The declarative sentence, as a “context-free” model of reality, offers us the possibility of an objective understanding of the universe. It is the foundation of scientific dis­course, which makes explicit and rigorous the decon­textualizing elements of the declarative model by calling for empirical verification/falsification, eventually in the controlled environment of the laboratory. But the declarative is by the same token the origin of fiction, which exploits its liberation from the discursive context in the opposite fashion, as a source not of objectivity but of the free representation and transcendence of desire.

The esthetic relation is constituted by our oscillation between contemplating the sign and its sacred referent that originates on the scene of representation. But this space and attitude of contemplation (Sartre’s pour-soi) once constituted can then be evoked by real-world objects, which are no longer perceived in the “instinctive” framework of appetite but can in various ways acquire an aura of sacrality. Although there is no point in speaking of mimesis itself as an esthetic phenom­enon, the forms of what René Girard describes as mimetic desire (see Mensonge romantique et vérité Romanesque, Grasset, 1961) include an esthetic dimension that depends on the scene of language, if not on language itself.

Imitation of another’s behavior can of course be performed for strictly practical reasons. But in this form of desire, the subject imitates a model whom Girard calls the mediator, whom he regards consciously or unconsciously as an esthetic object. The perceived reality of the model alternates with an imaginary being that embodies an inaccessible, sacred essence. The subject may subsequently attempt to acquire the objective components of this being for himself, in which case his mimetic activity rejoins the domain of the practical. But in the esthetic moment of mimetic desire, the non-equivalence between being and act, or substance and appearance, presents itself to the imagination of the subject.

The esthetic imagination as so defined is not practical, anticipating possi­ble future activity, but paradoxical. On the one hand, posses­sion of the appearance appears to include within itself the possession of the being that is expressed by it; but on the other, because the appearance is an expression of this being, the being must be acquired before the appearance and is therefore inaccessible through it.

This esthetic element is not dependent on language, nor indeed on any formal system of representation, except insofar as such systems lend significance to the appear­ances to be imitated. Yet it is important to insist that the moment in which the “being” of the other is contemplated, as opposed to “instinctively” imitated as in animal mimesis, is a mode of deferred action dependent on the human scene of representation. Thus it is no accident that the desiring subject’s contemplative “possession” of the real or remembered image of his model is homologous to the speaker’s “possession” of the sacred object in the originary event through the intermediary of the sign, the object of the ontological faith expressed in the inappropriate ostensive.

For the subject of mimetic desire, the image of his model in the act of appearing is a kind of private representation. This appearance is like an ostensive sign that, because its utterance accompanies in principle the presence of its referent, comes to be employed inappropriately, with the same paradoxical conse­quences, in order to obtain the presence of this referent in its absence. Hence although the desire is purely mental, it is nonetheless an effect of representation, a cultural phenomenon.

Yet the means of expression furnished by the ostensive, although adequate to this task, can scarcely be said to constitute an esthetic object in itself. The ostensive sign does not specify a particular mode of appearance of its referent, but the appearance of the referent per se; it stands for the substance of the referent, just as in mimetic desire a particular phenomenal aspect of the model can be said to stand for his being as a whole. Ostensive language does not express the content of the mimetic esthetic imagination but is only its formal sign, and what we have called ontological faith in language is precisely that this sign, given the conditions of linguistic presence, is sufficient.

A case in point is the magical use of a person’s name to “possess” his “essence,” a quasi-universal phenomenon in archaic societies, which is often guarded against by the expedient of maintaining this essence in a secret, sacred name re­vealed only to privileged members of the family group. The name is a merely formal attribute. But if its use is thought to give access to the substantial being of its bearer, then contemplation of the name may be taken as a linguistic equivalent to the contem­plation of some more palpable manifestation. Mimetic desire is ultimately concerned after all not with appearance but with substance, appearance being only a means to an end.

The inadequacy in our eyes of the esthetic that expresses itself in the evocation of a name in order to “possess” its bearer comes, not from the impotency of ostensive lan­guage, but, on the contrary, from its excessive power, its proximity to the sacred. The desiring subject only contemplates as much of the model’s appearance as he needs in order to assure himself that it expresses the latter’s being: if a name suffices, he need go no further. Indeed, the concrete appearances that function in the same fashion in the extra-linguistic imagination tend to be of a similar fragmentary, un-esthetic character, as in the possession through a piece of cloth­ing, a lock of hair, etc., exercised by practitioners of voodoo and similar rites.

Yet in the iconic realm, there is no clear frontier between the magical and the esthetic. Although a stick figure or voodoo doll might suffice to permit “possession,” the cave painters of the animals whose beauty still amazes us today were almost certainly not displaying their skill for the sheer esthetic satisfaction of their fellows; these paintings, like the art displayed in cathedrals, clearly had a reverential purpose, as well as an ultimately alimentary one.

