Issue #64 August 2023

Between Historical Reduction & Metaphysical Construction: Critiques of Representation in Foucault and Deleuze

Gustave Courbet, "The Wave", (1869)

I. Introduction

Our question will concern the critique of representation. We will need to account for what it is that representation means in the works of Deleuze and Foucault who are here considered, for we may find that even if its general concept is given a more or less similar description by them, the question of its function, which implicates the question of its possibility, diverges. From this, the order of a question may emerge that cannot be filled by either of their accounts, but emerges uniquely in the difference that separates them.

It should not be surprising that this would be the case, for despite the historical moment that comprehends Deleuze’s and Foucault’s projects as intimately bound, the field and consequences of their investigations are in fact quite different. This is to say, if we reside at the level of method and the field to which each methodological form refers, then we arrive at very different levels of conceptual experience whose unfoldings resist assimilation through certain presuppositions their analyses share by virtue of which this intellectual moment is, rightly or wrongly, superficially or profoundly, identified. The dislocation of subjectivity from the conditions of its speech, the destruction of metaphysical unity, the severing of the problem of origin and essence…, in other words, all of what had been prepared by previous writers for their further elaborations should not allow one to forego the philosophical accounting of difference, especially where the sense of difference strikes with the power of creation: in the singular contours of a work.

Why does the question, then, concern representation? In what way does representation as a concept and practice of language bring to the surface the divergent strands of inquiry we are to follow? It is this question that will open a field of problems whose solution will only be given in the consolidation of the problems as such, to use the expression of Deleuze. To further anticipate through appropriation of a certain line of his work, it shall here be a matter of establishing a certain type of seriesotherwise known as a dialecticthat may be able to produce at the level of propositions the differences that differentiate the levels proper to the bodies of work we will follow. What should not be expected from the following elaborations, of course, is a systematic rendering of the problems at stake, nor a proposed unification of the strands that unfurl in the specificity that is their own: it does not require much reading of the writings of the two thinkers to comprehend how this would be inappropriate to the sense of what is pursued. What is unsystematic, however, should be so for a reason, and it is this reason that is intimately bound, though in different ways, to a critique of representation.

For the sake of a certain clarity towards which the English language is often bent in pursuit, but more importantly to allow the greatest space possible for the expansion and layering of concepts, we shall begin by characterizing the different series to be constructed in accordance with conventional names, conventional because they are among the most general names made possible through a certain use of concepts that are here at issue. That one is historical which Michel Foucault embodies in his method, and metaphysical which Gilles Deleuze elaborates. In accordance with the good sense provided by a common sense of termswe shall not deny yet this evidence, one can say that what is historical concerns a determination of time with respect to a series of actual events, and what is metaphysical, a series outside of actuality though not outside of conditions of time. With these very general definitions, we pose a provisional horizon that interpretation will later differentiate. For it is the case, beyond this naivete of terms, that representation of actuality through language is led back to conditions that challenge the notion of historical progression itself, and in the other series, the metaphysical one, is subjected to a critique as it were from within, by means of a language that undoes the possibility of a metaphysics whose conditions of stability are given through representational form.

It is our view that between the work of each, the theory of language developed, though consonant in many ways with respect to a critique of representational norms and their moral and political correlates, nevertheless decisively depart concerning the type and genesis of the structures that perform this critique. Indeed, we will attempt to identify fundamentally different levels of analysis that bring with them separate fields of experience that involve the structures that shape them. This is to say, the forms of evidence that shape the possibility of language and conceptual development unfold at different levels. The question will be to what extent these levels can be thought coextensively, which is to say, co-existentially.


II. Elements of Critique

i. Énoncé

It is the interpretive aim of Michel Foucault to rid the interval between word and thingor, at the level most of his analyses implicate, discourse and eventof an internal genesis of sense, that is, a synthetic principle valuable as an explanatory mechanism for the binding of linguistic elements to aspects of a material reality. If such a characterization is apt as a global description of an orientation, this amounts to a type of critique of the representational function, which in its diverse forms may be understood as a form of explanation that locates the linguistic-material interval as a site of synthetic creation, regardless of which extreme, linguistic or material, is given greater determining force. To the extent Foucault’s analyses admit of the necessity of synthetic activity in the production of sense and thereby grant validity to established representational mechanisms that determine causal influence, the importance falls elsewhere when it comes to determining not the internal conditions of knowledge, but what a given knowledge is formally unable to determine yet is nevertheless indispensable to it. Knowledge and its objects are taken as the material for another kind of analysis which is, in a very specific sense, a historical analytic of knowledge claims, an “archaeology of knowledge”. An “analytic” as it is understood through common philosophical usage is a gathering or exposition of the essential aspects of a thing under consideration. But what determines the specificity of this analytic is a very different sense of what constitutes such essential aspects: in many ways, what appears inessential at the level of representational form will take the place that determinations of essentiality normally occupy, not so as to displace the validity of scientific knowledge, but to explore a domain it cannot think yet is in some sense bound to its possibility.

We would do well to begin with an exploration of a term developed by Foucault in Archaeology of Knowledge that goes the furthest in laying out a systematic approach to the study of linguistic entities: the énoncé (statement, enunciation). The challenge raised by Foucault’s elaboration of the énoncé is that of thinking the possibility of the repetition, or reproduction, of language and its attendant practices in a way that does not rely on a causal order of explanation given through the grammatical relations of linguistic phrases, the logical forms of propositions, or the subjective order of psychological processes. Enunciative analysis is developed as a type of critique that takes what transpires in language at the level of an existence that restricts the set of linguistic possibilities at a given time, such that the “enunciative field” that comprehends modes of existence is not beneath the representationally evident discourses that reproduce this existence. The énoncé, as Foucault understands it, produces an enunciative possibility that is not identical to the conditions of a representational system, yet is only given through these systems. “Discourse” or “discursive formation” is the name through which we are to understand this complex of interactions comprising both representational forms and their enunciative conditions. Discourse, thus,

“is not an ideal, timeless form that also possesses a history; the problem is not therefore to ask one-self how and why it was able to emerge and become embodied at this point in time; it is, from beginning to end, historical – a fragment of history, a unity and discontinuity in history itself” (AK, 131).

It should first be said that the enunciative field as a site of intersecting discursive formations is that wherein historically specific games of truth and recognition are put into play and is not considered a sum total of these systems or an in principle total description of their emergences and interactions. For,

“the analysis of statements [énoncés] does not claim to be a total, exhaustive description of ‘language’ [langage], or of ‘what was said’. In the whole density implied by verbal performances, it is situated at a particular level that must be distinguished from the others, characterized in relation to them, and abstract. In particular, […] it is another way of attacking verbal performances, of dissociating their complexity, of isolating the terms that are entangled in its web, and of locating the various regularities that they obey” (AK 121).

In other words, it would be absurd to imagine such a total level of analysis possible, not only because of the impossible demand it would make on the empirical knowledge necessary to describe it, but also because such an analysis is contrary to what the enunciative field is meant to show: the modalities of recognition and validity any given system assumes in such a way that remains unthought within its terms. Such modalities are called the “historical a priori” of discourse, or what remains both within and outside of representational systems. The historical a priori is thus not a condition of validity, but the force of rules within an historical existence that orient and bring together empirical laws and their objects to make appear as unified the disparate levels of which a representational system is composed, “an a priori that is not a condition of validity for judgements, but a condition of reality for statements [énoncés]” (AK, 143). Foucault’s analytic of discourse thus circumscribes, or reduces, representational truth claims so as to arrive at a level where their internal criterions of validity no longer define their conditions of possibility. The study of discursive formations is one that understands representational forms within a discursive set that comprehends the dissolution and repartition of what is utterable as language, which is to say, the forms of exclusion and identification necessary for the pronouncement of truth.

In more concrete terms, through enunciative analysis one is to think the creation of discourse and the repetition of its forms at the level of institutional enactments: in Foucault’s rendering of what constitutes an énoncé, the historical institution is both the discursive condition and the set of material consequence of this condition, as the énoncé is nothing outside of its institutional inscriptions and prescriptions. Precisely for this reason, the discursive relations it identifies are neither elements of language nor of practice, but a frontier that divides them through establishing their relation: ” not the language (langue) used by dis course, nor the circumstances in which it is deployed, but discourse itself as a practice. […] [A]nd what we discover is neither a configuration, nor a form, but a group of rules that are immanent in a practice, and define it in its specificity” (AK 51). This forms the reason why Foucault’s attempt to characterize his conception of the énoncé at a purely linguistic level brings forth a series of negations (neither a phrase, nor a proposition, nor a speech act…), as it is neither the specificity of its spatio-temporal enunciations, nor the generality of the linguistic form it transmits. But if this is the case, why should this term “énoncé” be given to what is really a description of the historical specificity of knowledge claims that the material forces of institutions determine?

A response may be given on the basis of the idea that Foucault’s theoretical descriptions of the relations between institutional power and truth necessarily entail a central importance given to the mediating dimension of language as such, insofar as the materiality of practical formations could find no site of deployment outside of the linguistic forms that open up a generalizable conception of order. Yet it is the nuance of his position to insist, in opposition to phenomenological interpretation with which his work is bound up, that this horizon of ideality is the effect of certain historically sedimented material configurations. Of course, this line of reflection expresses a form of explanative regress, but we might say that this is invited on the condition that it escapes a merely metaphysical figure and takes on the visage of complex form of causality that characterizes historical immanence as a movement of the involution and transcendence of material and ideal events. The énoncé in Foucault’s account is the theoretical crystallization of this movement expressed in the descriptive architecture of his other works.

The other significant pathway treaded by Foucault’s formalization of the type of historical intervention he had attempted to date of Archaeology of Knowledge has to do quite naturally with what any other study of language ultimately registers as, if not its end, then at least an insuperable term in a series of analysis: the subjective reference. Yet based on Foucault’s understanding of an énoncé as the existence of rules that shape the deployment of language, human consciousness, whether individual or collective, is considered neither their origin nor destination: it is not their origin because what the enunciative field concernswhat an “historical a priori” expressesis the unification of heterogeneous domains that have no necessary identity, which is to say, no ontological filiation immediately gathered or made accessible by a horizon of consciousness. When Foucault speaks of “rules of formation” at work in a discourse, he is referring to a form of unification that has no synthetic agent, but rather what is produced by a non-localizable coordination of discrete entities:

“when one speaks of a system of formation, one does not only mean the juxtaposition, coexistence, or interaction of heterogeneous elements (institutions, techniques, social groups, perceptual organizations, relations between various discourses), but also the relation that is established between them – and in a well determined form – by discursive practice” (AK 80-81).

