a logic of operative clo- sure--just like any other functional system. We are not primarily con- cerned with problems of causality, of society's influence on art and of art on society. (Such issues are of secondary importance. ) Nor do we advocate the defensive attitude that die autonomy of art ought to be upheld and protected. Modern art is autonomous in an operative sense. No one else
The Function of Art and the Differentiation of the Art System 135
does what it does. This is why questions concerning die independence or dependence of art can arise in a causal sense. The societal nature of mod- ern art consists in its operative closure and autonomy, provided that soci- ety imposes this form on all functional systems, one of which is art.
We base the following analyses on a distinction, namely, on the distinc- tion between system/environment relations, on the one hand, and system/ system relations, on the other. When dealing with system/environment re- lations, the system constitutes the internal side of the form, whereas the environment is its unmarked space. "The environment" is nothing but an empty correlate of the system's self-reference; it provides no information. If, however, we are dealing with system/system relations, then the other side can be marked and indicated. In this case, art no longer deals with "everything else" but with questions such as whether and to what extent the artist is motivated by political convenience or by wealthy customers.
Insofar as system/environment relations are concerned, system differen- tiation merely replicates the difference between system and environment within the system, that is, it reenters the two-sided form system/environ-
5
ment into the system. The decisive question is whether, and in what ways,
other autopoietic systems, endowed with their own autonomy and their
own operative closure, can emerge within the autopoietic system of soci-
ety (which is closed with regard to its own operation of communication).
The answer lies in the problems that occur in the system at large, problems
that take over the subsystems as their own functions, because these func-
tions can be fulfilled nowhere else. Older social formations provide exam-
ples of such operative closure--urban communities based on center/
periphery differentiation, and aristocratic societies based on stratification.
But if at times centers of privileged life partially differentiate themselves,
this does not mean that autopoietic, operationally closed subsystems are
established within society at large--except within the dominant frame-
work of segmentary differentiation. Only given functional differentiation
do the subsystems generated according to this principle become opera-
tionally autonomous, because none of these systems can fulfill the func-
6
tion of the other.
In system/system relations, the concept of form becomes relevant in a
different way. Only in such relations can one speak of a "form of differen- tiation" in the sense that a system's type of differentiation informs the sys- tem of the other systems it must expect in its environment: systems of the same type in the case of segmentation, systems of a different type in the
136 The Function of Art and the Differentiation of the Art System
case of a center/periphery differentiation, and both similar and different systems in the case of functional differentiation. The differentiation of a functional system already indicates that there must be other such systems in its environment--whatever else the environment might contain--for the simple reason that all the necessary functions of the system as a whole must be taken care of in one way or another.
These reflections suggest a certain developmental logic in the evolution of forms of differentiation. This is not a matter of decomposing a given whole into its parts. Forms of differentiation are not principles of decom- position. If they were, then the transition from one form to another would be difficult to imagine. Rather, the system of society as a whole involves the possibility for differentiating operatively closed subsystems. When, and only when, this happens, the subsystem assumes a form that presupposes another side. The specific system type suggests what kinds of other systems can be expected on the other, external side of the form: other setdements if the form is a settlement; systems of lower rank if differentiation rests on a claim to higher rank; and eventually other functional systems if the differ- entiated system specializes itself along functional lines. In this way, religion was crucial for the development of the early modern state, first as ammu- nition for civil wars and later--after reorganizing itself during the triden- tium and within the corresponding structures of a state church that evolved in the Protestant world--as a partner in another, political function.
The relations between art and stratificatory differentiation are certainly more complex than one might expect in retrospect. When one distin- guishes individual genres and then asks how they can be connected, the
7
problem presents itself as one of hierarchical ranking --that is, in the
terms in which the unity of society or of the world is described. Hierarchy makes people look upward--even if it increasingly conflicts with the self- image of art. On the one hand, the nobility certainly did commission art- works. Art finds appropriate objects, persons, and destinies only in the
8
highest social circles. There is a connection here to the moral-pedagogi-
cal function of art. At the bottom of the social hierarchy, there is not enough room for free action, hence no example for excellence. Stylistic forms of rhetoric and poetry vary accordingly, depending on the rank of
9
the persons depicted. According to Henri Testelin, even drawing must
take social status into account and draw rustic country folks with rough
10
strokes while using clear lines to represent grave and serious people. Ludwig Tieck's novels, princes and dukes are still indispensable, but
In
The Function of Art and the Differentiation of the Art System 137
poverty contributes equally important possibilities for action. On the
other hand, the indispensability of social rank for the novel does not nec-
essarily mean that the upper classes developed an understanding or inter-
est in art.
29
as a slice of time with a self-determined beginning and end.
tional art, by contrast, begins by projecting an imaginary space or time in order to gain a free hand in employing this self-created medium for pur- poses of ornamentation and representation. Since the early Renaissance, European art has preferred this option, relegating ornament to the func- tion of decoration, of placing accents and emphasizing the essential. A closer look, however, reveals that the ornamental is always the work's in- frastructure, even in deliberately representational art forms, for the fol- lowing reason: if one employs space and time (how else could an artwork appear? ), then these media must be organized, whatever might subse- quently be represented in them.
IV
No matter how we place the accents, no matter how much a work's fig- urative or ornamental aspects may capture our attention at first, we must assume that the forms that constitute the work by virtue of their distin- guishing force diverge, depending on which medium of perception or in-
30
tuition they deploy. There is no commensurability between painting and
music, or between sculpture and dance, or between the lyric and the novel--which is not to say that there can be no "onomatopoeia" in music or that dancers cannot become sculptures. These formal differences are conditioned, not by the choice of form (if they were, then they could be avoided), but by the medium that underlies each form and that, by virtue of its loose coupling, makes tight couplings possible. Perceptual media do not display a spectrum as broad as that of artistic genres. Painting and
Representa-
n6 Medium andForm
sculpture, theater and dance, depend on light--a visual medium--whereas the lyric, like narrative (the epic, the novel), relies on language as a medium of fixation for intuition. But the manner in which art shapes and makes use of perceptual media differs in each genre. To radicalize our previous question: How can one speak of the unity of art, or of the unity of an artis- tic medium in view of such diversity?