In the formal, minimalistic domain of language, however, the distinction between the elementary linguistic forms and the declarative is critical. Although the elementary forms produce an esthetic effect, the dependence of the sign on the speech situation precludes the development of an esthetic internal to linguistic communication, that is, a literature. This is merely to reformulate in esthetic terms the dependence of the elementary linguistic forms on the linguistic scene that makes them incapable of furnishing objective models of reality. The “excessive power” of the ostensive-imperative that transcends the esthetic sphere is the exact counterpart of its non-objectivity; the failure to distinguish between desire and reality is in both cases inherited directly from the originary scene of representation.

Hence what we may call the esthetics of the elementary forms can only be grasped by going beyond the linguistic models themselves to the entire scene on which they are presented. The pragmatic paradoxes generated by the use of these forms in specific situations can then be understood in esthetic terms and the emergence of first the imperative and then the declarative “solu­tions” to these paradoxes as steps in the evolution not merely of the objective representation of reality, but of esthetic expression.

In this perspective, because the ostensive is dependent on the presence of its object, its own “esthetic value”—its capacity for evoking when contemplated by its hearer the being-for-desire, or simply the significance of this object—is limited to the moment of deferral in which it presents itself as a representation of the object. In the development leading to the constitution of the impera­tive, this deferral is prolonged in an awaiting in which the sign designating the object becomes for the hearer the stimulus to a practical performance. Here, for the first time, the utterance can function outside its practical use as an object of esthetic contemplation. Insofar as the imperative remains an “inappropriate ostensive,” its utterance, rather than being realized in practice and thereby annulled as an expression of desire, may be merely contemplated as a sign of the absent desire-object.

But it would not be accurate to speak of the object of such contemplation—the linguistic expression of mimetic desire—as an imperative utterance. Rather, the dialectic of the inappropriate ostensive may be said to lead to two complementary results. One is the imperative, in which ontological faith in language is made the basis for a praxis that resolves the paradox it contains. But the other is the esthetic contemplation of the linguistic sign, in which this paradoxical faith is not tested in practice but enacted, without resolution, in the imagination. This enactment may be said to be the birth of esthetic expression as such; but its effectiveness remains dependent on the linguistic presence that gives the sign a potential albeit unused power over its real referent. Thus “inappropriate ostensive” language can be said to afford linguistic expression to mimetic desire, but only to the extent that it takes real beings as its models, for the relation between the substance of the model and its (linguistic) “appearance” or attribute must be guaranteed in the real world, and not, as in esthetic representation proper, within a fictional universe.

With these considerations in mind we may now turn to the declarative form, which by providing a context-free model of reality, possesses by the same token a “context-free” esthetic. Here the linguistic object presented for contempla­tion is not a mere sign attributed to its referent by the speaker under the guarantee of linguistic presence, but an articulated model consisting of topic and predicate, in which the topic sign refers to what may at this point truly be called a signified, to which the predicate furnishes a context-free attribute. The declarative begins where the ostensive-imperative ends, with the designation of an object of desire guaranteed implicitly by a preceding imperative, that is, by the implicit interest of the hearer, and not merely in general terms by “ontological faith.” But instead of replacing the sign by its referent as with a successful imperative, the declarative “comments on” the referent with a predicate that explicitly constitutes it as inac­cessible to the desire of the (imperative) interlocutor. Thus the declarative model, because it is indifferent to the “magical” power of desire as expressed in the impera­tive, obliges the subject of this desire to contemplate its object esthetically, within the representational confines of the model, producing the mental oscillation between sign and imagined referent that defines the esthetic experience.

The relation thereby established between speaker, hearer, and object is structurally identical to the “triangular” model of mimetic desire put forth by Girard, which it can be said to “express” in the same way that inappropriate-ostensive language expressed the desiring subject’s ontological faith in the sign. Thus in the sense in which desire may be defined as an intersubjec­tive relation, it can only be said to emerge at this stage. But to define desire by its fully evolved configuration would lend itself to the same criticism as the choice of grammarians and linguists to define linguistic form on the basis of the declarative sentence. Such definitions foreclose the possibility of generative analysis.

In Girard’s model, the mediator openly or covertly designates, by means not specifiable in advance, the desire-object to the subject. The vagueness of this designation is analogous to the indeterminate nature of the relationship between the declarative speaker and the predicate attributed to the object. This predicate, as we have seen, is necessarily temporal, and its temporality designates a moment, real or imaginary, of the declarative speaker’s experience that is presumably not shared by his interlocutor. But this experience of the object is not presented as such in the declarative model, from which the speaker qua speaker is absent; the model is merely under­stood to be founded on temporal experience as the source of its predicate.

In what Girard calls “internal” mediation, the medi­ator may well be invisible to the subject, or even an anony­mous on/Man of social judgment. Here the field is less the domain of an overtly privileged model than that of social experience in general, and the specific value of individual objects of desire is wholly dependent on their status in a given social group. Triangular desire is thus both more objective and more subjective than the desire expressed by the imperative. On the one hand, it depends on the “objective” form of predication, but on the other, because the source of this predication—that is, the media­tor—is absent from the declarative model, the valorization of the object is cut off from the public presence implicit in the ostensive sign.