Yet, what can we say carries out the establishment of relation? How is the unification of a discourse in fact produced if not through an individual or collective subjective synthesis? To this, Foucault responds that it is through the discourse itself that this production occurs, which cannot be led back to a form of subjectivity for the simple reason that it produces the location we understand as subjective. Thus,

“[i]n the analysis proposed here, the rules of formation operate not only in the mind or consciousness of individuals, but in discourse itself; they operate therefore, according to a sort of uniform anonymity, on all individuals who undertake to speak in this discursive field” (AK 69-70).

But again, this anonymous uniformity, what is it?

“He is not in fact the cause, origin, or starting-point of the phenomenon of the written or spoken articulation of a sentence; nor is it that meaningful intention which, silently anticipating words, orders them like the visible body of its intuition; […]. It is a particular, vacant place that may in fact be filled by different individuals” (AK 107).

Strangely, we seem to be led to the type of subjectivity that the propositional form defines in the field of representation: a determinate yet empty place that can effectively be filled by different individuals. What is the difference between these conceptions?

We would have to say that the formal difference between the subject of the énoncé and the subject of representation is slight, but that a certain inflection creates a fundamental change. Linguistic performance, with respect to the demands of a representational system, is essentially a matter of correctness, where the criterion of validity established by the set of rules specific to a language are related to the objects already in part defined by the rules, but whose spatio-temporal iterations open up both the possibility of the extension of rules and that of their error, a space of error which, beyond simply mistaken attribution, is necessary insofar as the variations in objects can never be fully circumscribed by syntactic form. The regularity of a linguistic system thus exercises its regulative use by distributing truth and error among the subjects who perform it, where the good subject is the one who not simply reproduces the general rules in particular cases—though this be a necessary condition of the subject’s recognition and acceptability—, but also the one who extends the rules to those things and situations that have not yet been brought within their scope, thus confirming apparent irreconcilability as belonging within the scope of a system. The subject of the énoncé, by contrast, occupies a level where such judgments of value that necessarily inform the rules of a system do not fully penetrate, for it brings into relief certain conditions of subjective formation that make such evaluation possible. This is not to say that there is an “enunciative subject” independent of the layering of rules and their evaluative prescriptions, but there is, for Foucault, at least the possibility to circumscribe the conditions of emergence of the subject of representation, its “subjectivation”. The difference thus essentially resides in knowing “what position can and must be occupied by any individual if he is to be the subject of it” (AK 107).

For the moment, we will hold to the concept of the subject of the énoncé in its generality: as an analytic category in the analysis of discourse, posing what is required for subjective recognition. To what extent this relies on principles of synthetic production remains a question. But to move closer to the possibility of fully posing it, we take up a different way of exploring the problems posed by linguistic form.

Gustave Courbert, "The Wave", (1869)

ii. Exprimé

In the work of Gilles Deleuze, we find through various figures the attempt to sustain a few principles that are at all times in the process of disappearance. That which evades propositional form, or what cannot be circumscribed by the progression of representational signifiers, is the ground from which his thinking forms a current and to which it returns. The project of a dislocation of concepts from the subjective and objective coordinates of identification that define the most intrinsic demand of rational progression is perhaps nowhere better articulated within the terms of that idea of reason than in his writings. Perhaps nowhere else do we better find philosophically expressed the impossibility of philosophy, an impossibility that forms the condition and dissolution of conceptual movement.

When taken at the level of a consideration of the deployment of language, what is it we can say of the dimensions invoked in his work? At what levels must we reside to be susceptible to the demands it makes? It is necessary to traverse the works most forthrightly oriented by a critique of representational language, a tentative shared between Difference and Repetition and Logic of Sense. We will begin with the latter to understand how this is carried out by his reevaluation of the production of linguistic sense, effectuated through a certain status given to the aspect of “expression” and its object, the exprimé.

Deleuze’s logic of sense can be paraphrased in the following way : the productive (synthetic) dimension of language resides neither in the material things that inhabit the words and propositions we proffer to one another and towards which these linguistic entities would have no other sense than to evaporate in the presence of their disclosed being (standpoint of designation); nor in the ideality of the linguistic construct taken as the bearer of a formal reality that differentiates the flux of material states according to necessary rules and their reproduction (standpoint of signification); nor, finally, in the ground of a subjectivity that would act as the synthetic relay between the particularity of the existent and the universality of law (standpoint of manifestation). There is thus sought a fourth order of language that is not only not reducible to either of the other three but in some way traverses each of them as a superior instance in the production of sense as the genesis of the difference between words and things (standpoint of expression). According to Deleuze, it is from this standpoint that the identity and difference of the ideal and material event is expressed, posing the constitution of a transcendental plane outside of the conditions necessary for a statement to be considered true or false, which is to say, in accordance not with representational validity but with the ontological univocity of signifier and signified. It is here that “nonsense” is seen not as lack or mistaken attribution, but as an inherence of words and things that expresses a primacy with respect to the state of causal identification made possible through the difference in nature introduced between the signifying and signified orders, their ontological equivocity. It is in this way that the productivity of representation through which self and world are generally known, and by which positive knowledge is established, is led back to an instance that signifies nothing, and from the standpoint of representation, is completely sterile or inefficacious. It is however, for Deleuze, sense in its aspect of emergence, which he gives various names: pure event, incorporeality, quasi-cause, paradox. All of these terms can be comprised within that of the exprimé.

The basic concepts of this critique of representation are relatively simple: in the chain designation-manifestation-signification, a circularity is created such that each asserts itself as the condition of the other, thereby returning to the form of possibility created in the image of the other. Designation asserts the primacy of the object whose possibility is then claimed by the manifestation of a subjective field which in turn is referred to a signifying order of causation that opens the ground of subjective presence, returning to a primary objective order that in turn can only found its veracity by referring itself to the presence of a designation. And so, the signifying order is supposed to found the veracity of designated states of affairs, yet can only claim this veracity insofar as the designation is considered true. Manifestation is supposed to found the presence of the object, but can only do so insofar as it makes use of causal ideas that go beyond the form of subjectivity (world, God). Thus, Deleuze wants to say that the representational proposition can only generate sense through a sort of recycling of conditions. It is difficult to say exactly how this critique applies to empirical knowledge claims as developed in the sciences, but at the level of a critique of the metaphysical grounding of such claims carried out largely by the philosophical tradition, it has intelligibility. According to him, signifying propositions are such that

“[i]n discussing the conditions of truth, we raise ourselves above the true and the false, since a false proposition also has a sense or signification. But at the same time, we define this superior condition solely as the possibility for the proposition to be true. This possibility is nothing other than the form of possibility of the proposition itself” (LS 22).

The form of possibility as transcendental key is, then, nothing more than the image of what is conditioned. And so, in constructing a theory of conditioning possibility, one does no more than describe in general the conditions under which a designation may be true or false, which is ultimately something empirically decided.

Now, one could say that this is precisely the basis of knowledge: there is nothing other than empirical events and the conceptual forms that generalize their evidence. This is, one could further say, the genetic power representational form locates, in that the linguistic-material interval is the condition of the proliferation of linguistic and material series in their possibility of relation. Yet Deleuze wants to insist that sense is not created at this level, that one must search elsewhere for its irruption. One must engage a metaphysics of truth, yet do so outside the order of conditions of possibility produced in the image of the conditioned:

“For the condition of truth to avoid this defect, it ought to have an element of its own, distinct from the form of the conditioned. It ought to have something unconditioned capable of assuring a real genesis of denotation and other the other dimensions of the proposition […] sense, the expressed of the proposition, is an incorporeal, complex, and irreducible entity, at the surface of things, a pure event which inheres or subsists in the proposition.” (LS 22).

Sense, as truly outside of resemblance with the conditioned, escapes all forms that the other dimensions of the proposition entail to the point that “we may not even say that sense exists either in things or in the mind; it has neither physical nor mental existence” (LS 23), all the while retaining “its genetic power as it animates an a priori internal model of the proposition” (ibid.). What exactly is the status of this nonexistent entity? How is it generative of the other orders of the proposition?

The problem concerns the existence of nonexistent entities. Yet the question that guides it is not formed in the abstraction and arbitrariness of examples (unicorns, etc.), but in the “transcendental sensation” of a primary sense generative though not determinative of what is normally understood through the differentiation of signifier and signified. Sense, for Deleuze, is the folding of the linguistic proposition and the thing designated—signifier and signified, or representation and represented—in such a way that reduces to neither, subsisting as the frontier between them: this is its “event”, the emergence of which is the occasion for the type of differentiation comprised in representational language. A rather delicate theory, because it relies on an ontologically undifferentiated level to explain the development of difference, risking always to fall into the depths from the surface it creates (the schizophrenic standpoint).

It is necessary to here specify what Deleuze understands by a transcendental condition, especially under these circumstances where he charges sense with a genetic power while at the same time describing it as an “effect”. For one could without much difficulty ascribe to his own theory a circularity that, beyond merely a regress of conditions, would express a simple illogicality, an endless and arbitrary swapping of the sense of cause and effect. However, it would be too facile to make such a judgment even if it often appears to be the case. We must keep in mind that “transcendental” does not mean for him the conditions under which a proposition is considered true, which in effect means that it does not concern the conditions of possibility of the identity shared between the representative concept and a material existence, which, finally, implies that it does not comprehend the grounds by which empirical knowledge is generated. And so, one has reason to doubt the possibility of his proposed derivation of the representative functions of the proposition, yet there is no reason to thereby doubt that he locates a dimension of language that is quite different than representation and even genetic in a certain way. But in what way?

“Sense is both the expressible or the expressed of the proposition, and the attribute of the states of affairs. It turns one side toward things and one side toward propositions. But it does not merge with the proposition which expresses it any more than with the state of affairs or the quality which the proposition denotes” (LS 25).

As we have said, Deleuze considers sense as a limit that resides at the point of differentiation between the series of signifying and signified terms while remaining ontologically identical to the differences thereby created. In this vanishing state of both…and…, neither…nor…, sense is both the exprimé of a proposition and an attribute of things, but derives neither from the proposition nor from a state of existence. Does this thereby mean a proposition or an existent is derived from it? Clearly, sense presupposes both the linguistic element and a state of existence. It is thus not genesis in an absolute sense that is at issue, but the “insistence” of a type of sense that presents an identity of expression between ideal and material terms while simultaneously presenting an internal structural difference between the series developed. This is Deleuze’s fundamental interpretation of signifier and signified, the development and relations of which can only be grasped serially because the genetic power of sense for him does not reside in the conditions of representational correlation inherent to scientific description, but rather in a type of excess and lack determinative of language as a temporal movement, one which he identifies uniquely as a mode of becoming irreducible to succession.