In anticipating this question, we have taken great care to elaborate the distinction between medium and form. We can start from there. Percep- tual media are media of the psyche in Heider's sense, rather than of a so- cial or communicative nature. No one can share the perception of others, although one can perceive that others perceive or that others perceive one's perceiving. We shall make the controversial claim that the same is true in the realm of intuition, that is, in the realm of imagined perception. It is true when language is used for the purpose of stimulating intuition
(rather than communicating information) and thus holds also for art- works created in language, especially the novel. No one knows what an- other experiences intuitively when reading about Odysseus and how he had himself tied to the mast, or about Siebenkas, who searches for his own
31
grave only to find the woman he will eventually marry.
how others imagine Robinson's surprise when Friday appears, or how Napoleon (in War and Peace) is struck by the events of the Russian cam- paign. One tends to speak of "fictional" literature in these cases. Whatever that means, the fictional medium is primarily constituted by the private nature of an intuition that requires no "continuation of communication" and therefore no special effort on the part of consciousness and memory, but instead allows them to operate freely.
How, then, is art nonetheless possible as communication? And what, in this case, would be the medium of communication?
The key to this question may well reside in the intentionally created observational relationships we analyzed in the previous chapter. Once
? someone (no matter who) recognizes, from the manner of presentation, \ an arrangement that is produced for an observer, a social medium has \ come into existence--whether or not this aspect is communicated in the
work of art. Literary texts in particular often distinguish themselves by self-referential clues of this sort. (Incorporating the text's production into the text, addressing the reader, attacking the reviewers in the manner of Jean Paul are still rather crude stylistic means, aimed at differentiating the text at the level of an observation of observations. ) In the wake of these
No one knows
Medium and Form
117
developments, it becomes possible to establish an "artificial" form that si- multaneously serves as a medium for forms within the form--such as the space contained in a painting, the potential movement of a sculpture ar- rested in time, or the realm of possible events in which a narrative estab- lishes sequences that present in tightly coupled forms--in this and no other way--what might well turn out otherwise. Or consider the decep- tive maneuvers of baroque architecture, which are enjoyable only when seen through, or dance, which does not take the direction of its move- ment from ordinary life but instead presents it so that it appears, from moment to moment, as if it had been selected only for the sake of the dance.
Although perceptual media and artistic genres differ greatly with regard to their concrete materialization, they share a common ground in the manner in which they construct novel medium/form relations that are in- tended to be observed and are intelligible only when this is understood. The unity of art resides in that it creates for the sake of observation and observes for the sake of being observed, and the medium of art consists in the freedom to create medium/form relations.
The possibility for combining forms and thereby tightening the work, as it were, from within, suggests another parallel between individual gen- res. We recall that forms are always two-sided. No matter what they indi- cate and fixate in the work of art, they always simultaneously present an- other side that must cooperate in rendering visible what is determined by the work. This is true if the artwork itself is meant to be recognized as a specific object (and nothing else). It holds for every detail that constitutes the work in collaboration with others.
The unmarked space is the indispensable other side, a reference to pos- sibilities that, for their part, point to an infinity that cannot be contained
32
in one place.
unmarked space into a marked space, and it creates a boundary by cross- ing that boundary. Spencer Brown speaks of "drawing a distinction. " At die same time, a difference between medium and form comes into being, an enclosed, specially prepared marked space, in which the artwork fol- lows the pull of its own distinctions and determines its own forms.
The determination of one side does not entirely leave open what can happen on the other side--this accounts for the specificity of individual art forms. While it does not determine the other side, it renders it nonar- bitrary. What can happen there must "fit" or it will cause dissonance, a
The first step in the making of an artwork leads from the
n8 Medium and Form
flaw, or disruption (which can, of course, be intentional, in which case it requires a balancing fit on its part). The determination of one side, as we just stated, does not determine the other. But it facilitates decisions and the observation of decisions concerning what can happen or has been estab- lished on the other side by the artist. If an indication is to become a work of art, then the other side must remain accessible--which assumes that one can demarcate another unmarked space in relation to this other side.
Whenever an object is intended to be a work of art, the indication does not refer merely to itself (to this and no other object). It also refers to the crossing of the boundary that divides the form into two halves and in- structs the observer to search for and fix what has not yet been decided. This holds for the artist as well as for the observer of art, and ultimately for any observation that depends on time. The indication, we might say, is used as meaning. Crossing the boundary never leads into the unmarked space, never into the world as such, but always executes an indication, a new indication. As we know, indications can only be executed as distinc- tions. They specify only one side of (another) distinction that, for its part,
33
has its own other side. This is why an artwork cannot reject the world.
do so, it would first have to indicate and distinguish the world, that is to say, it would have to execute operations that can occur only in the world.
In this sense, the artwork forces both the artist and the observer to ad-
vance from one form to the next in order to return eventually to the form
from which they began, which is now the other side of another form.
Forms play with forms, but the play remains formal. It never arrives at
"matter," it never serves as a sign for something else. H Each formal deter-
mination functions simultaneously as an irritation that leaves room for
subsequent decisions, and advancing from one form to the next is an ex-
periment that either succeeds or fails. ? This is why, as we shall elaborate be-
low, a "code" emerges in art, a continuously maintained binary orientation
concerning the "fit" or "lack of fit" of forms. This is why every artwork
contains "information" in Gregory Bateson's sense--differences that make 35
a difference. And all of this holds for any kind of art!
Accordingly, for all genres the medium of art is the sum total of possible
ways of crossing form boundaries (distinctions) from within toward the outside and of discovering fitting indications on the other side that stimu- late further crossings by virtue of their own boundaries. The medium of art is present in every artwork, yet it is invisible, since it operates only on the other side--the one not indicated--as a kind of attractor for further
To
Medium and Form 119
observations. The process of discovery transforms the medium into form. Or else one fails. In working together, form and medium generate what characterizes successful artworks, namely, improbable evidence.
