By his acceptance of the speaker’s desire, the addressee of the imperative becomes not merely the latter’s hearer but his inter­locutor. Thus he hears not only the utterance but the person, and by hearing the utterance as a personal one, he comes to assume on the scene of representation the role of member of the community and respecter of its norms of significance. In this fashion, the addressee of the imperative creates the possibility of dialogue with the speaker.

At the same time, this dialogue demonstrates the inadequacy of the imperative model to fully comprehend the fundamental asymmetry of the speech situation. In the exchange at the operating table—”Scalpel!”-“Scalpel”; “Forceps!”-“Forceps,” the ostensive serves as a reply to the imperative. The hearer’s reply “corrects” the first speaker’s “inappropriate ostensive,” universalizing his expression of personal desire by asserting the objective presence of the object. Yet this reply in no way transcends the asymmetry of the impera­tive intention. The second speaker responds in a form differ­ent from the first, and which permits of no further dialogue; the content of his utterance is entirely determined by that of the first speaker, which he simply mimics. The identity in linguistic substance of the two utterances reflects the dependency of the second speaker’s role, in which making an ostensive reply is only possible upon prior comple­tion of the requested performance. The imperative-ostensive dialogue represents the expres­sion and annulment of desire, the successive pairs of utter­ances marking the beginning and end of successive periods of awaiting; the fundamental asymmetry of the two speakers with respect to this desire and this awaiting is never called into question.

The addressee may of course refuse the imperative and put an end to the conversation. But even if he accepts the role it designates for him, he may not be able to carry out the required performance. In the absence of performance, no linguistic response to the imperative is possible, even if the addressee has no desire to violate the terms of the imperative dialogue—a violation that might lead to unfortunate consequences.

In this situation the contradiction inherent in the imperative model becomes explicit. This contradiction is not a purely private one, as was that between individual desire and the communal sacralization of the originary object. Nor does it stand­ in an ambivalent position between the “private” and “public” spheres, as we found in the second stage of our dialectic, where the “inappropriate” ostensive, an expression of individual desire, came to be interpreted as a legitimate imperative speech act. Here the contradiction overtly involves the distinction between the speaker’s and hearer’s model of the imperative. It is not yet a fully dialogic conflict because it cannot be assimilated to a symmetrical disagreement in which one speaker contests the other’s objectivity. The hearer’s model merely permits him to understand the situation in a way closed to the speaker, for whom the possibility of nonperformance cannot arise because per­formance is already included in his intention.

It would be a mistake to view this situation as that of a contradiction between the imperative model on the one hand and “reality” on the other. This contradiction of course exists, but from a representational standpoint it is mediated by the speaker’s relationship to the hearer. For the referent of the imperative model is not merely an object of desire, but the object of a desire expressed on the scene of representation, and its impossibility of fulfillment, before leading to a contradiction in the sphere of reality, provokes a contradiction within the limits of this scene. This fact would certainly be obvious to the addressee, to whom “reality” is apt to afford little protection from the wrath of the speaker.

The hearer’s model of the imperative was never blind to the possibility of non-fulfillment, because the speaker’s request appeared in it from the beginning not as an objective model of reality but as an expression of desire. But the non-identity of the hearer’s intention with the speaker’s remained only latent so long as the former was able to bring the two into coincidence by fulfilling the expressed desire of the latter—a coincidence expressible in an ostensive reply. Now the non-coincidence becomes a contradiction, although, as befits the asymmetrical structure of the imperative, one visible only to the hearer. Even if “reality” is the ultimate source of this contradiction, the situation would be little different if this source were the latter’s recal­citrance, for in any case his problem is to maintain, despite the contradiction, his relation of linguistic presence with the speaker. But this implies that instead of using an ostensive to express the presence of the object, he must find a way to communicate its absence. It is this need that will lead to the creation of the declarative.