In effect, this recasting of the sense of signifier and signified as a serial movement of their mutual displacements consists in a doubling of each series such that the linguistic element is both designation and expression, the material element both the designated and the expressed:

“On the side of the thing, there are physical qualities and real relations which constitute the stae of affairs; there are also ideational logical attributes which indicate incorporeal events. And on the side of the proposition, there are names and adjectives which denote the state of affairs; and also there are verbs which express events or logical attributes” (LS 30).

Expression and what is expressed, the exprimé, is thus a doubling of both the proposition and the designated state of existence commonly understood through the terms of signifier and signified; but what is doubled does not have the characteristic of resemblance: the doubling, we are to understand, is conditioned by a fundamental difference, which is precisely the difference between word and thing. Yet for this difference to no longer remain extrinsic, that is, as a conditioning and conditioned term formed in the image of each other, Deleuze needs to introduce the identity of this difference. What separates this gesture from what is most typical of philosophical idealism, however, is that he doesn’t understand this identity as an essence. He understands it, rather, at the level of the ideality of language, which discloses its power in the figure of what cannot be said to exist in any determinate mode: paradox, or the fundamental possibility of impossible determinations. Language is a being-towards the paradoxical; it is this figure of the impossible beyond contradiction that animates the entirety of Deleuze’s logic of sense and is for him at the basis of the other determinations of the proposition, for he understands it to be that without which the movement between signifying and signified series would not come to be. It is here that we are to find the genetic power of sense:

“It is a two-sided entity, equally present in the signifying and the signified series. It is the mirror. Thus, it is at once word and thing, name and object, sense and denotatum, expression and designation, etc. It guarantees, therefore, the convergence of the two series which it traverses, but precisely on the condition that it makes them endlessly diverge. […] They are simultaneous without ever being equal, since the entity has two sides, one of which is always absent from the other. It behooves, therefore, to be in excess in the one series which it constitutes as signifying, and lacking in the other which it constitutes as signified” (LS 48).

The state of excess and lack of which Deleuze speaks is not to be interpreted as an essentially symbolic or material production, but as an ideal effect of the movement of language insofar as it is unequal to itself in the series of which it is composed. Language thus produces itself as a series through the events of impossible convergence that determine the signifying and signified orders as inverting sites of excess and lack that establish the sense of difference between word and thing. The exprimé and the level of expression, ideal effect of the proposition and its correlates, assert a primacy at the level of sense, determining language as an ideal occurrence, though an ideality without essence, for it returns neither to the signifying nor signified orders as its foundation: it resides at the surface. It remains to show how this understanding of the production of sense is related to Deleuze’s reinterpretation of the Platonic Idea developed in Différence et répétition. For the moment, we rest with the notion that the series of signifier and signified form a structure wherein they are differentiated by impossible points of convergence, what is in fact the power of the divergent. It is in the movement between these events that we are to understand the development of a history that is nothing other than the structure of the series itself: “the structure includes a register of ideal events, that is, an entire history internal to it” (LS 60). But what are we to do with this ideal history? Where are we in relation to historical determinations, those, for instance, of which Foucault’s énoncé is charged to disclose?


iii. Consequences

Between the énoncé and the exprimé, there forms a difference that we abstractly took into account at the beginning, where it was said that a certain sense of the historical and the metaphysical determinations of language, each turned in their own way against representational form, presided over the fundamental gesture and development of the writings we are following. The énoncé, as Foucault elaborates it, defines the position of an analysis that lodges itself between representational systems so as to draw out the discursive conditions necessary for the appearance of uniformity between them and cast off the idea of an internal principle of development shared by these systems that would integrate different types of empirical knowledge through fundamental ideas expressive of a society, most notably, the ideas of continuity and progress. However, the question of the synthesis of the disparate—given that subjectivity and its related psychological projections onto history are vacated of synthetic power—must fall to the discursive conditions themselves and the material realities which they both produce and by which they are in turn produced. Historical institutions present the necessary material support without which the organizing force of the énoncés specific to them would never be enacted. Inversely, such material structures could never be conceived without énoncés that organize their development. We are thus unable to escape the question of synthesis in Foucault’s account, insofar as it remains a question of how either a material or ideal form produces its correlative field of deployment. The synthetic possibility of their coproduction is ascribed to the anonymous machinations of power that furnish the motor of his historical reduction of representational systems, philosophy included. But as we before mentioned, the circularity of conditions between the material and ideal instances in the analysis of discourse and its énoncés is supposed to lead beyond an explanation produced by a prime mover installed ex machina in a series of terms. That is, one is led to a consideration of the simultaneity of material and ideal series outside of resemblance of conditioning and conditioned terms, that is, to a complex figure of historical movement and a radical questioning of historical actuality. As Deleuze rightly puts it in his study of Foucault: “the two formations are heterogeneous, even though they may overlap: there is no correspondence or isomorphism, no direct causality or symbolization” (F 31). Yet it is the question of the state of such an inherence that we have raised from the outset and to which we still are not able to give a sufficient response. We can, however, note a fundamental difference between what it would entail in relation to the conception of language that Deleuze provides in his logic of sense, which is largely unconcerned with the historical actuality of sense and subjectivity.

The exprimé: what is it? An ideal effect of the propositional form, or an excess in relation to a lack identified as a power of language that escapes representation in its basic function as the identification of the conditions under which a signified is related to a signifying term, that is to say, what comprises the entire field of knowledge claims and many of their accompanying philosophical explanations. Yet as we have already mentioned, this does not mean that Deleuze is not interested in giving an account of such a relation; it is simply that his account unfolds at a level completely heterogeneous to that normally carried out by metaphysics: the identity of words and things, rather than an ideal focus of convergence with respect to which the series of signifier and signified are mobilizedthe fundamental gesture of the Kantian tradition, there is substituted an identity of materiality and ideality vacated of all essentiality. It is nonsense that operates the donation of sense, the impossible determinations of which are the cathected events from which he understands the difference of signifier and signified to be derived and mobilized. Of course, he doesn’t carry out anything resembling a derivation in the strict sense, precisely because this would be impossible in the commonly understood sense of the term. The level of expression he locates cannot be reduced to the order of representation, but neither can it reduce representation to its conditions, for it is ultimately not something representable. It is a metaphysical construction that inverts the order of conditions: the paradoxical identity of the difference between word and thing is an effect of this difference that insists or subsists as cause, producing the convergences and differentiations between them. It is in this Deleuze understands the genetic power of sense, which we must read not as explanative of the conditions of representation, but as a determination of language that resolutely differs from such conditions, posing a problem beyond resolution of the largely theologically motivated summoning of classical metaphysics to ground the represented.

To the extent that the conception of the énoncé and exprimé can be read together it is in this: the representational function of language presupposes more than it can represent. Representation cannot represent its possibility outside of representational form, a tautology that expresses both the productivity of empirical knowledge and the limitation of a philosophy based on it. Between the historical reduction of representational form and the construction of a pathway that fulfills in a very different way the demands of metaphysics, there is the branching of a choice concerning the way by which to think otherwise the traditions that have produced our concepts and history. It remains to further accentuate the incompatibility of these directions while posing the question of their coexistence.

Gustave Courbet, "The Wave", (1870)

III. Idea and Épistémè

i. Pedagogy of the Idea

Sense is what is expressed in a proposition as its exprimé: neither the object represented through the act of designation by a subject with the self-identity necessary to manifest its truth, nor an ideal series of signifying propositions whose entailments develop the evidence of designations, the mode of description proper to scientific explanation. Sense is outside the conditions of identity by which the subjective and objective orders converge in representational figures whose formal and material contents find synthetic organization and extension. This is to say then, from the vantage of representation, that sense, when understood in this way, is unproductive or without the conditions of comprehension necessary to establish the differentiations undergone by conceptual identities in the field of empirical difference. Yet as we know, it is just this dimension of sense that is supposed to claim a ground that with respect to representational form is unrepresentable: an ungrounding of the conditions of possibility for a proposition to claim its truth value, which amounts to the sounding of conditions outside of propositional form presupposed by it. The exprimé and the theory of sense developed in Logic of Sense performs at the level of linguistic analysis what Deleuze’s reinterpretation of the Idea does at that of ontological inquiry in Difference and Repeptition. The two levels interpenetrate in each text, but the emphasis given respectively entrains a different perspective and vocabulary used to elucidate the problems. We shall now enter into how this unfolds.


· · ·


The pursuit of a reformulation of the conceptual dynamics of difference and repetition with respect to their formal ideality, posed as a preliminary requirement for the reformation of empirical concepts as used in various sciences, or at least their ontological reinterpretation, would appear to be a certain repetition of one of the longstanding ambitions of philosophical metaphysics, or “transcendental inquiry” in its sense determined by the tradition initiated by Kant. Indeed, the conceptual premises and terrain of analysis that this text covers would be unthinkable outside of this tradition. The prevailing ambition, however, or the “critical turn” it presents, is that of a repetition of many of the ideas and conditions of this tradition that constitutes a definitive difference from it—or within it. However, for as much as this text is vowed to a condemnation of dialectical form as an emanation of the power of abstract negativity by which traditional metaphysics is often the expression, Deleuze proceeds by what is most basic to that form: the reversal of perspectives with their resulting claim to the absolute. Repetition is to be understood as not that of the same but of the different; difference becomes not a specification of a concept but what remains outside of concepts; not solutions but problems become the object proper to questions. And then, there are the basic conceptual oppositions that structure the text: virtual and actual, singular and regular, convergence and divergence, conscious and unconscious, etc.

So in what constitutes the difference? If we follow the impetus and course of this work, it is to be found in the contention that the empirical determinations of beings afforded by the mechanism of representational form in no matter which domain of knowledge is rooted not in a formal possibility that anticipates these contents as the horizon by which an empirical world comes to be, but rather in what is called the “virtuality” of the Idea, a transcendental principle that comprehends various conditions, all of which turned against those of representation as the general mode of possibility of empirical givenness. The virtual presents the inversion of the category of the possible, the latter understood as comprising the conditions of identity necessary for empirical determinations to be made at all, a function whose elaboration is understood as the province of nearly the entire philosophical tradition whose conceptual failures are a result precisely of this vocation. The challenge that Deleuze forms to this dynamic amounts to the hypostatization of a transcendental dimension inaccessible to representation—thus, one outside of conditioning possibility in the form of subjective and objective unity—which provides the real ground of its method, a real that is necessarily unconscious because it exceeds the formal identity required to be made an object of reflection. Thus,

“[t]he true opposition lies elsewhere: between Idea (structure-event-sense) and representation. With representation, concepts are like possibilities, but the subject of representation still determines the object as really conforming to the concept, as an essence. That is why representation as a whole is the element of knowledge which is realised by the recollection of the thought object and its recognition by a thinking subject. The Idea makes a virtue of quite different characteristics. The virtuality of the Idea has nothing to do with possibility. Multiplicity tolerates no dependence on the identical in the subject or in the object” (DR 191).