Does this always have to be another distinction with another side? One may well imagine an artwork that has two components but only one form (or, more accurately, a precise congruence of two inverse forms that over- lap one another). In this case, one focuses on the side that is the other side of the other side and vice versa. The formal asymmetry necessary for ob- servation is canceled in symmetry. One can only oscillate between die two sides. Any further dynamic is blocked. We have, in other words, the pre- cise image of a logical paradox, a kind of time trap that compels the ob- server to move back and forth between its poles in the shortest possible in- stance. This is not meaningless, nor is a logical paradox meaningless. But the meaning of such a figure--of its form--must be sought in the clue that allows for unfolding this paradox and reintroduces asymmetry into the form.
Paradoxes present the world in the form of a self-blocking observation. It is certainly possible to stage works of art as paradoxes, but only in order to demonstrate what doesn't work, only to symbolize the unobservability of the world. Works consisting of only two components are not yet gen- uine artworks. But to the extent that they carry along the other side of the
36
form as an empty form, as mere exteriority, they are works of art after all. This is particularly true where paradox becomes the work's theme--as in Escher's etchings, in Magritte's work, or in certain forms of sixteenth- and
37
seventeenth-century lyric poetry, especially that of John Donne.
cases, the paradoxical oscillation of truth is introduced deliberately--not to represent the world, but to invite the viewer or reader to search for an innovative exit that remains undetermined in the work and about which
38
even the artist himself may have his doubts.
may work together to accomplish this effect, paradox remains the super- form, the frame suggestive of what remains unsaid in the work and what is only marked as unsaid.
Let us return to the ordinary. What remains open as the other side of a form is generally determined by an indication that actualizes another dis- tinction for which the same is true. This process keeps going until a system of references closes itself off, in which nothing remains undetermined. Oc- casionally, things may go wrong in the process, and the remaining discord must be covered up or minimized. This is a fact of life. Our concern here
While several aesthetic forms
In these
IZO Medium and Form
is not art criticism but the principle of form, the process of observing (the making and observing) of artworks.
It should be clear by now that this analysis precludes comprehending an
artwork in terms of the relation between a whole and its parts. Dividing a
work and judging die relationship between parts misses its internal nexus.
Nor are we concerned with the primacy of the whole in relation to its
39
parts.
bution to the work consists in what they are not, what they make available for further elaboration. The artwork closes itself off by reusing what is al- ready determined in the work as the odier side of further distinctions. The result is a unique, circular accumulation of meaning, which often escapes one's first view (or is grasped only "intuitively"). Eventually, it may turn out that a determination plays a role in several distinctions at once, that it is multifunctional and therefore not interchangeable. This creates an over- all impression of necessity--the work is what it is, even though it is made, individual, and contingent, rather than necessary in an ontological sense. The work of art, one might say, manages to overcome its own contingency.
Several more or less standardized, genre-specific formal models are available for this process. The basic form for generating forms from odier
40
forms is the (misleadingly so called) ornament.
on the problem of broken symmetry, on the problem of form. At stake here is the projection of asymmetries that still exhibit traces of the sym- metries from which diey emerged. Ornaments are recursions tliat keep go- ing by recalling previous and anticipating further forms. They display the
41
unity of redundancy and variety. Transitions are effaced; at least they are
not emphasized as breaks, since each place in an ornament is at the same time the place of another place. This includes the repetition of forms at other points in space or time, whereby their positional difference suggests nonidentity within the ornament's overall identity. Continuation is the principle that integrates what is excluded at first, that defines the excluded as a motif, reiterates it, or connects other motifs. This clearly shows that art is neither a sign for something else nor the mere form of the material. The ornament generates its own imaginary space by continuously trans- forming formal boundaries into transitions that have more than one meaning. It prevents the work from falling apart into isolated figures, on which one can focus or from which one can turn away. The ornament, in other words, holds the artwork together, precisely because it does not par- take in its figurative division. One can call this "mere decoration" only in
If one wants to isolate parts, then one discovers that their contri-
All ornaments are based
Medium and Form 121
social situations in which the differentiation of art is already under way, whereas decorative ornaments also adorn utilitarian objects, jewelry, sa- cred objects, or "crafts," so that mere decoration must be distinguished from art. In such a situation, the ornamental structure can be granted only
42
a subordinate role.
supplements his observation by referring to the masking function of dec-
43
oration:
tinct and enables faster comprehension; on the other hand, it suppresses contradictory, confusing information. The ornament accomplishes all this with a luxurious excess that tends to become an end in itself.
The basic idea is still one of "hierarchical opposition," of a contrast be- tween balanced proportion and mere adornment. One expects art to keep decoration under control. This places the burden on the dominant side of the distinction, on what is supposedly balanced proportion or is later called symbolic meaning. But the inability to resolve the problem of hier- archy eventually "deconstructs" the distinction itself. The ornament, ini- tially meant to fulfill a secondary function, takes on the burden of sup- plying meaning. If one wants to observe artworks as art with reference to their play of forms, then one must pay attention to their ornaments.