The speaker of the imperative awaits a performance by which his utterance will be realized as an ostensive (whether or not actually spoken by either party). The addressee, lacking the possibil­ity of producing the object, must produce an utterance that will have to be accepted in lieu of the object. What is required is, so to speak, a negative ostensive—the contradictio in adjecto being merely a reduced form of the contradiction between the hearer’s and the speaker’s intention. Once the reduction is accomplished, however—that is, once the possibility of expressing the contradiction by an utterance is recog­nized by the addressee of the imperative—the negative-ostensive resolution of his paradoxical situation will become possible. It suffices that the materials of which this solution is constructed be available: the ostensive, and the concept, or more precisely the operator, of negation.

Negation and the Imperative

The ostensive admits of no negation because its referent is required to be present, and this is true even if the referent is itself “negative” in character. Thus, for example, we may consider the familiar utterance “Help!” despite its imperative appearance as in fact an ostensive. Although from a semantic standpoint help is obviously what is being requested, the word embodies no representation of the aid to be furnished; what it expresses is rather the presence of a help-requiring situation. And in the absence of such representation, no imagined help exists either in the mind of the hearer, who knows only to come to the rescue.

Yet the notion of interdiction is as old as the originary sign itself, which is in a sense its chief component. The originary sign designating the sacred object can be seen as a negative imperative avant la lettre, indicating that its referent is not to be appropriated by any of the members of the group. From this perspective, in the passage from osten­sive to imperative, a communal interdiction is transformed into one imposed by individual desire. Just as the object of the ostensive, whether an item of value or of danger, is in general not to be appropriated individually by any member of the group, the object of the imperative is designated for appropriation by the speaker and by the same token refused to the hearer. It is the explicit formulation of this refusal, once the ostensive can be used “inappropriately” as an expression of desire, that will constitute the negative imperative.

We have referred to the desire aroused by the ostensive as the motive force for its inappropriate use and thereby for the emergence of the imperative form. But the expression of desire by one speaker should not be taken as a sign of its extinction in others, but rather of the contrary. The power of the scene gives a supplemen­tary force to expressed desire and thus permits the constitution of the imperative; but this expression of individual desire can only mimetically strengthen the interlocutor’s own unexpressed desire for the object. To the extent that the (ostensive) sign can be used imperatively to demand an object, there would arise the need for an operator of negation to permit the speaker to make explicit his supplementary interdiction of the object to his interlocutor.

The speaker of an interdiction implicitly recognizes in the other a desire similar to his own, yet within the intentional structure of the interdiction, no symmetry is established between this desire and that of the speaker which contravenes it. On the contrary, interdiction addresses itself not to the desire of its hearer but to his action. If anything, the asymmetry between speaker and hearer is even greater here than in the prescriptive (positive) imperative, because the desire of the former is now realized explicitly in the negation of the activity of the latter.

The interdictive imperative thus tends to imply more read­ily than the prescriptive the preexistence of a relation of authority between the interlocutors. The strength of this relation is most evident in a phenomenon we may call “normative awaiting,” which is particularly although not exclusively characteristic of interdiction. In such cases the imperative scene is not terminated by any specific performance, but prolongs itself indefinitely into the future. Thus a mother who tells her son “Don’t play in the mud!” does not await any specific act, even an act of renouncement, although a sign of such renouncement might be expected to terminate linguistic presence stricto sensu. She simply states a general rule of conduct, and will consider the interdiction to be violated if contravened at any future time. A similar situation, we may note, is created by a positive normative imperative such as “Don’t go out without your scarf!” or “Keep your hands clean!”

In normative awaiting, the linguistic pres­ence of the speaker is in effect indefinitely prolonged—one might say, as a “superego”—so that any offending conduct becomes a violation not merely of the speaker’s desire but of the scene of representation, guaranteed by the community through the mediation of the sacred. This form of the imperative thus plays a major role in the maintenance of the social order. The normative propensity of the interdictive form is of interest here because it serves to emphasize the asymmetrical attitudes toward desire implicit in the imperative model. The speaker’s desire, here as always, is identified with the maintenance of the communal scene, whereas the interdiction makes that of the hearer incompatible with this presence. In the normative imperative, whether prescriptive or interdictive, the expression of the speaker’s desire may well be the repeti­tion of a generally accepted norm, the original pronouncement of which may even be attributed to a sacred being. But even in this case, the communal authority of the norm and the awaiting it imposes is realized in the speech situation only through the intermediary of the speaker’s own authority.