The question is how such an Idea claims distinction in this state of obscurity, how empirical states are related to them, and how representational concepts are recast in their light. In its basic premises and exposition, the theory of the Idea Deleuze puts forward fulfills in many ways the requirements stipulated by Kant—that of progressive determination, transcendental excess, differentiation of the object and subject of cognition—yet does so in a way that subtracts what he views as their epistemological and ultimately moral limitations: that of commonality or concordance among our subjective faculties in their specification of identity, which ultimately means the progressive cancellation of difference through its absorption in conceptual form in view of the equilibration of subject and object. And so, rather than progressive concordance, the conception of the Idea Deleuze creates introduces a progressive discord that stipulates among other things the impossibility of a unified world, that is, one in which a formal identity would determine the mode by which concept and object align:

“There is indeed a serial connection between the faculties and an order in that series. But neither the order nor the series implies any collaboration with regard to the form of a supposed same object or to a subjective unity in the nature of an ‘I think’. […] The transcendental operation of the faculties is a properly paradoxical operation, opposed to their exercise under the rule of a common sense” (DR 145-146).

In this way, the Idea (and here, for Kant as well) is the immanent mode by which transcendence is given as what cannot be thought yet must be thought in some way if there are to be concepts. However, Deleuze insists that this state of excess with respect to the determinate concept has nothing to do with its regulation, but rather with a certain force that leads the finite concept beyond its representational coordinates in the realm of the faculty in which it is exercised, whether it be that of knowledge, moral, or aesthetic experience:

“We saw how the discord between the faculties, which followed from the exclusive character of the transcendent object apprehended by each, nevertheless implied a harmony such that each transmits its violence to the other by powder fuse, but precisely a ‘discordant harmony’ which excludes the forms of identity, convergence and collaboration which define a common sense” (DR 193).

In this discordant accord between the faculties of transcendental experience, there is to be read a characteristic that we have already approached in Logique du sens, one which forms in many ways the architectonic of Deleuze’s thought: that of serial differences based on the movement of a displaced term that brings them into contact. As we saw before, it was that of nonsensical figures, of paradox, that imbues the structures of signifier and signified with their respective movement, the exprimé being just this entity that interiorizes the difference between word and thing, conjugating the series as perpetual and respective excess and lack. We have now another formulation of this structure, where the linguistic notion of the exprimé becomes an ontological one that in many ways expresses the same movement. The virtual (or partial) object, derived from Kant’s metaphysics and from psychoanalytic theory, is that which will be used to articulate the repetitions and differences between series of terms.

We find such an object where a convergence of series, whether they be comprised of empirical events, psychological fantasies, or the relations between them, produces not a causal figure used to either retrospectively or prospectively explain the movement between the terms based on origin or finality, but locates rather the constitutive lack or excess of each, the “object” being their mutual displacement that produces what is lacking or excessive. Thus, “[t]he virtual object is a partial object – not simply because it lacks a part which remains in the real, but in itself and for itself because it is cleaved or doubled into two virtual parts, one of which is always missing from the other” (DR 100). What is “virtual”, then, is not a state that precedes and transcends the movement of a being’s actualization as a set of possibilities its development could or must take, but rather that which comprises its actualizations as a set of virtual figures whose series present different degrees of realization with respect to each other. There is thus no “set of possibilities” for existence, but only degrees of realization of the Idea. Series of terms are thus not deployed in or comprehended by a common image, but seized only as the discordances developed between them. As it was for Bergson, so it is for Deleuze: the actual and its states of affairs, qualities and quantities attributable to the real, are forms by which the virtual expresses itself, a virtuality that is of a past that can never be made present, yet in which all actual time and states find their ground. The virtual object is what brings series into communication as expressions of this ground:

“The virtual object is never past in relation to a new present, any more than it is past in relation to a present which it was. It is past as the contemporary of the present which it is, in a frozen present; as though lacking on the one hand the part which, on the other hand, it is at the same time; as though displaced while still in place” (DR 102).

The virtual object, or exprimé when taken at the level of the proposition, is that which develops sense as the transcendence immanent to its unfolding, as the dimension that fragments the objective unity of cognition into a disparity between series: “We call this state of infinitely doubled difference which resonates to infinity disparity. Disparity – in other words, difference or intensity (difference of intensity) – is the sufficient reason of all phenomena, the condition of that which appears” (DR 222). At the level of the proposition, it is what doubles designation in the dimension of expression, producing paradoxical figures that interiorize the difference between word and thing. The exprimé, neither word nor thing, is the identity that differentiates them, “the event occurring in a state of affairs and the sense inhering in the proposition are the same entity” (LS 209). But for both the virtual object and the exprimé, “sense is like the Idea which is developed in the sub-representative determinations” (DR 155). What are these “subrepresentative” determinations of the Idea?

If representation is what brings to bear the conditions under which a word can designate a thing, an idea a causal relation between beings, a subject a world within which representations coexist, then that which is subrepresentative in the Idea as it is understood here would be that which precedes these determinations as a certain condition of their appearance without reducing to the image of what is thereby conditioned. This is, of course, the conceptual refrain that carries Deleuze’s entire approach. What it amounts to is a hypostatization of a transcendental principle (the Idea) whose virtual combinations produce the realm of empirical configurations circumscribable by the process of representation: “The transcendental principle does not govern any domain but gives the domain to be governed to a given empirical principle; it accounts for the subjection of a domain to a principle. […] It is the transcendental principle which maintains itself in itself, beyond the reach of the empirical principle” (DR 241). The difference between the actual and the virtual is terminologically noted in accordance with the opposition différenciation/différentiation, or what expresses the same, the order of the solution and the order of the problem. The subrepresentative conditions of the Idea are fundamentally tied to a sense of the problematic as such, that whose progressive determination fulfills a pedagogy beyond the determination of solutions. Deleuze’s theory of Ideas presents with the most universalizing reach the bifurcation essential to his thinking of the metaphysical grounds of representation as a method of explanation at work in empirical inquiry: the actual and the virtual are to be resolutely distinguished, the resemblance between them destroyed. The virtual is not the formal possibility of the actual, but rather the reality of its differences which can only be approached as what virtually subsist within any finite representation as its subrepresentational ground. And so,

“[w]hereas differentiation determines the virtual content of the Idea as problem, differenciation expresses the actualisation of this virtual and the constitution of solutions (by local integrations). […] Every object is double without it being the case that the two halves resemble one another, one being a virtual image and the other an actual image. They are unequal odd halves” (DR 209-210).

The passage from the virtual to the actual as an integration is to be understood as that by which a purely logical event—a singularity, or the problem as such—condenses around it cases of solution that, as expressions of quality or quantity within the real, are both the actualization of a problem and its eternal imparity: solutions do not actualize a problem without thereby being differentiated from it. The problem, or the obscurity from which clarity issues, absconds to another level, determining other integrations that will never become equal, for as a virtuality, it is of a dimension that can never be made present. It is in this way that the problem as such and the category of intensive magnitude that Deleuze develops are aligned, if not the same thing. Intensity determines what can be seized empirically as qualified or quantified without itself being able to be seized. Taken together, these various aspects form what Deleuze calls a pedagogy of sense which, in view of what it develops, we may call the pedagogy of the Idea:

“The point of sensory distortion is often to grasp intensity independently of extensity or prior to the qualities in which it is developed. A pedagogy of the senses, which forms an integral part of ‘transcendentalism’, is directed towards this aim” (DR 237).

But given that a fundamental dehiscence of the actual and the virtual can be traced, why should it be thought of as a transcendental principle? As with the attempt to derive the orders of designation, manifestation, and signification of the proposition from that of expression, is it not likewise here the case that, in the attempt to attain a sufficient reason for the development of phenomena on the basis of an Idea of differential multiplicity severed from the coordinates typical of the metaphysical tradition—identity of the concept, opposition of predicates, analogy in judgment, resemblance in the object—one cannot but remain in a certain abstraction by which a theory is charged with an explanative power it can never obtain? Deleuze is able to show how the kind of virtuality he develops philosophically can be seen at work in certain mathematical and empirical domains, but it does not follow from this that such a conceptual rendering of these aspects should be taken as a transcendental basis for that out of which it is developed. In other words, his critique of foundations formed in the image of what they found, “[t]he shortcoming of the ground is to remain relative to what it grounds, to borrow the characteristics of what it grounds, and to be proved by these” (DR 88), seems no less applicable to his theory of the Idea.  This is why the Idea of difference as it is put forward in Différence et répétition is best understood as a philosophical postulate whose ultimate value resides not in the promulgation of a doctrine nebulously linked to certain empirical domains as their transcendental foundation, but in the incitation it develops within the philosophical enterprise susceptible to extension beyond philosophy: a pedagogy of sense that leads beyond the finality of essence, which is to say, beyond the metaphysics of a kingdom of ends. The question that opens at the margins of this text is whether such a beyond has any need of the type of “transcendentalism” Deleuze ascribes to it. For our purposes, we may simply ask: Is it enough to think representational form as the degradation of an ideal state of difference? Is there another way to elucidate the problem of representation itself? For such an elucidation, we return to the work of Foucault.

Gustave Courbet, "The Source of the Loue", (1864)

ii. Forms of Dispersion

As an expression of the fundamental intuition of Foucault’s archaeological analysis—that which situates neither expression nor intuition in the space it traverses—, the movement of the historical planes conjoined between the pages of The Order of Things serves not to demonstrate a thesis, such as that of the disparition of the representative function at the end of the 18th century and with it the specification of the human being as a finite form in the process of disappearance, nor to propose an alternative by which language and things are to be brought into relation. Rather, we are presented an accumulation of texts produced during a certain period of time in a certain culture that are used to show a series of changes that occurred in how language is brought into relation with that of which it speaks within the domains of knowledge constitutive of modern biology, economics, and linguistics, which together form the resources, epistemological and empirical, for the formation of the human sciences.