Only then can one return to the question of how the work is made and which secondary meanings serve the ornament while receiving from the ornament the electrical charge tliat accounts for their artistic quality. Painting, too, eventually pushes its ornaments to the margins or into the background--which needs to be filled anyway--in order to foreground its figures. Assisted by unified perspective, painting develops the back- ground into an open space--a landscape, for example--only to discover the need to compensate for the ornaments function by filling its imagi- nary space with nonarbitrary objects until, in the end, even the landscape
44
is no longer necessary. While the ornament is marginalized as mere dec-
oration that can also adorn nonartistic objects, a functional equivalent of the ornament emerges from within artworks themselves--an inner "line
45
of beauty"
more distinctly curved and of much greater condensation than any nat- ural line. Even as ornamental adornment is excluded from art, the orna- ment is reborn within the work. Similarly, in poetry sound and rhythm are more easily replaced by meanings, which requires that the play with distinctions be reproduced as a formal nexus among narrative elements. By indicating an action, narrative, for example, can serve two distinctions
Gombrich speaks of an "explanatory division" and
on the one hand, decoration renders the information more dis-
that joins what has been separated figuratively, a line that is
122 Medium and Form
The au-
The intense use of these possibilities leads to the modern novel of the
47
individual. "Flat" heroes become "round,"
parent, the strength of motives is demonstrated (typically those preferred by the author, such as the profit motive in Robinson Crusoe or Moll Flan- ders). At the same time, narrative can induce changes in a character; it can motivate learning, religious conversion, or regret; and it can recommend its result to the reader as a laudable attitude. This development allows the novel to free itself from moral guidelines, and the reader is confronted
48
at once: it characterizes the agent and moves the plot forward.
thor arranges the events, that is, he shapes the medium in such a way that the reader can infer the changing mental state of the hero. The fictionality of the arrangement hides behind the contingency ofevents and actions that must serve as the starting point for the reader who follows the narrative.
with live models and experiences that could be his own.
mal combination of character and plot by means of actions--a combina- tion that informs both of these distinctions--has established itself and guides the reader's expectations, it becomes possible to write narratives that still present themselves as novels but break with this combination. These narratives seek to distinguish themselves by precluding any infer- ences concerning the character and motives of the hero or by refusing to move the plot along by means of actions. This development begins as early as Flaubert's L'Mucation sentimentale (1869).
The internal ornament serves the artworks self-description; it beautifies
49
because it is beautiful.
it can bind. On the one hand, the form combination that is selected indi- vidualizes the artwork and qualifies it as a distinct object. This allows for technical reproduction, which neither affects the work's recognizability nor "damages" its individual form, but instead makes it more accessible. Technical reproduction, in turn, generates a new form: the distinction be- tween original and copy. On the other hand, observing how the work is
50 made yields an observation of a more general type often called "style. " At
the level of stylistic forms, the art system is able to evolve; it can replace form combinations that have already been tested or derive new forms from the rejection of what has become all too familiar. Moreover, the sys^ tem can turn rejection itself into a form that is intelligible only to those aware of what used to be customary and what, accordingly, is the expec-
51
tation to be disappointed.
itself the avant-garde has taken this backward-looking manner to an ex-
their motives become trans-
It absorbs as much variety as possible, as much as
The movement that, strangely enough, calls
46
Once the for-
Medium and Form 123
treme--like oarsmen, who face the direction they are coming from and have the goal of their journey behind their backs.
We will not elaborate on these reflections. Instead, we would note that situations of this sort--in any artistic genre--presuppose the artwork as a specific combination of forms, no matter how this presupposition is then attacked in the search for new forms. This protest is possible only when (and so long as! ) a loosely coupled medium is available for the creation of possible forms. If the previous reflections hold, then this medium may well reside in the need for another, yet undefined side on the other side of the indications that make up an artwork. This need would explain why an artwork, in closing itself off and in determining its open side by means of other distinctions, coagulates the medium in such a way that a tightly coupled form emerges, though the medium always reproduces itself at the same time--that is, the question of what lies on the other side of the form is posed anew in each instance.
V
We began with the medium of perception, mentioning the medium of
language only in passing. But in order to demonstrate the unity of art and
the formal similarity of all artistic genres, we must pay attention to the
verbal arts as well, especially to poetry. Here, we are dealing with words as
medium, as a loosely coupled set of elements. Words serve not only as el-
ements to be coupled but also as a means for the artwork's self-descrip-
tion, a means of bringing about the unity of the description and the de-
52
scribed.
formal combination of words that calls upon words to display an unusual meaning--whether or not it relies on the structure of grammatically cor- rect sentences. This can happen only if distinctions are placed in words. There is no need to paraphrase these distinctions; nor can they be para- phrased, for to do so would create the possibility for rejection--just what art wants to avoid.
Ordinary language uses the same words in multiple contexts and there- fore relies on wearing down their meanings and on phrases as an aid to understanding. It seeks to establish unambiguous denotations and ac- complishes this goal via naming and the construction of abstract objects, conceptual correlates, or ideas. Poetic language operates in reverse--with or without the aid of phrases. It reflects on the usage of language--as if
Accordingly, the creation of poetic form consists in a special
124 Medium and Form
language were one material among others that one finds in the world.
54
53
Poetry has no use for denotations. Instead, it relies on connotations, ploying words as a medium in which the connotations that select one an- other assume form. Rather than forcing the diffuse referential richness of words into the most unambiguous relation to facts in the world, poetry posits this richness (including the resonance of what remains unsaid) over against the facts. Just as atoms alter their internal electronics when com- bining into molecules, poetry modifies the meaning of words. It may gen- erate striking new nuances or create estrangement, but it can also surprise by restoring the original meaning of everyday words. Overgeneralized forms are pried open and reconstructed. Poetry contextualizes such forms so that they become structurally determined and can no longer easily be recognized. Like any use of form, this strategy aims at what it excludes. Other words can pick up what remains unsaid, but they can also serve as a reminder that many important things must be left unsaid. The other side of the form always plays along--as a boundary that can either be crossed by further guided observations or else is fixed, again and again, as the same boundary, as the unmarked space of different words.
Poetry shifts from a denotative to a connotative use of words because of its need for poetic closure--a closure for which the poetic genre does not choose the form of narrative. Referential meaning would refer the reader into the world, where it would lose itself among the multiple references
55
that make up the meaning of reality. The poem becomes a unity only at
the level of connotation, by exploring the liberties that come with using
56
words exclusively as a medium.
can be secured at this level, forms one must relinquish when moving from one poem to the next. Poetry requires an intensification of memory, that
57
This means that only temporary forms
is, a retroactive reading (if one still wants to call this "reading"
and reader must leave behind the linearity of the text and apprehend its structure in a circular manner; they must be able to take apart the net- work of multiple circles that makes up a text. But such an effort is feasible and can be accomplished only within a single poem.
