1The philosophies of Merleau-Ponty and Deleuze are traditionally understood as being antipodes. Michel Foucault, for example, claims that Deleuze’s Logic of Sense “can be read as the most alien book to The Phenomenology of Perception imaginable” (Foucault 1994, 79, my translation). One of the fundamental differences between both philosophies seems to be their stance towards structuralism. Merleau-Ponty is said to be structuralist, whereas Deleuze is considered a poststructuralist. Why?
2Merleau-Ponty was “one of the first French philosophers to become interested in Saussure” (Barthes 1977, 24). And he was not only “interested” in Saussure: when Merleau-Ponty describes his own philosophical project in the period between 1949 and 1959, he constantly refers to Saussure. But also in his later work, in The Visible and the Invisible, the lines of force are, as I will show, structuralist. Deleuze, on the other hand, is traditionally subsumed under the heading of poststructuralism, especially in the Anglophone world. This label stems from the fierce critique of structuralism Deleuze and Guattari articulate in the two volumes of Capitalism and Schizophrenia.
3I will show that this critique is the result of a process in which Deleuze and Guattari extracted and preserved for their use all that was good and worthwhile in structuralism. This is, by the way, also the reason why the label “poststructuralism” is well chosen to designate their work. However, they had no illusions about structuralism’s limitations. I argue that Merleau-Ponty stumbled upon the same problems in Saussure and introduced notions that bear resonances to the concepts Deleuze and Guattari invoked. As will be clear from the length of my description of them, however, Merleau-Ponty’s notions are not as developed as Deleuze’s and Guattari’s.
4To find out what this comparison of Merleau-Ponty’s and Deleuze’s references to structuralism implies for the more general characterization of their relation, I refer to my book Thinking Between Deleuze and Merleau-Ponty (Ohio University Press, 2018) of which this article is a revised excerpt.1 My description of Merleau-Ponty’s stance towards structuralism borrows heavily from James Schmidt’s excellent book Maurice Merleau-Ponty: Between Phenomenology and Structuralism.
5Merleau-Ponty started reading Saussure after the publication of Phenomenology of Perception, around 1949 to be precise, and his interchange with linguistic structuralism ended in 1959. He initially turned to structuralism as part of his attempt to formulate a critique of Sartre’s phenomenology. However, structuralism soon became a catalyst for his challenge of phenomenology in general and for the ontological turn that followed this challenge, because it implied the defenestration of the subject and consciousness, as well as a rejection of philosophy as a search for transcendent origins.
6What did Merleau-Ponty write between 1949 and 1959? In 1950, Merleau-Ponty embarked on a new project, called The Origin of Truth: this would leave the world of perception extensively dealt with in Phenomenology of Perception and enter the world of signs. More specifically, Merleau-Ponty hoped to answer the following question: how can signs have a meaning for others and how are significations institutionalized and transformed over the course of history? The first part of this project, entitled The Prose of the World, consisted of a study of linguistic and aesthetic signs and was geared towards preparing the question: how are institutions signs, how are they the expressions of historical powers? The conception of institutions and history that Merleau-Ponty developed at that time is fundamentally structuralist: just as linguistic signs are determined by the differential relations between them, so too institutions only have a meaning in relation to one another. History evolves, not because it is driven by internal or external principles, but because of changes in “symbolic matrices” (Merleau-Ponty 1973a, 101; Schmidt 1985, 137).
7The Prose of the World turned out to be the only part of The Origin of Truth Merleau-Ponty actually realized. At the end of 1951, Merleau-Ponty stopped the project of The Origin of Truth and made a new attempt to write a structuralist philosophy of history, this time without taking the detour of linguistics: Adventures of the Dialectic (1955). He soon realized, however, that deconstructing a philosophy of consciousness by focusing on the differential genesis of structures could only be completed through a total rejection of the notion of the “self” and implied notions, such as “other”, “thing”, “concept”, “idea”, “representation”, “experience”, “projection”, “intentionality” and even “significance” (Merleau-Ponty 1968, 176, 167). The Visible and the Invisible (1964) would become the book where the remains of the dualistic ontology presupposed by the aforementioned concepts completely disappear in favour of a being that is differentiated through hollows and divergences (écarts).
8Let us now turn to the exact references to structuralism in Merleau-Ponty’s writings. These are for the most part to be found in his linguistic-philosophical writings from the period 1949-1959, namely The Prose of the World (1950-51), “Indirect Language and the Voices of Silence” (1952), “On the Phenomenology of Language” (1951) and the course “The Problem of Speech” (1953-54). In general, we can say that these references bear upon two aspects: the differential and the immanent dimension of Saussure’s structures.2 The differential nature of a structure resides in what Saussure calls the “diacritical” nature of the sign:
When Saussure used to say that linguistic signs are diacritical – that they function only through their differences, through a certain spread between themselves and other signs and not, to begin with, by evoking a positive signification – he was making us see the unity which lies beneath a language’s explicit signification, a systematization which is achieved in a language before its conceptual principle is known (Merleau-Ponty 1964, 117).3
9The diacritical nature of the sign refers to the fact that the sign is not determined by the external reality to which it refers, or by the concept it expresses, but by its difference from other signs. A linguistic element does not exist in itself, but only in relation to other linguistic elements. This idea has severe theoretical implications: it goes against positivism – “taken singly, signs do not signify anything” (Ibid., 39) – and against any atomist and causalistic thinking – “the learned parts of a language have an immediate value as a whole, and progress is made less by addition and juxtaposition than by the internal articulation of a function which is in its own way already complete” (Ibid., 40). If a sign receives its meaning from how it differs from other signs, its meaning is not grounded in something – the referent to which it refers, for example – but in the interval, spread (écart) or emptiness between signs (Ibid., 42; Merleau-Ponty 1973b, 43).