The extreme expansiveness of the field of investigation is tempered by a relatively specific, ultimately philosophical concern whose elaboration cannot be separated from its historical exposition: How is it that the difference between word and thing is produced by the discursive formations in question? This will be to ask: How does the representative function change in relation to the enunciative possibilities that define the modalities of knowledge proper to a specific historical strata? Far from a thesis concerning the historical supersession of representation or an injunction for its surpassing, we find rather the conditions of its complexifying figures in the discourses under consideration, the elicitation of the human to which they are attached forming the touchstone for Foucault’s reading of the philosophical response to the question of the foundations of knowledge thereby produced, the “human sciences” taken as an entity forming a part of this response. In what follows, we will chart the significant points of his analyses of the representative function of linguistic form in the changes it undergoes between the classical and modern period, leaving aside the concomitant analyses of biological and economic theory.


· · ·


For Foucault, the threshold of the Western classical age is found where language breaks its tie with the form of “commentary”—that which proceeds by a conception of a fundamental resemblance between the beings of perception that words are given the task to decrypt—and establishes the statute of “discourse”, a term which here will be used to tie together various theories of this period on the basis of a shared reevaluation of the epistemological and ontological coordinates of representation. Classical discourse is to be understood not as a certain form of exposition, nor even a more complex form of ordering phenomena, but as a system that interiorizes the possibility of order and whose collection of signs forms the genetic possibility of rational progression. Representation comes to be understood as that which, in representing something through the sign, makes fundamental reference to its own power of representation, and thereby, the inclusion of each representative act in a system of signs that determines its movement:

“nothing is given that is not given to representation; but, by that very fact, no sign ever appears, no word is spoken, no proposition is ever directed at any content except by the action of a representation that stands back from itself, that duplicates and reflects itself in another representation that is its equivalent” (OT 86).

The act of designation or signification, then, is not something that is given through the differences and resemblances expressed by phenomena of which linguistic elements are the transcriptions, but that which elaborates the progression of investigation proper to the artifice of signs, which are rather the representations of the natural enchainment of ideas. It is in this way that discourse is not simply a form among others, but the method of a mathesis, a universal language: “the possibility of defining the natural and necessary progress of the mind from the simplest representations to the most refined analyses or the most complex combinations: this  discourse is knowledge arranged in accordance with the unique order laid down for it by its origin” (OT 93). In brief, the sense of ontological univocity typical to Renaissance thought wherein the sense of word and thing interminably echoes in the depth of resemblances is brought to a close:

“And that uniform layer, in which the seen and the read, the visible and the expressible, were endlessly interwoven, vanished too. Things and words were to be separated from one another. The eye was thenceforth destined to see and only to see, the ear to hear and only to hear. Discourse was still to have the task of speaking that which is, but it was no longer to be anything more than what it said” (OT 47-48).

The power of discourse, as it is understood in the classical age according to Foucault, is thus to be found in a synthetic production of sense whose epistemic grounds are almost entirely attributable to the signs that deploy its function. It is of course obvious the type of being whose existence must occupy the ontological foundation of this production: this subjective being cannot be understood as simply a form of being among others, but as the commencement of the possibility of attribution given through the verbal form. The verb and the subjective being coalesce as a constituted part of language while simultaneously constituting its possibility of deployment, forming “[a]n invisible canvas, entirely overlaid by the brightness and design of the words, but one that provides language with the site on which to display its painting” (OT 105). For its part, the paint that fills the canvas of representation has an origin not entirely reducible to the pure form of manifestation guaranteed by the verb and the subject who occupies it: the thickness of things, their relative opacity in face of the invisible structures that orient them, is to be accounted for in such a way that comprehends the transition between the pure givenness of a material reality and the structures of designation and signification that invest the power of naming proper to the noun. It is in the discovery of subrepresentational elements of language—the mechanisms that produce the inflections, syllables, and sounds that compose the elements proper to the representational totalities of words and phrases—that account could be taken of the movement from the primal revelations of things to their consolidation in formal elements allowing for the diversity and expansion of representation. The subrepresentational is thus, for the classical grammar, the deposit of material which has no other destination than its formalization by developed orders of representation under the aegis of the subject who first grants the possibility of attribution:

“Throughout its density, even down to the most archaic of those sounds that first rescued it from its state as pure cry, language preserves its representative function; in each one of its articulations, from the depths of time, it has always named. It is nothing in itself but an immense rustling of denominations that are overlying one another, contracting into one another, hiding one another, and yet preserving themselves in existence in order to permit the analysis or the composition of the most complex representations” (OT 113-114).

The pure verbal function thus has as its correlate a conception of the nominal form that is to account for the way by which subrepresentational structures at once translate the primal language of things and transform it into figures that bear no resemblance to their origin yet guarantee an independent and internal development of the representational function of signs as those which represent the progression of ideas. The various pathways that Foucault treads in his exposition of the épistémè of the classical age with respect to its conception of language is thus done with the intention to show its reducibility to the enunciative possibility by which it produces itself, that is, the symmetrical elaboration of the subject who manifests the possibility of language in the act of attributing being through the pronouncements of signs, and in turn, the objects of the signs whose differentiations are translated by the artificial structures that compose the possibility of their representation by the subject as a form of knowledge. The sovereignty of naming, however, brings to the fore the instability of this enunciative regime, insofar as the identity of word and thing acts as a limit towards which it proceeds but which it must constantly efface so as to maintain itself as a discourse constructed in its autonomy and nonresemblance to that of which it speaks:

“to speak or to write is not to say things or to express oneself, it is not a matter of playing with language, it is to make one’s way towards the sovereign act of nomination, to move, through language, towards the place where things and words are conjoined in their common essence, and which makes it possible to give them a name. But once that name has been spoken, all the language that has led up to it, or that has been crossed in order to reach it, is reabsorbed into it and disappears” (OT 129).

This is to say, the classical theories of language are still indexed by a certain notion of resemblance that will be transfigured only with a change in the épistémè, that which can be understood as a global rule of formation that marks the historical a priori of an epoch.

With his analysis of the theories of language produced in the classical age, Foucault marks the elements of its representational schemes that bind it to inherited conceptions of the origin and ontological reality of language while simultaneously preparing the conditions for its transformation. The inventory and analysis of inflections, syllables, sounds, etc., or what could be called the subrepresentational elements of language that determine the sense of nouns and verbs in their role as the representations of the movement of thought, open the space of a certain mechanics of language whose proper representational value can only be imposed through the appeal to the sovereignty of naming—even if this is just the transcription of a primal cry—, and to the verbal form itself, which ought to bear the manifestation of the presence of a subject. For the classical grammar,

“[e]very language, however complex, was situated in the opening that had been created, once and for all, by archaic cries. Lateral resemblances with other languages […] were noted and listed only in order to confirm the vertical relation of each to these deeply buried, silted over, almost mute values” (OT 253).

It is thus this laterality or horizontality of forms severed from the largely imagined vertical injunction imposed on the material elements of language that establish what is called the “pure grammar” of the modern age, or what is generally called linguistic science.

Thus, the discoveries of the modern analyses of language Foucault will show to condense around a transformation in the conception of the origin of the representative function. Such a function comes to be understood as effectuated on the basis of something that does not represent the representative power: the representative function is no longer the representation of the enchainment of ideas. This amounts to an understanding of a historical dimension to language that is as if the inverse to the history of its ideational representations with their originary limit of primal articulations:

“But from now on there is an interior ‘mechanism’ in languages which determines not only each one’s individuality but also its resemblances to the others: it is this mechanism, the bearer of identity and difference, the sign of adjacency, the mark of kinship, that is now to become the basis for history. By its means, historicity will be able to introduce itself into the density of the spoken word itself” (OT 256-257).

Such a historicity is defined by the tracing of formal transformations interior to the developed structures of language rather than by regressing along the history of ideational forms to the point where the vocalic or written element would transcribe the basic sense of the thing in its immediacy. The canvas of representation constructed in the classical age as the ordering surface of contact between words and things, even if cleansed of certain notions of resemblance characteristic of pre-classical thought, nevertheless presupposes a point of contact: it is in this that Foucault situates its archaeological filiation with what preceded it. What is then particular to the modern épisémè is the fundamental difference, or absolute non-resemblance, between word and thing that forms the historical a priori of knowledge in which new positive sciences as well as both positivist and transcendental philosophy find their ground.

And so, for linguistics to develop as a science, analysis sought to remain in the historicity proper to the transformations of the sub-representational elements of languages, seeking neither the archaic elements within a given language that would translate the language of things, nor the supposed original forms shared by all languages: “it is enough to study the modifications of the radical, the system of inflections, the series of variable terminations. […] [I]t is therefore not possible to relate a language to the form or the principles that render all other languages possible; they must be grouped according to their formal proximity” (OT 317). A radical form within a given language thus does not represent an origin, but simply a modification relative to the radical forms of other linguistic systems. The internal, or formal, historicity of languages is then nothing other than the heterogeneity of forms at each level of linguistic history and the discontinuity between the development of signs and the development of representations produced through their combinations:

“the heterogeneity of the various grammatical systems emerged with its peculiar patternings, the laws prescribing change within each one, and the paths fixing possible lines of development. […] Here, as elsewhere, the arrangements into chronological series had to be broken up, and their elements redistributed, then a new history was constituted” (OT 319).

Finally, the pure verbal form as expressive of a subject who attributes being to things through the intermediary of representations can no longer be sustained: “The ontological transition provided by the verb to be between speaking and thinking is removed; whereupon language acquires a being proper to itself. And it is this being that contains the laws that govern it” (OT 322).

The being proper to language as it arises in modernity is what opens the space for the development of a positive science of linguistic forms which, along with the development of other domains, leads to a reformulation of the epistemological and ontological status of the human being that Foucault will spend the remainder of the text circumscribing. His claim is simply that the human being as forming a finite domain of empirical knowledge did not exist in the classical age or in previous ages because the epistemological resemblance between knowledge of self and knowledge of nature made the human in its concrete finitude unthinkable. For the classical age, “[i]f human nature is interwoven with nature, it is by the mechanisms of knowledge and by their functioning; or rather, in the general arrangement of the Classical episteme, nature, human nature, and their relations, are definite and predictable functional moments” (OT 338). But for modernity, the human, rather than that which occupies the place of attribution whose canvas of representation serves as an ordering medium and surface of contact between words and things, becomes that which obscurely precedes all determinations of knowledge, not as the conduit of universal reason, but as an empirically limited, though irreducible, field of phenomena: the human, “as soon as he thinks, merely unveils himself to his own eyes in the form of a being who is already, in a necessarily subjacent density, in an irreducible anteriority, a living being, an instrument of production, a vehicle for words which exist before him” (OT 341). It is thus a transformation in the conception of limits that is at issue insofar as human finitude is placed at the basis of a knowledge that becomes radically delimited by objects which autonomously preside over the formation of representations. The concomitant to the being proper to language is the being proper to things, or the revelation of an unmasterable exteriority:

“things, in their fundamental truth, have now escaped from the space of the table; instead of being no more than the constancy that distributes their representations always in accordance with the same forms, they turn in upon themselves, posit their own volumes, and define for themselves an internal space which, to our representation, is on the exterior” (OT 259).