Connotations start out from the familiar meanings of words, truncating only their hetero-reference so that opposites, for example, can appear as a unity, even though and even because this is not how they appear in the ex- ternal world. This technique requires a suspension of the ordinary referen- tial meaning of words, which catches the reader off guard. Last, but not least, this is possible because of the ornamental quality of verbal constella-
). Author
em-
Medium and Form 125
tions. This quality may reside in the sound of words (nevermore 2nd vast [English in the original] are well-known examples), or it may be an effect of the relation between short and long syllables or of repetitions, echoes, stereotypes, contrasts, anagrams. In Finnegans Wake, the ornamental qual- ity of language, the resonance between sounds and other words, over- whelms the text to an extent that intelligible words communicate only that they do not matter. Rhythms are complicated to the point where they es- cape reading altogether and require recitation to be apprehended. To con- vince, poetry appeals to perception, not to thinking. And the function of the ornament, in poetry as elsewhere, is to intensify redundancy and vari- ety in ways that would otherwise hardly be possible.
Poetry, then, is not just rhymed prose. If one reads poetry as a sequence of propositions about the world and considers the poetic only as beautifi- cation, adornment, or decoration, one does not observe it as a work of art. Nor can one apprehend in this way the formal combination the poet uses
58
to compose his work. Only at the level where symbols,
ings, and rhythms conspire--a level that is difficult to "read"--do poems refer to themselves in the process of creating forms. They generate contex- tual dependencies, ironic references, and paradoxes, all of which refer back to the text that produces these effects. Supported by the text, poetic self-
59
reference may eventually articulate itself explicitly
statement, but as a form within the nexus of forms that constitute the text.
The problems presented by the materiality of words have often been discussed, at least since Mallarme and frequendy with reference to him. It seemed plausible to locate the problem in the relationship between con- sciousness and language, in the poet's access to language, or in the self- sacrifice commanded by the shaping of language. This general insight, which ultimately points to the distinction between psychic and social sys- tems, can be supplemented by the distinction between medium and form. This distinction is a projection of art, here a projection of poetry, a form of its autopoiesis. It is given neither as matter nor as Spirit. It has no on- tological substrate, which makes it tempting to observe how observers handle this distinction.
All of this needs further elaboration. William Empson and Cleanth Brooks brought these problems to the attention of a literary theory that
60
calls itself "critical. "
form, this awareness yielded only the formal analyses of the New Literary Criticism and subsequently led to the critique of what these analyses ig-
But instead of producing a general terminology of
sounds, mean-
--not as a flat, abrupt
Medium and Form
nored.
literature, the promise of a unified theory of artistic kinds went unful- filled. Currently, however, literary theory is sufficiently receptive to inter- disciplinary suggestions that this separation is unlikely to prevail. Thus af- ter this excursion into the special domain of the medium of language, we return to more general analyses.
VI
The medium of art renders the creation of forms at once possible and improbable. The medium always contains other possibilities and makes everything determined appear to be contingent. This improbability is em- phasized when everyday purposes and utilities are bracketed as the guid- ing principles of observation. Artistic form (backed by aesthetic reflection) goes out of its way not to appear useful.
In this way, the artwork directs the beholder's awareness toward the im- probability of its emergence. If attention is drawn to poetic constructions,
then it is only because they do not seem very likely, whereas the likelihood of using other constructions is, on the contrary, very high. "Poetic is that
62
which has not become law," writes Julia Kristeva.
texts, one might add that their improbability must not be based on their informational value, which always implies a certain quality of surprise, but consists in their renunciation of information in the sense of mundane utility.
During the past two centuries, a number of doctrines derived from diis enforced improbability surfaced in poetic reflection--the rejection of a rule-based poetics, for example, or the emphasis on the individuality and originality of an authentic artwork, and eventually the search for an alter- native explanation of the improbable that points to the "genius" of the artists. But these are collateral circumstances, secondary phenomena that accompany efforts to come to terms with the improbable. When focusing on the improbability of form itself, one is primarily concerned with the observer's fascination, his staying-put-with-the-work in a sequence of ob- servations that attempt to decipher it.
One might expect the sequence of medium-form-medium-form for- mations progressively to constrain the medium's possibilities, thus leading to an increase in redundancy. Many potential sculptures come to mind if one considers only the media of space and material. When one represents
126
61
Caught in the conventional division between die visual arts and
Especially for poetic
Medium and Form 127
a mobile living being, the limitations of its body constrain what can be rendered. Lessing's analysis of the Laocobn shows that the artist is not en- tirely free in selecting the moment from which the before and after of movements can be rendered visible. Once we are dealing with the Dying Gaulor with stage productions of Lucia di Lammermoor, very few perfor- mances are conceivable that could give form to precisely this medium. The improbability of the composition must be wrested from these limita- tions. This may result in a "structural drift" within the art system that turns into an end in itself not only the artwork but above all its improba- bility. Eventually, one begins to experiment with the idea of declaring everything a work of art so long as an artistic claim can be asserted and maintained. The work's probability then boils down to the credibility of such a claim.
But artistic credibility is still a relationship between medium and form. The difficulty of creating forms shifts to the difficulty of claiming a work as art and maintaining this claim. Yet the medium remains a medium of art by virtue of its ties to the history of art; it continues to function as a medium that propels the historical machinery of the art system beyond its current state with new and ever more daring forms. The medium might absorb decontextualized historical references, as it does in postmodernism, whereby the improbability resides precisely in this decontextualization, in free selection from a historical reservoir of forms. What used to be bound historically is now up for grabs on the condition that it remain recogniz- able as such. One might equally well continue the project of the avant- garde in an effort to expand, via the production of art, the concept of art itself. In both cases, art turns into the artistic medium insofar as, and so long as, it is capable of making the observer recognize the improbable as improbable. In the end, the observer might even be challenged to com- prehend the incomprehensibility--created especially for him--of an art- work as a reference to an incomprehensible world.