The negative imperative, as it appears from this discussion, is not an independent linguistic form; it differs from the prescriptive variety only in its content. It would further be a mistake to classify the positive/negative dichotomy as a gram­matical paradigm like that of person. As we have defined the grammatical, it functions to objectify linguistic intention­ality by discounting the particular temporal or spatial condi­tions of speech acts. The positive-negative dichotomy is not paradigmatic in this sense, affirmation and negation not being in any sense a pair of “shifters.” As an operator, negation affects  the entire content of the performance re­quested by the speaker.

The effect of an operator must be distinguished from that of a simple modifier. In the imperative, modification, like governance, receives a primitive form of grammaticalization. Requests for physical objects, for example, must in practical contexts distinguish between the category of things and that of qualities; once more, the performative nature of the impera­tive model provides the impetus for grammaticalization. Thus if a big hammer is requested, a small hammer will generally be more acceptable than a big basket. The same analysis evidently applies to constructions like “Come quickly!” where a verbal rather than a nominal request is qualified.

Hence we can consider imperative modification to be, like governance, in a state of incipient grammaticalization. Whether or not specialized lexical terms existed, the intentional structure of the impera­tive provides in its asymmetry a model for asymmetrical relations of both kinds, although this asymmetry is not fully realized in the linguistic model. Thus as in the case of governance, analysis into, for example, “noun” and “adjec­tive,” although implicitly carried out by the speaker, is structurally speaking a matter of concern only for the hearer, because it develops from the analysis of his performance. The asymmetry that provides the foundation for this analysis is at the same time an obstacle to its formalization in the linguistic model, because the speaker’s words, unlike the hearer’s actions, are intended to produce fulfillment through their mere presence on the scene of representation—the imperative remaining always, in essence, an “inappropriate ostensive.”

The use of operators to modify the imperative perfor­mance model as a whole constitutes the limit of semantic polarization possible within imperative language. In “Big hammer!” or even in “Run fast!” the requested performance is merely more specific than in “Hammer!” or “Run!” In a case like “Run again!” however, what is added by the operator again is not a specification of the object/action requested, but of the performance in which this object/action is included. An operator takes the substantial part of the linguistic model not merely as designating an object or action, but as a performance complete in itself. And in negation, the operator most clearly distinguished from a simple modifier, the polarization must be fully conscious to the speaker as well as the hearer. On hearing “Don’t walk!” not only the hearer but the speaker as well must consider that the command cannot be separated into the substantive action of walking and a secondary but indepen­dent quality that attaches to it.

Since interdiction operates on the imperative model as a whole conceived in a positive sense, we may represent this situation by the equation:

Perf (~X) = ~ [Perf (X)]

where “Perf” refers to the performance requested by the speaker. We must be careful to distin­guish this equation from the apparently similar but incorrect:

Imp (~X) = ~Imp X

where “Imp” stands for the entire intentional structure of the imperative, the sense of which would be that to forbid something is simply not to order it. To claim that this distinction is a merely logical one would miss the point that we have specifically defined the negative operator in imperative language. For the only thing wrong with the second equation is that it contradicts the normal functioning of imperatives as we know them. Declaratives do not function in this manner: There is ~ man = ~ [There is (a) man], and the “illogicality” of such constructions as “must not,” which ≠ “~ must,” is indeed traceable to their connection with the imperative. In the imperative, the operation of the negative is fixed at the level of performance. In interdiction, the performance of the addressee is the negation of a possible (“positive”) perfor­mance, but the fundamental elements of the imperative intentional structure remain the same, the nature of the awaiting merely being altered to fit the non-perfor­mative nature of the request. This represents the closest thing to a logical paradigm within imperative language and thus its high­est level of what we may call “thought” or “reasoning”—manipulation of linguistic models as context-free substitutes for reality.

To be continued…