The conditions of knowledge in modernity are lodged irremediably outside of the general form of subjective reflexivity as the relay or point of contact between the interior and exterior world. One only comes to know to the extent one traverses an exteriority radically separate from the representational system used to account for it.

Foucault will thus say that the cogito in modernity is defined not by the thinking of what is thought, but by thinking the limit between what is thought and cannot be thought. The question such a cogito poses is “[h]ow can man think what he does not think, inhabit as though by a mute occupation something that eludes him, animate with a kind of frozen movement that figure of himself that takes the form of a stubborn exteriority?” (OT 352). Such an unthought or unconscious term is not an imaginary structure, but the essence of an empirical positivity that the human cannot ideally delimit, which Foucault will ultimately identify as the conditions of its conceptual death. What is unthought or unconscious is here not a primordial region that surfaces at the extremity and in spite of the determinations of positive knowledge. Rather, the unconscious, in its modern specificity, is the reverse side of such determinations: a superseding exteriority that contaminates the human essence. The source of positive science is also what exposes such an essence as fundamentally separated from itself, a separation whose limit is infinite differentiation, or the dispersion of essence.

Gustave Courbet, "Grotto of Sarrazine near Nans-sous-Sainte-Anne", (ca.1864)

The modern cogito is thus an inversion of the relations established by the representative order between the exteriority of things and the interiority of words such that words assume an exteriority that is precisely the internal structure of development proper to their material history, and things, for their part, assume an interiority that defines an exteriority beyond all possible unification with words in the representative function. In other terms, this situates an inversion of the sense of the empirical and transcendental: the representative power is not an ideal surface that defines the order of the empirical depth of things, but rather is of this depth an effect whose internal laws are nothing but the historically variable differentiations of an aleatory ensemble of material. Signification is reduced to the series of transmutations the material body of the signifier undergoes, the ideality of structure nothing more than an effect of these changes. The transcendental concomitant to this is the determination of things as preceding and exceeding the interiority of language such that the human being becomes understood in a sense that is strictly speaking finite, that is, without recourse to transcendental conditions of representation: “everything that had functioned within the dimension of the relation between things (as they are represented) and words (with their representative value) has now been drawn back into language and given the task of providing it with an internal legality,” this interiority being the exterior, the outside: “in which man appears as a finite, determined being, trapped in the density of what he does not think, and subject, in his very being, to the dispersion of time” (OT 368).

But it is precisely between the human finitude that the unconscious registers as the positive ground of the empirical sciences and the dispersion of time to which such a being is given over that together form the possibility for what Foucault will treat en masse as “the human sciences”, that constellation of practices within which his own work is ambiguously lodged. After the destruction of the surface of representation proper to the classical age through the modes of knowledge developed in modernity, we find that, for the deployment of concepts proper to the human sciences, the representative function returns as an essential problem insofar as the ideas and practices of human beings in their finite and, with respect to the conditions of positive knowledge, derivative forms, pose the problem of the totality within which they are articulated. The canvas of the classical age through which particular representations could be hierarchically situated becomes for the human sciences the system of largely unconscious structures through which particular significations can be interpreted. Representation thus returns as a productive function, but one which, rather than claiming the ground of knowledge claims, is understood as an irreducible ensemble of effects issuing from the empirical dimensions discovered by modern sciences. It is thus for the human sciences that representation is not what assigns an order of existence to its logical place in a hierarchy of knowledge as was attempted by the mathesis of the classical age, but that which coordinates forms of human existence—interactions with words or things materially independent of such an existence—with their unconscious conditions of deployment. For the human sciences, the unconscious is thus invested with the power of the representative function that was for the classical age the power of consciousness: representation is understood as a realm prior to empirical knowledge that defines sets of practices subjacent to rational inquiry. Language can thus be understood, while not in the least denying the findings of linguistics, as nevertheless an attempt to say the sense of a thing independent of language, an attempt which implies certain forms of human dispositions and practices. These attempts can thus be taken as significations produced within a system that remains largely unconscious:

“The role of the concept of signification is, in fact, to show how something like a language, even if it is not in the form of explicit discourse, and even if it has not been deployed for a consciousness, can in general be given to representation; the role of the complementary concept of system is to show how signification is never primary and contemporaneous with itself, but always secondary and as it were derived in relation to a system that precedes it […] in relation to the consciousness of a signification, the system is indeed always unconscious since it was there before the signification” (OT 394).

Between the unconscious expressive of the modern épistémè and that uncovered by the human sciences, Foucault traces the continuity of an enunciative possibility that is not without a difference, one which concerns the possibility of its repetition. For if it is true that the reverse side of the empirical sciences produced in modernity is the unconscious as unrepresentable exteriority that forms the dispersion of time within which the human being is placed and through which it interminably repeats its determinations, the unconscious discovered through the discourses of the human sciences would locate a representative power proper to language and its attendant practices that condition the repetition of its significations. It is the status of this representative power that marks the ambiguity of the origin of such discourses and their mode of relation to the sciences of modernity. While Foucault does not provide a derivation in the strict sense of their forms from those of the empirical sciences, it is clear what he understands to be the archaeological ground they share: a historicity internal to both object and subject that defines their mutual exteriority and necessary dispersion:

“[History] takes place on the outer limits of the object and subject; it designates the erosion to which both are subjected, the dispersion that creates a hiatus between them, wrenching them loose from a calm, rooted, and definitive positivity. […] History shows that everything that has been thought will be thought again by a thought that does not yet exist” (OT 406).


iii. Images of Thought

As was asked before concerning the coexistence or possible codeployment of the enoncé and exprimé developed in the texts we have traversed, the same question may be extended to the conceptual space within which the ontological determinations of the Idea as Deleuze conceives it and Foucault’s historical analysis of the épistémè find the possibility of mutual expression. This is to ask whether one can think coextensively the accounts they provide of the conditions of knowledge claims irreducible to a structure of representation. If not, it would be the case that we are given accounts that reside at levels whose possibility of coexistence would remain to be determined.

When we characterize Deleuze’s approach as being that of “metaphysical construction”, what does this mean? By this we are to understand that the movement of critiqueas it is ultimately from the register of the critical tradition descending from Kant that his work is inseparableis generated, as it was for Kant, by a form of self-movement that constructs the dynamic by which series of empirical events become situated as experience and find a basis of intelligibility. That the possibility of this transcendental basis is fundamentally tied to a reality exterior to the subject is what the Idea is meant to show. Yet for Kant, the Idea is functionally a form of mediation between conditions of an inner world and those of an outer world, whereas for Deleuze there is never any such separation: the Idea does not mediate but constructs the conditions under which series of empirical terms are deployed, whether they be “inner” or “outer”. In his vocabulary, Ideas comprise the “singular points”, or metaphysical nodes, that determine the formations of “ordinary points”, or all within the actual that is expressive of qualities determinable in perceptual resemblance and propositional opposition and of quantities involved in numerical identification and the form of analogical judgment. Together, these determinations form what he describes as the “fourfold illusion” of representationthe illusion residing in the interpretation they are given of ontological primacywhose promulgation in the philosophical tradition Deleuze’s reconstruction of the Idea and his theory of sense gathered around the function of the exprimé are to displace.

As we have suggested yet not fully drawn out, the critique of representation developed through the ontological questions Deleuze poses and the figures of explanation given in the theory of the Idea should be read as following from a practicalin Kant’s senseintervention that can be understood as a pedagogy, insofar as this no longer means a method by which something is taught by a master and assimilated by the pupil, but rather the mode by which the repartitions of difference and repetition is undergone, which fundamentally involves a transformation in the sense of temporality and the form of subjectivity that arises out of it.

If repetition is to become a power of concepts rather than remain in the element of exteriority within which a conceptual identity is repeated (bare or material repetition), there is necessary an internalization of difference by which repetition becomes repetition of the different rather than of the same. But how is this concretely effectuated? Even if the type of difference and its repetition Deleuze describes is understood as what essentially belongs to the realbare repetition being a form of representation resulting from a logical or natural blockage of the concept—, it is nonetheless certain that it must be conceptually produced, or in other words, that it is not of the order of phenomena discoverable through a method, but is itself a method whose primary object is itself insofar as it concerns a reorientation of temporal order out of which the modes of discovery as practiced in empirical domains may be resituated or at least conceptually reconsidered.1placeholder Deleuze’s pedagogy of the Idea is foremost a practical method in that it irreducibly involves a dimension of reflexivity tied to a certain supra-conceptual exigency that surpasses the limits of what can be rendered representationally. Yet unlike the in principle separate realms of pure theoretical and practical reason in Kant’s work, it is the exigency of the practical that not only subtends but determines the realm of the pure metaphysics Deleuze creates in such a way that they belong to the same form of thinking. The imperative to affirm the irreducibility of difference and all that expresses the divergent is thus not the rhetorical dimension of a conceptually separable theory, but the foundational imperative that is identical to the theory itself, which is revealed in the final instance to be inseparable from the rhetorical order of a demand that invests the basic sense of what it performs. Thus, in the imperative that

“[e]ach term of a series, being already a difference, must be put into a variable relation with other terms, thereby constituting other series devoid of centre and convergence. Divergence and decentring must be affirmed in the series itself. Every object, every thing, must see its own identity swallowed up in difference, each being no more than a difference between differences. Difference must be shown differing” (DR 56).

There is no purely rational reason why this ought to be done, for the concern of falling into contradiction if not followed only applies to the sphere of representation. There is thus a supra-conceptual exigency at work that has not to do with reason’s correctness or correlational validity, but with a certain potentialization of a capacity to attain the limit of what concepts can do, a limit which marks the event of exposure between the faculties of discordant reason. The boundaries of a pure metaphysics are thus never distinct in Deleuze’s work and structurally cannot be. It is in this way that the type of truth demanded by representational thoughtwhether this be the closure of a metaphysical system or the indefinite progress of empirical investigationis disbanded in the movement by which pure philosophy, ethical reason, as well as aesthetic experience communicate to each other their forms of evidence through the exigency of difference.