The recognition that every form is a form-in-a-medium dates back to romanticism. A medium suited for artistic forms had to be sought, dis- covered, and eventually constructed by dismanding interpretive aids taken from everyday life. The fairytale-like incredibility of backdrops served this purpose and simultaneously indicated that henceforth only absolute self- reflection--a reflection that includes the observer--could be presupposed
63
as the ultimate medium. Fantastic art,
the events and forms presented can be explained naturally, is subject to
by leaving open whether or not
128 Medium and Form
similar conditions. But the general situation was still conceived in terms of the subject, an authority that underlies itself and everything else. In the meantime, the dynamic of the art system has evolved in ways that can no longer be attributed to a subject. The observer--the self-reference under- lying all distinctions and the distinctions underlying all self-reference-- seems a more appropriate basic figure. From the position of the observer, one can unfold this circle by distinguishing the distinction from the indi- cation of one of its sides, and self-reference from hetero-reference. This distinction specifies the operation of observing as a distinguishing indica- tion and defines the concept of the self-referential system as a system that copies the operatively generated distinction between system and environ- ment into itself and bases its observing operations on the distinction be- tween self-reference and hetero-reference.
Once the "subject" gives way to the observer, we no longer need the counterconcept of an object. Regarding the manner of operation, a wide range of possibilities is now conceivable besides intentional awareness (a state of consciousness). The observer can be a social system, and observa- tion can be communication. The artwork is not necessarily a device that causes the perspectives of producer and observer (or of theories of produc- tion and reception) to oscillate. Still, nothing speaks against starting out from references to psychic systems, either to the artist or the observer. But the emergent unity of the art system and its unique medium cannot be grasped in this manner. The art system is a special system of social com- munication. It has its own self-reference and hetero-references tliat indi- cate forms existing exclusively in a medium unique to art. This medium is the improbability of the combinatory structure of form that art wrests from everyday life and that refers the observer to other observers.
These reflections eventually raise the question of whether an artwork
64
has to be difficult and, if so, why. Like everything else, this proposition
can be questioned today, and there is a tendency to separate art from craftsmanship. Pushed to its extreme, difficulty might ultimately boil down to the problem of how one can work as an artist in a manner that is still recognizable. Pointing to the essence of art--to the idea of art, the rarity of genius, or the like--is of no help in this matter. The question, rather, is whether and for what reasons the mediums potential for creat- ing forms must be limited, and how this limitation is accomplished.
Within a theory of symbolically generalized media, Talcott Parsons as- sumed that each of these media, just like money, requires a real backing
Medium and Form 129
that can be overdrawn by confidence but not expanded at will. Using the
medium below or in excess of its capacities is certainly possible, but it
leads to an inflation or deflation of the medium that jeopardizes its func-
65
tioning.
ing of art, especially of modern art? Obviously, nothing external to the medium can fulfill this function; what backs the medium of art is the works triumph over its own improbability.
This is why the trend toward facilitating the creation of forms and re- ducing forms to simple distinctions cannot be countered by judgments of taste or values. Even tlie concept of art apparendy no longer sets limits to what can count as art. But one can know that the medium/form dynamic requires constraints and that expansive trends lead to inflation. How much inflation the art system can tolerate boils down to an empirical question. Sanctions are evident not in the reaction against violations of the norm but in the loss of interest in the observation of observations.
VII
As we suggested earlier, the distinction between medium and form is based on a complex relationship to time. On the one hand, medium and form must be actualized simultaneously. On the other hand, the medium reproduces itself only by alternating the forms that an observer employs as distinctions. The stability of the medium rests on the instability of forms that repeatedly realize and dissolve tightly coupled relationships. Media are constant, forms alternate. In the basic medium of meaning, all other media are subject to variation as well, but only to the extent that they are observed as forms in another medium.
The paradoxical "simultaneity" of invariance and variability corre- sponds to the general problem of structuring the autopoietic reproduction of systems. Only actualized elements that assume the form of events (op- erations) can reproduce the system, which requires a recursive recapitula- tion of the past and anticipation of future events; in other words, nonac- tualized events must be actualized as nonactual. The actualization of the nonactual requires (and is made possible by) a selectivity that employs the logic of the distinguishing indication. Selections capable of actualizing the nonactual always function as structures--in the moment of their actual-
66
ization--by virtue of references that transcend the actual.
In art, the work's material substratum guarantees that the observing op-
If we follow Parsons's suggestion, then what would be the back-
130 Medium and Form
erations can be repeated. It ensures that the potential for repetition is per- ceived along with the work and that it actualizes what is momentarily nonactual. At the same time, the work indicates the nonidentity of the repetition. We are aware that we experience the same work (without ques- tioning its sameness) differently each time--for example, as recognizable or familiar, as confirming our attitudes rather than presenting us with as- tonishing information. Redundancy and variation collaborate in their ef- fects. Repetition alters what is repeated--especially when the repeated content is recognized and affirmed as the same. Identity is necessary--but only to allow for the nonidentical reproduction of the observing opera- tion. Observational sequences can build up comfortable redundancies and suppress provoking irritations; they can search for confirmation in one as- pect of the work and find it in another. In the visual arts, the stability of the material secures this process. In texts, writing--and in music, the re- peatability of the production (with or without notation)--does the same. We need not pursue the details of this art-external (material, memorylike) anchorage here; what should be stressed is that it requires a separation of individual artworks. The horizon of reference must be interrupted in or- der to allow for recursion, for the return to the same, and for the struc- turing anticipation of this return. But if this is true, doesn't the art system disintegrate into a disjointed ensemble of individual works?
This question enforces the recurrence of the temporal problem at the level of the systems autopoiesis, a level that transcends the individual art- work. At this level, the temporal paradox of structuring, the paradoxical actuality of the nonactual, recurs at a higher level. It is no surprise that un- folding the paradox once again boils down to a distinction--not between externally secured constants and the fluidity of observation, but between change and conservation in what counts as art.