The declarative sentence, taken as an objective model of reality, is the foundation of metaphysics. (See “Plato and the Birth of Conceptual Thought,” Anthropoetics 2, 2, January 1997; http://anthropoetics.ucla.edu/ap0202/plato .) Predication, its distinguishing feature, is both the source of its objectivity (as it situates the topic outside linguistic presence), and at the same time, the expression of the mediating role of the (declar­ative) speaker between his hearer (the speaker of the original imperative) and the object of his desire (the topic). When this role is grasped explicitly by the hearer, the intersubjective situa­tion is that of the dialogic schema discussed a propos of the imperative (see Chronicle 545). But in the esthetic contemplation of the declarative model, the speaker does not appear; instead there is only predication “in itself.” Thus it is not the predicate but its predication of the topic that reveals to the detached observer what it hides from its desiring hearer: the presence of a second subject as mediator of his desire.

We can now understand more concretely the position of the declarative in the dialectic of desire. The first (impera­tive) speaker of the imperative-declarative dialogue already desires the object, because he pronounces its name in order to acquire it. The declarative response transforms this desire into its “triangular” form by providing a mediator in the person of the second speaker, and a predicative attribute independent of linguistic presence. The subject’s original desire was founded on the magical power of language to make present its referent; now this making-present­ becomes purely imaginary, and at the same time, defined for him by another. The second speaker acts as an unavowed mimetic rival, maintaining the desire-object in its inaccessible position in the declarative model through his act of predication, although at the same time not revealing his own agency within his utterance.

In practical terms this utterance carries information that may be of use to the first subject in realizing his desire. But insofar as the utterance itself becomes, on the scene of representation, an object of con­templation, the declarative expresses the inaccessi­bility of its topic within the real world. Thus to realize the desire, the speech situation would have to be abolished rather than, as in the imperative, fulfilled. Worldly fulfillment thus now appears as incompatible with the scene of representation, which generates an imaginary, or more precisely, a fictional universe. The pro­longation of the linguistic presence of the declarative thus leads in the opposite direction from that of the imperative, because the hearer remains thereby immersed in a fictional world in which his desire is incapable of practical fulfillment. The fiction of the declarative sentence can only prolong itself in fictional discourse, which is to say, in literature, where the imagined object is constantly reconstructed synchronously with the temporal progression of the work.

The great flaw of “esthetics of literature” is that they begin with an anesthetic notion of language proper and are then obliged, in order to understand its literary use, to posit a “literary language” defined by some mysterious “esthetic” difference from “ordinary” language, like the “opacity” by which Sartre distinguished poetry from prose in his 1948 Qu’est-ce que la littérature? The indefensibility of this position is apparent even from the limited perspective of pragmatic linguistics. But it is not enough to speak vaguely of the “esthetic element” in language without making clear the structural relation between linguistic form and the esthetic.

Thus the question must be turned on its head. Language is from the outset an esthetic phenomenon as much as a communicative one, the two functions being only fully sepa­rable at the discursive level. Even the ostensive, which in linguistic presence merely designates a present referent and thus can operate as an esthetic expression only during the quasi-instantaneous (although repeatable) time of deferral, functions esthetically in signi­ficant memory, and lacking this function, its inappropriate use as an imperative—but also, as we have seen, for the esthetic contemplation of the ostensive sign—would be inexplicable.

The marginality of the esthetic, as exhibited in such phenomena as the bohemian life of artists and the estrange­ment of esthetics from the mainstream of philosophical analysis, perpetuates its original and never-quite­-forgotten connection with the sacred. This connection is as real, if less salient, in the minimal, formal domain of language as in that of iconic representation. Yet no student of painting or sculpture considers the pictorial or plastic components of artworks as in themselves devoid of esthetic interest, apprehended and integrated by means of a formal “compe­tence” to which the esthetic is irrelevant. If we are less likely to insist on the fundamentally esthetic nature of language, it is because the esthetic’s irrelevance to the practical world of science and technology makes it appear as a “supplement” to language rather than part of its original essence. This mind-set is that of meta­physics become no longer an auxiliary but a substitute for religion. The arbitraire du signifiant in the world of declarative language is a rootless world of “structures” that are “applied” to nature but shorn of their own natural, let alone anthropological, ontology.

The revelation of the internal contradictions of the meta­physical world-view, long relegated to the marginal domain of esthetics, has since Nietzsche become a staple of continen­tal philosophy, of which Derrida was no doubt the ultimate exponent. Analytic, “Anglo-Saxon” philosophers have safely ignored this trend, which they not unjustifiably dismiss as tainted with estheticism. Thus as both schools increasingly concentrate their attention on language, one attempts to deconstruct its basis without examining its forms, whereas the other attempts to construct its forms—increasingly in the context of “artificial intelligence”—without examining their basis. Generative anthropology, by demonstrating the centrality of the esthetic/sacred element in the genesis of language, and understanding the “context-free” declarative as a product of linguistic evolution that liberates us from the immediacy of the speech-situation, would substitute a historical dialectic for this dialogue de sourds.