As it was for Kant, necessity is not an external constraint but the imperative of a method; yet here, the imperative is inseparable from a reversal by which the opposition of internality and externality, or subject and object, is suppressed by the monistic nature of a differential multiplicity from which the opposition is taken as an illusory effect. That the imperative of rhetoric and the apodicticity of science converge around figures of the expropriation of consciousness is the inversion essential to Deleuze’s Copernican mask. There is thus understood irony and intended humor in the fact that this elevates his form of address outside of a determination of truth and falsity. The pedagogy is this movement of inversion by which the developed orders of subjectivity discover the real through creating the conditions of subjective dispersion, a creation which is just as much a destruction of itself considered as illusory effect. For this pedagogy to be carried through, it is not surprising that a certain pre-critical naivete is required insofar as the injunction toward subjective erasure demands a lack of discrimination or even identification of the orders of being proper to the object and those of the questioning subject. It is in this way that things are taken as signs whose objective determinations are the expressions of the propositional structure of a problem issuing from a question tied to the displacements of the virtual or partial object. At the level of linguistic structure, the exprimé is what is expressive of the unity of this difference, registered as paradox. At the ontological level, it is the sign (or simulacra) that comprises the “singular points” of the Idea which are themselves the objective nature of the problem developing itself in the displacement of the question. Thus, as we have it,

“[t]he problems ‘correspond’ to the reciprocal disguise of the terms and relations which constitute the reality series. The questions or sources of problems correspond to the displacement of the virtual object which causes the series to develop” (DR 106-107).

The sign thus develops in the depths what the exprimé develops at the surface: a time of unlimited becoming where essence is nothing other than the displacement of the identical by a difference more profound or superficial than the most minimally representable depth or surface of things. It is this vanishing state that is the ultimate term of the pedagogy: a teleology of dismemberment wherein each subjective faculty attains the destruction inherent to the Idea it develops. In other words, one quits the sphere of resemblances, oppositions, recognitions, and analogies proper to the world of representation and enters the fatality of an immanence that a thought without image demands.

And so, returning to our question concerning the coexistence or codeployment of the conceptual space of the Idea and that of épistémè, what can be said? For the historical transformations that the épistémè registers, does it not imply a very different image of thought than that of the thought without image? When we speak of the historical reduction proper to Foucault’s approach, what does this mean? By this it is certainly not sufficient to understand a method characterized as “historicist”, that is, what simply reduces to the interactions between series of historical events the movement that animates concepts. Rather, the nature of his historical reduction cannot be thought apart from the type of originality and productivity by which pure phenomenology establishes its transcendental-synthetic order, even if Foucault’s envisions a very different origin and space of deployment. Indeed, as we before mentioned, the notion of synthesis is entirely displaced within his conception of the historical determination of ideas and nearly banished from his theoretical vocabulary to the point that its absence becomes a structuring negativity of his texts, a lack that cannot be filled but which is effectively filled by a given enunciative regime or discursive formation: the productivity of the énoncé is inseparable from a lack of determination that is the condition of its coordination of heterogeneous domains of existence. It is by no means a question of its synthetic production of these domains and the beings that occupy them, but rather its effective subordination of a multiplicity to a unity of production that is called “discourse” (in the sense determined by archaeological method). As Foucault says in reference to that of his own,

“my discourse, far from determining the locus in which it speaks, is avoiding the ground on which it could find support. It is a discourse about discourses: but it is not trying to find in them a hidden law, a concealed origin that it only remains to free; nor is it trying to establish by itself, taking itself as a starting-point, the general theory of which they would be the concrete model” (AK 226).

That this nonetheless implies a certain notion of synthesis we have already raised as a problem, one that is not dispelled by invoking the productivity of power. Yet for such a research, it is less a matter of identifying agents of production, individual or collective, than determining the points of intersection or collaboration between developed forms. In this, his approach is more analytic than it is synthetic. Foucault searches for conditions that for representational knowledge must be considered inessential insofar as they do not directly circumscribe the synthetic conditions of the emergence of representations but rather indirectly illustrate the historical complicities necessary for the emergence of a discursive domain in its apparent unity. That this approach be considered antirational is thus both true and false: true insofar as Foucault’s work is not concerned with the internal development of representations, and false insofar as the possibility of knowledge is its sole concern.

“Reduction” must therefore be seized in its productive aspect: as a function of unification that neither has synthetic essence in the consciousness of those involved in the formation of discourse, nor in the institutions that materially embody a discursive regime, but as what forms the frontier between discourse and its material support as the space of their interpenetration. Reduction in this sense is what the énoncé exposes. Even though he does not grasp the significance of the category of unity in Foucault’s rendering of the énoncé, Deleuze nevertheless rightly characterizes the form of relation its frontier establishes as neither a vertical form of parallel expression through symbolic transfer, nor a horizontal form of unidirectional causal determination, but as what he calls a “diagonal” relation, that is,

“discursive relations become associated with non-discursive milieux, which are not in themselves situated either inside or outside the group of statements but form the above-mentioned limit, the specific horizon without which these objects could neither appear nor be assigned a place in the statement itself” (F 10).

That’s what we spoke of earlier as the historicity of knowledge wherein its material and ideal determinations form a series of causal influence that determines historical immanence in the absence of origin.

Gustave Courbet, "The Beach", (1865)

However, even if such a method of reduction displaces the metaphorical hermeneutics of vertical relation as well as determination in the final instance of horizontal causality in favor of the involutions of heterogenous series within enunciative regimes, it is nevertheless the case that a certain form of origin is indispensable to Foucault’s historical reduction, even if it be nothing more than the threshold of a transition. It is precisely this form of relative origin that the épistémè registers. The épistémè does not describe the conditions of the unity of discursive regimes within a given historical period as does a history of ideas or a philosophy of history, but rather the series of transformations necessary for a given domain or number of domains to be considered unified as a science. The unity that the form of science imposes is thus never exposed in its representational positivity as a history of science would demand, but is instead located at the margins of its development in the transformations of epistemic possibility determinative of the epochal indexing of knowledge claims, which ultimately implies nonscientific determinations brought to bear on those of science. The conventional epochs that the épistémè identifiesRenaissance, Classical, Modernare descriptively necessary yet misleading insofar as they are understood as expressing within a unified historical time the essence of ideational structures whose truth or falsity is determinable relative to an absolute measure by which the progression of forms becomes comprehensible and not, rather, as historically relative discursive conditions that determine the threshold of truth.2placeholder

But how are we to understand the reduction performed in the elucidation of the transformations undergone by the representational form in The Order of Things? Unlike Foucault’s other work, we are given a series of enunciative positions relative to the establishment and transformations of domains that claim a certain objectas we traced above, that of languagethat seems to proceed in accordance with an internal logic of development wherein the crises undergone are presented as if resulting from a complication in the essence that negates its limitation through a re-specification of totality. The descriptive movement of Foucault’s reduction thus seems here to operate a certain inverted Hegelianism where the spirit of the concept becomes progressively lodged within an exteriority it can no longer internalize: the teleology of absolute knowing is transformed into that of an absolute death. And it would thus be within the interminable separation of essence and the finitude of the empirical human it makes possible that science in its modern specificity emerges as what resolutely disperses the possibility of its absorption by a mathesis of knowledge, a canvas of representation, a metaphysics.

Such a characterization has an aspect of truth even if its rendering is pat: for with respect to what we have identified as the dynamic of the reduction proper to Foucault’s historical method, we would have to specify that what is particular to this text is an elaboration from an interior perspective of the identity a discursive regime produces and the transformations to which it is given over. The exteriority or material support that forms the other side of the frontier of énoncés, both in what remains constant and what changes throughout the period of time the text crosses, is largely implicit in the descriptions of its internal movement: the structure of the academic institution, forms of technology, and new knowledges and objects made possible through colonial expansion form a subtext that could be read through what Foucault describes at the level of noetic transformation. It is thus a certain aspect of” the set of conditions in accordance with which a practice is exercised, in accordance with which that practice gives rise to partially or totally new statements, and in accordance with which it can be modified” (AK 230) that the text addresses, that is, the enunciative positions that in their identities and differences form the unity of production characteristic of an épistémè, which is itself the global coordination of discursive practices that the apparent unity of an epoch expresses.

Foucault thus attempts to expose the discursive unity of certain statutes issued from the “anonymous depths of language” that determine the possibility of coordination between words and things, determining in turn the authors representative of the épistémè that produces them. It is thus for Foucault that the form of representation is not something that ought to be surpassed, or that can be surpassed through assuming a more agile metaphysics, but that which is effectively surpassed in its historical forms through the development of sciences and rewritten in the creation of new forms of conceptual production, such as those of the human sciences. Through this frame, it is difficult to see how Foucault could not but regard Deleuze’s theory of the Idea as forming a belated part of the post-Kantian tradition of metaphysics arising out of the empirical limitation of the sphere of representation established by the sciences in the 19th century, or

“the origin of those metaphysical doctrines that, despite their post-Kantian chronology, appear as ‘pre-critical’: they do, in fact, avoid any analysis of the conditions of knowledge as they may be revealed at the level of transcendental subjectivity; but these metaphysics develop on the basis of transcendental objectives (the Word of God, Will, Life) which are possible only in so far as the domain of representation has been previously limited; they therefore have the same archaeological subsoil as Criticism itself” (OT 265-266).

Or again, how his conception of archaeology and its identification of the historical determinations of the épistémè could be reconciled with the Idea insofar as “to deploy a dispersion that can never be reduced to a single system of differences, a scattering that is not related to absolute axes of reference” (AK 226), a relating back which is what the Idea effectively does, even if its system of differences is removed from axes of absolute reference or subjective centrality.

In other words, between the Idea and the épistémè there is a very different understanding of the representative function of language and the space of deployment of representational concepts. Deleuze’s theory of the Idea provides a resolutely metaphysical, and in many ways transcendental, account of the formation of empirical concepts while forming a critique of the philosophical tradition as the long history of an error in its misrecognition of difference as such and of repetition as repetition of what differs. Destroying the old concepts enchained to the reduction of difference is thus a matter of the creation of new forms which more adequately express the nature of the real as unbounded divergence and unlimited temporal differentiation, the conditions of a “thought without image” that is also in many ways a thought without history. Or rather, insofar as historical time can be maintained within this conception, it would be that of the succession of common sense enacted by reactive forces within which singular points of difference would appear as ruptures produced by certain authors. In this way, Deleuze can read, for example, Lucretius, Spinoza, and Nietzsche as if they were contemporaries engaged in the same project.3placeholder This approach appears at odds with what Foucault attempts to perform with his notion of the épistémè and with archaeological analysis in general. Such a notion as the épistémè has nothing to do with the elucidation of a unique system of differences that poses a genetic dimension that can be recovered through following representational differences to their virtual source, and nothing no less to do with the affirmative ethics the conceptual reorientation it outlines requires. In other terms, the épistémè is not the foreclosed truth of the metaphysical tradition that returns through various guises as does the Idea. Rather, it is inseparable from the convergence of representational forms in an epistemic identity that conditions not their possibility of existence, but the possibility to treat as unified that of which they are composed. The reduction of representational forms through the discovery of their common archaeological soil is thus a mode of exposing their proliferation and eventual transformation through the énoncés that structure them. In this way, appeal is never made to an instance that would comprehend discourses on the basis of something that the discourse in question did not already present. Representation is thus not, for Foucault, difference expressed at a base degree, but rather that which is engaged in a politics of truth that defines it as the contingently composed though discursively necessary assignment of its historical roles and rules. What is aleatory in a throw of the dice is most certainly abolished.