To observe the changes in a domain common to many artworks, the (historical) concept of style has been available since the last three decades
67
Long before that, the concept of style had been 68
of the eighteenth century.
used to designate ways in which the elements of an artwork are coupled. Rhetoric, following a general trend of hierarchization, proposed a ranking
69
of styles and prescribed styles according to the dignity of their objects. Not until Winckelmann was the concept of style--which concerns factual differences such as "writing," manner, and presentation--anchored in a temporal dimension and claimed to reveal (and cause) historical differ- ences. The distinctions employed by artworks, the "against what? " of their
Medium and Form 131
manner, were subjected to the pressure of innovation. Not only did indi- vidual works have to distinguish themselves, but what did not distinguish them needed to be distinguishable at another level of comparison, as be- ing sanctioned by the concept of style. One expected style to legislate it- self--not to succumb to a prescribed canon, but rather to distinguish itself by deviating from models. A style prolongs the half-life of the publics in- terest in a work of art; one is reminded of similarities in other works and can observe each work anew with reference to similarities and differences. Style respects tradition by deviating from it. Deviation is a specific form of acknowledging relevance; it is not indifference or ignorance. Deviation re- quires knowledge of the subject matter, circumspection, and precision in selecting aspects where deviation matters, and often it is necessary to re- formulate the unity of the preceding style with disregard for what was rel- evant and accessible to this style. The procedure is a typical case of recur- sive reconstruction!
At the same time, there is the opposite trend: to hold on to what is worth preserving, precisely because of deviation. Objects are put into mu- seums or, when this is not possible--as with textual art--they are identi-
70
fied as "timeless" classics. Museums grow out of processes that decide
what is and what is not accepted. Today, even the most recent art can be defined as given (= already dated) by the mere fact of being accepted and displayed in a museum. The decision observes observers, that is, it belongs to the level of second-order observation. Classicism, too, is a construct,
71
created by observers for other observers,
has always been to reverse time: in contrast to other works, classical works
72
improve with time.
from stylistic change, which is of no significance whatever unless there is something against which the preservation of worthy objects is directed, namely, the perpetual historicization of styles. The notion, implicit in the idea of style, "that one is no longer able and will never again be able to work that way" reinforces the conservation of resources that are no longer reproduced, and each loss becomes an "irreplaceable" loss. One needs in-
73
stitutions of mourning, of the "nevermore" [English in the original]. This diagnosis shows that even at this level the paradoxical unity of the distinction between medium and form seeks identifications, plausible dis- tinctions that can be carried on and prove one another. Style as form, the museum as form, classicism as form: all of these forms respond to the same fundamental situation (concealed by forms themselves), namely, that loose
and the intent of this construct
Museums and the classics symbolize an art removed
132 Medium and Form
and tight couplings are reproduced simultaneously in a manner that is in- variant and invisible in the medium and variable and visible in the form. What reacts to this situation is not a supermeaning, a principle of art, an ultimate, convincing idea, but yet another distinction plausible enough to enable convincing identifications. The form of "style" processes the bur- den of innovation and along with it the temporality of all forms while cast- ing a secret glance toward an eternal life beyond its own time. The form of the museum and the form of classicism live off a work's ability to outlast changing styles and find therein its meaning.
Although art collections have been around for a long time and there have always been preferred authors and composers, the museum and clas- sicism as forms of conservation presuppose an art system that operates at the level of second-order observation. It is therefore no accident that these forms, along with a historicized concept of style, appear in the final dec- ades of the eighteenth century--at a time when the differentiation of the art system reaches the level of second-order observation, when it estab- lishes itself and begins to solve its problems at this level. Now one begins to inquire into the unity of the arts regardless of the different media of perception in which they realize their primary forms. Only now is art, no matter what kind, defined in temporal and historical terms. This period also introduces a reflexive concept of culture: it situates culture within the context of historical and regional ("national") comparisons for the pur- pose of self-evaluation. Once the game of observation is played at this level, it finds rules and opportunities for self-affirmation, which, for the time being, provide sufficient orientation. An "analytical" terminology ca- pable of more rigorous analyses is nowhere in sight. At any rate, such a
terminology would only reveal the paradox that informs any operation with distinctions.
? 4 The Function of Art and the Differentiation of the Art System
I
One of the few constants in the century-old academic history of sociol- ogy is the assumption that modern society is characterized by a certain de-
1
gree of social differentiation and by some unique form of differentiation. Historical development, it turns out, is subject not only to differentiation
2
but also to de-differentiation.
ory and the theory of evolution have altered the manner in which the the- orem of differentiation is justified and how it is conceptually framed. To- day, one no longer works with analogies based on the paradigm of the division of labor, which was believed to occur spontaneously whenever there was an opportunity, simply because of its yields or its productive ra- tionality. It is doubtful that more differentiation, or differentiation at the level of the division of labor, is generally desirable. (Adam Smith already pointed out its disadvantages. ) Currendy, an overall critical, more skepti- cal, and doubtful attitude prevails. This does not change the fact that the theorem of differentiation posits a crucial accent, if not the main criterion for distinguishing modern society from its predecessors. However, if dif- ferentiation in its specifically modern form turns out to be not as benefi- cial as was previously assumed, then one needs to revise one's judgment of modern society. Many indications point toward this.