IV. Conclusion

Throughout this pursuit—which is indeed only an introduction—to the conceptual figures emergent in these texts that define among the works of their authors the most sustained investigations of the sources and conditions of the form of representation, we have taken as a problem what they understood as generative of their discourses understood as critique: the form of representation as a practice of language determinative of certain epistemological and ontological positions that, taken together, constitute what could simply be called “reason” in its most general acceptation. Such a critique is by no means a negation of representational knowledge, but rather, in the manner that Kant understood it, an exploration of a ground that cannot be made fully evident to the empirical deployment of the representational form. Yet unlike the way that the Kantian tradition proceeds, both of them propose as a task the thinking of the possibility of assigning truth through the order of a concept in a way that does not resolve to a representation of the representational act itself. This is to say: it is not a matter of identifying the possibilities internal to the production of the sense that representation affordsthus, epistemology is not indexed by the question of correctness or correlational validity, ontology not expressive of the problem of the essence that unifies word and thing, but rather the various conditions that, while not simply external to representation, form a complex that does not reduce to its conditions. If we are to find a consonance between the works of Foucault and Deleuze, it is located here in the mutual understanding of the limits of a philosophy operative through abstracting the conditions found in other domains, or that which promulgates its truth through the assertion of a higher order reflection on that which gives itself in a more immediate, though for that reason, more limited way. It is thus clear the relation that their work forms at the level of a critique of the objective pretensions of the philosophical tradition and the subjective assurances it provides to those who engage in its terms. Clear as well the conceptual privileging of difference and discontinuity over the form of identity and continuity constitutive of the most commonplace notions of subjective and objective order. Yet beyond these alliances, it is difficult to see how what they effectively understood as forming the discordant ground of representation can be brought together at all, and it is here that we have lodged our own problem, which concerns what one might call an incompatibility of approaches.

At the outset, we spoke of the works of Deleuze and Foucault as respectively forming a metaphysical and historical series of displacements with respect to a certain conceptual method utilized in both historical and metaphysical reflection identifiable as that of “representation”. In full awareness that their fields of research are not entirely same, the manner in which they form the elements of their critiques and the conceptual forms that arise out of these considerations engage a certain philosophy of languagea reflection on basic determinations of the production of sense that doesn’t simply reduce to a parsing of representationsthat does in fact admit of incompatibility between conceptual possibilities. With his conception of the énoncé, Foucault develops the possibility of the utilization of linguistic structures and their repetition in discourse as generated by a frontier that brings into relation a historically limited set of linguistic phenomena with the materially specific forms of institutions. The énoncé is expressive of this identity and difference of one in the other whose mutual causality defines the possibility of diachronic, historical time through the synchronic implication and exclusion of enunciative possibilities and material practices that define a discursive regime, and beyond it, the épistémè of a culture.

As we tried to locate in Foucault’s archaeology of linguistics, the épistémè corresponding to each epoch he identifies is not understood as an ideal structure that generates the plurality of discursive acts and authors in their similarities and divergences, but rather produces certain representational possibilities through a series of defined, historically specific exclusions each epistemic transformation brings about. The “classical age” as the age of representation is thus defined through its exclusion of certain conceptual forms based in the old ontology of resemblance between words and things so as to develop an autonomous field of language that is able to perfect itself in unlimited development of scientific reach and sovereign control of the order of things. “Modernity” is understood as a transformation that locates in the classical épistémè its residual enchainment to what it attempted to banish; the modern epoch thus discovers the full measure of finite, empirical being through undoing the remaining vestiges of resemblance in representation. We can thus understand epistemic transformations in Foucault’s work as what define a series of transformations in enunciative possibilities and material practices that are always at work within the appearance of unity that the épistémè registers. The functional category of unity is thus necessarily involved in the coordination of discrete practices, objects, and énoncés that, for Foucault, have no essential filiation prior to their establishment in discourse. The subjective ground of this establishment, as we know, is emptied of intentional, synthetic activity, a void that is left open as the infinite negativity by which the reflective subject may encounter the labyrinth without center that precedes it.

Now, it could be said that Gilles Deleuze as well searches in such a labyrinth, but here we say that, in fact, it is not the same: there is a difference in level involved that cannot be reduced to the same structure. It is this lack of reducibility that condemns any facile periodization or intellectual inventorying of this historical moment to theoretical irrelevance. In effect, the exprimé and the level of expression it situates, an ideality beyond essence, forms the irreducibility of sense gathered within language to its fixation by the thing it designates, the subject who manifests it, or the causal order of concepts constructed through signification. As the identity of the difference between word and thing that language can only register through the figure of paradox, it constitutes an impossible determination that gives the orders of signifier and signified their mobility, their perpetual divergence. It is thus that the exprimé is regarded as a genetic instance, for it expresses an irreducible instance in language that seems to traverse its representational orders, establishing a communication between them within a temporal order of becoming that displaces the simultaneity and succession of represented beings. This temporal order is what the notion of virtuality expresses, which forms the coordinates for the theory of the Idea that Deleuze develops. The Idea forms a way of thinking difference and its repetition beyond representational mediation at the subjective and objective levels; it thus constitutes a “superior” form of difference by which thinking is to be reoriented through a liberation from the identity attributed to the subject of reflection and its object, the oppositions of predicates in both subject and object, the analogical distribution of orders of being in judgment, and the valuation of resemblances in perception.

It may now be clear to what extent the respective conceptions of language and the form of time they implicate pose analyses that are not easily reconcilable, that are even irreconcilable. We believe that this is due not only to a conceptual difference, but to a difference at the level of existence in which these conceptual forms unfold, a difference that marks the recurrent divergence between metaphysical and historical analysis. But to say that these works are incompatible, that they situate levels of existence that cannot be identified or reduced one to the other, is in fact to affirm a disjunction that marks the philosophical tradition in its specificity and which, in its proper irresolvability, is perhaps the only reason why such a form of thought persists and has significance. The dimension of time and language as they appear historically open a time and language without historical determinations as what is impossible with respect to the form of historical conditions, claiming a necessity that abolishes the limit, unfolding in the becoming of time a sense that transports the finite and actual by a line situated on the unlimited. The necessity of the impossible turns back towards its material element as what forms its irreconcilable possibility, the mass of historical strata fanning out in the time of past forms by which the impossible is one of its appearances. It is within the mutually divergent that a necessity appears, one which holds together these seemingly incompatible emanations of existence. Between the works of Foucault and Deleuze, the trace of its torn movement appears.

Turner Roth is an essayist and poet. He received a MA in Philosophy from Université Paris 8 and Kingston University London, and a BA from Bard College. His writings have appeared in Epoché Magazine, The International Magazine, People’s World, Movable Type, among others. He lives and works in New York City.

Works Cited & Abbreviations


DR          Difference and Repetition, Continuum, 2001.

F              Foucault, University of Minnesota Press, 2006.

LS            Logic of Sense, Continuum, 2004.



AK           The Archaeology of Knowledge, Routledge, 2013.

OT         The Order of Things, Routledge, 2005.


When Gillian Rose notices that a philosophy such as Deleuze’s which eschews the problem of subjective grounding nevertheless implies a dimension of self-reflexive orientation, she is correct. This problem, however, is never elided in his work, insofar as the dimension of sense with respect to the form of subjectivity is constantly at issue; the question falls to an origin of synthesis that is productive of such reflexivity. But that this approach in general involves a separation of metaphysics from ethics is by no means the case. Concerning the despair Rose understands as endemic to this position and the mourning it forecloses, see G. Rose, Mourning Becomes the Law: Philosophy and Representation, Cambridge, University Press, 1996, pp. 6-7, 122.


In her otherwise lucid and comprehensive reading, Judith Revel locates what she understands as an impasse in the conceptual form of such archaeological indexing that necessitated a turn towards a genealogy of power in Foucault’s later writings: the épistémè introduces a homogeneity of production within an epoch that seems unable to account for its own historicity, implying a notion of epochal unity given in advance: “l’homogénéité de l’épistémè semble malgré tout fondée sur la présence d’éléments invariants […] les invariants expulse en réalité du découpage épistémique la seule chose qui pourrait pourtant lui permettre de revendiquer une réelle consistance, c’est-à-dire le statut d’une véritable périodisation. Cette chose, c’est l’histoire.”  However, this only holds if one reads the forms of discursive unity expressive of an épistémè as substantial categories that produce history rather than as the effects of discursive practices that are themselves historical, which is to say, non-unified strata of linguistic and material elements that produce the appearance of unity. This production implies a genealogy of power inscribed within archaeological analysis, even if the status of power remains ambiguous at this stage of Foucault’s writings. Further, the épistémè, even though it is a static concept, implies a dynamic movement that we tried to show in the imbrications of epistemic paradigms in his archaeology of linguistics: an épistémè is never isolated from that which precedes it and that which replaces it. cf. J. Revel, Foucault, une pensée du discontinu, Paris, Fayard/Mille et une nuits, 2010, pp. 72-73.


It is this form of assimilation that Éric Alliez defends in a short text as the method of reading the philosophical tradition proper to a virtual philosophy, where historically distinct writers can be understood as a procession of simulacra issued from an ontological absolute: “Il n’est d’histoire philosophique de la philosophie qu’à développer des philosophies virtuelles qui dramatisent un jeu de concepts comme expression du jeu du monde. Elle n’a pas telle ou telle philosophie pour objet mais point de vue, pure effectivité comprenant son effectuation réelle, (trans)-historique, comme la prime inflexion et la Pliure originale d’une idéalité en elle-même inséparable d’une variation infinie“. É. Alliez, Deleuze Philosophie Virtuelle, Le Plessis-Robinson, Les Empêcheurs de penser en rond, 1996, p. 42.


August 2023


On The Validity of Normative Life: Habermas’ Discourse Ethics

by Rowan Anderson

Between Historical Reduction & Metaphysical Construction: Critiques of Representation in Foucault and Deleuze

by Turner Roth

Kant on the Problem of the Organism

by John C. Brady

Popular and Generational Shifts in Cultural Capital

by Diego Galán