A first step--which hardly improves our relation to the tradition--is to describe modern society as a functionally differentiated system. Generally speaking, this means that the orientation toward specific functions (or problems) of the social system catalyzes the formation of subsystems that
Moreover, the elaboration of systems the-
m
134 The Function of Art and the Differentiation of the Art System
dominate the face of society. If one wants to investigate the consequences of this development and wants to know how differentiation affects the subdomains of social communication (in this case art), then one must fo- cus one's conceptual apparatus more accurately. Most importantly, one must clarify how functions can serve as evolutionary "attractors," and in
3
precisely what sense subsystems constitute systems in their own right. A description of the art system that takes these background assump- tions for granted and analyzes the form of social differentiation in general systems-theoretical terms yields consequences that will accompany us from now on. Today, systems theory is a highly developed, albeit contro- versial, analytical instrument. It requires theoretical decisions that do not directly concern art. (This, of course, holds for other--for example, semi- ological--analyses of art as well. ) In conjunction with the thesis that soci- ety is a functionally differentiated system and is in this form historically unique, a systems-theoretical orientation has further consequences. It means that the different functional systems are treated in many respects as comparable. The terminology we introduced earlier demonstrates this in remarkable detail. Issues such as system formation and system boundaries, function, medium and forms, operative closure, autopoiesis, first- and second-order observation, and coding and programming can be investi- gated with regard to any functional system. As these investigations take shape and yield answers, a theory of society emerges that does not depend on discovering a unified meaning behind society--for example, by deriv- ing societies from the nature of man, from a founding contract, or from an ultimate moral consensus. Such propositions may be treated as part of the theory's subject matter, as different forms of self-description available to the system of society. What ultimately characterizes society, however,
4
manifests itself in the comparability of its subsystems.
In a domain such as art (just as for law, science, politics, and so on), we
discover not unique traits of art but features that can be found, mutatis mutandis, in other functional systems as well--for example, the shift to a mode of second-order observation. Art participates in society by differen- tiating itself as a system, which subjects art to a logic of operative clo- sure--just like any other functional system. We are not primarily con- cerned with problems of causality, of society's influence on art and of art on society. (Such issues are of secondary importance. ) Nor do we advocate the defensive attitude that die autonomy of art ought to be upheld and protected. Modern art is autonomous in an operative sense. No one else
The Function of Art and the Differentiation of the Art System 135
does what it does. This is why questions concerning die independence or dependence of art can arise in a causal sense. The societal nature of mod- ern art consists in its operative closure and autonomy, provided that soci- ety imposes this form on all functional systems, one of which is art.
We base the following analyses on a distinction, namely, on the distinc- tion between system/environment relations, on the one hand, and system/ system relations, on the other. When dealing with system/environment re- lations, the system constitutes the internal side of the form, whereas the environment is its unmarked space. "The environment" is nothing but an empty correlate of the system's self-reference; it provides no information. If, however, we are dealing with system/system relations, then the other side can be marked and indicated. In this case, art no longer deals with "everything else" but with questions such as whether and to what extent the artist is motivated by political convenience or by wealthy customers.
Insofar as system/environment relations are concerned, system differen- tiation merely replicates the difference between system and environment within the system, that is, it reenters the two-sided form system/environ-
5
ment into the system. The decisive question is whether, and in what ways,
other autopoietic systems, endowed with their own autonomy and their
own operative closure, can emerge within the autopoietic system of soci-
ety (which is closed with regard to its own operation of communication).
The answer lies in the problems that occur in the system at large, problems
that take over the subsystems as their own functions, because these func-
tions can be fulfilled nowhere else. Older social formations provide exam-
ples of such operative closure--urban communities based on center/
periphery differentiation, and aristocratic societies based on stratification.
But if at times centers of privileged life partially differentiate themselves,
this does not mean that autopoietic, operationally closed subsystems are
established within society at large--except within the dominant frame-
work of segmentary differentiation. Only given functional differentiation
do the subsystems generated according to this principle become opera-
tionally autonomous, because none of these systems can fulfill the func-
6
tion of the other.
In system/system relations, the concept of form becomes relevant in a
different way. Only in such relations can one speak of a "form of differen- tiation" in the sense that a system's type of differentiation informs the sys- tem of the other systems it must expect in its environment: systems of the same type in the case of segmentation, systems of a different type in the
136 The Function of Art and the Differentiation of the Art System
case of a center/periphery differentiation, and both similar and different systems in the case of functional differentiation. The differentiation of a functional system already indicates that there must be other such systems in its environment--whatever else the environment might contain--for the simple reason that all the necessary functions of the system as a whole must be taken care of in one way or another.
These reflections suggest a certain developmental logic in the evolution of forms of differentiation. This is not a matter of decomposing a given whole into its parts. Forms of differentiation are not principles of decom- position. If they were, then the transition from one form to another would be difficult to imagine. Rather, the system of society as a whole involves the possibility for differentiating operatively closed subsystems. When, and only when, this happens, the subsystem assumes a form that presupposes another side. The specific system type suggests what kinds of other systems can be expected on the other, external side of the form: other setdements if the form is a settlement; systems of lower rank if differentiation rests on a claim to higher rank; and eventually other functional systems if the differ- entiated system specializes itself along functional lines. In this way, religion was crucial for the development of the early modern state, first as ammu- nition for civil wars and later--after reorganizing itself during the triden- tium and within the corresponding structures of a state church that evolved in the Protestant world--as a partner in another, political function.
The relations between art and stratificatory differentiation are certainly more complex than one might expect in retrospect. When one distin- guishes individual genres and then asks how they can be connected, the
7
problem presents itself as one of hierarchical ranking --that is, in the
terms in which the unity of society or of the world is described. Hierarchy makes people look upward--even if it increasingly conflicts with the self- image of art. On the one hand, the nobility certainly did commission art- works. Art finds appropriate objects, persons, and destinies only in the
8
highest social circles. There is a connection here to the moral-pedagogi-
cal function of art. At the bottom of the social hierarchy, there is not enough room for free action, hence no example for excellence. Stylistic forms of rhetoric and poetry vary accordingly, depending on the rank of
9
the persons depicted. According to Henri Testelin, even drawing must
take social status into account and draw rustic country folks with rough
10
strokes while using clear lines to represent grave and serious people. Ludwig Tieck's novels, princes and dukes are still indispensable, but
In
The Function of Art and the Differentiation of the Art System 137
poverty contributes equally important possibilities for action. On the
other hand, the indispensability of social rank for the novel does not nec-
essarily mean that the upper classes developed an understanding or inter-
est in art.