10All these theoretical consequences (non-positivism, non-atomism, non-causalistic thinking), however, are unified by the fact that they are implied by the immanent nature of the sign. It is because Saussure wants the genesis of the sign to be immanent – meaning that the ground of the sign cannot be sought outside the sign (transcendent) – that he cannot consider this ground to be something that precedes the sign. However, this immanence cannot be understood along classical lines to mean an absolute transparency between the sign and its origin, sense, or to mean that the sign contains its sense once and for all. An example of this classical immanence would be Husserl’s conception of sense as originated in the subject and thus without any secret for the subject. For Husserl, or at least the Husserl from before The Crisis, the genesis of meaning begins and ends in the subject. Merleau-Ponty, on the other hand, remarks on numerous occasions that “the genesis of meaning is never completed” (Merleau-Ponty 1964, 41-2), that “the idea of complete expression is nonsensical, and that all language is indirect or allusive – that is, if you wish, silence” (Ibid., 43). Thus, Merleau-Pontian immanence does not imply transparency. It simply means that nothing precedes language, that there is nothing that transcends language: “language never has anything to do with anything but itself”, and “It is words that words arouse” (Merleau-Ponty 1973b, 115). The ground of the sign is neither the subject who creates signs in order to express his feelings or refer to reality, nor the reality or the referent of the sign. There are only differentiations within a structure of signs that already exists. Hence, the ground of the sign does not transcend the sign and is not fundamentally different from it. The following passage, with its very Derridean conclusion, indicates that it was also Saussure who inspired Merleau-Ponty in this respect:
these elements [of the “verbal chain”] form a system in synchrony in the sense that each of them signifies only its difference in respect to the others (as Saussure says, signs are essentially “diacritical”); and as this is true of them all, there are only differences of signification in a language. The reason why a language finally intends to say and does say [veut dire et dit] something is not that each sign is the vehicle for a signification which allegedly belongs to it, but that all the signs together allude to a signification which is always in abeyance. (Merleau-Ponty 1964, 88)
11Merleau-Ponty’s differential and immanent conception of the sign thus clearly bears a structuralist stamp. However, there also exists a fundamental point of divergence, which Merleau-Ponty never presents as such. On the contrary, he incorrectly attributes his adaptation of structuralism to Saussure.4 This divergence has to do with Saussure’s distinction between language system (langue) and speech act (parole), between, on the one hand, the socially-shared set of abstract conventions within a linguistic community, the a-temporal and universal structure of a language, and on the other, the particular choices made by a speaker deploying that language, the unique and singular events governing actual speech (Merleau-Ponty 1973b, 23). According to Saussure, linguistics has to concentrate on the langue.5 Merleau-Ponty, on the contrary, proposes to complement the diachronic study of the langue with the synchronic study of the parole. He is not interested in the logical a prioris of language, like Husserl was, but in the rules or universalities that govern speech and that do not presuppose a common essence. Merleau-Ponty is not interested in a vertical universality that presupposes an underlying essence, but in a lateral universality that is based on a shared existence, on something common within our speech acts (Merleau-Ponty 1964, 120; 1973b, 23). In contrast to the logical a prioris of language, this shared existence cannot be identified, let alone formalized, because it is always coherently deformed through these singular speech acts. In other words: parole cannot be considered a simple effect of langue because parole supports and modifies langue.6 Hence, when Merleau-Ponty says he wishes to complement the diachronic study of the langue with the synchronic study of the parole, this does not mean he wants to concentrate on the accidental events themselves, but rather on how a speaker animates the structures of language, how he resumes and innovates langue with every speech act. A synchronic study does not examine the immediately visible aspects of parole, but how these expressions form an answer (and thus become signifying systems) to the questions or problems of human existence in general, or of human existence in a specific era. It focuses on the structure or “style” (Merleau-Ponty 1964, 61; 1973b, 71) that is eminent in and immanent to a specific expression. As such, it always already involves langue or, better said, some kind of generality or structure. Hence, parole indicates a level that precedes the clear distinction between, on the one hand, the singular, accidental facts or events and on the other, the universal, static laws or a prioris.
12If we want to examine Deleuze’s references to linguistic structuralism, we have to turn to his early work, more specifically to Difference and Repetition (1968), The Logic of Sense (1969) and the essay “How Do We Recognize Structuralism?” (published in 1972, but written in 1968). All of these texts were written during the heyday of structuralism. Not surprisingly then, in these texts we see Deleuze trying to determine his stance towards this new school of thought.
13It is striking to see that Deleuze picks and copies exactly the same ideas from Saussure as Merleau-Ponty did: he’s equally interested in the differential and immanent dimension of structuralism's conception of signs. With respect to difference, we can, for example, refer to Deleuze’s statement from Difference and Repetition that Ideas – the virtual and problematic conditions of thought – are “structures” or “a system of multiple, non-localisable connections between differential elements which is incarnated in real relations and actual terms” (Deleuze 2004a, 231). Moreover, just like Saussure’s structures, Deleuze’s Ideas “have neither sensible form nor conceptual signification, nor, therefore, any assignable function. They are not even actually existent” (Ibid., 231). They do not have an identity in themselves: they are merely connections.
14In The Logic of Sense, Deleuze refines the idea of the differential nature of a structure by organizing the elements of the structure into series. Every structure has at least two series, one of which is the signifier and the other the signified. These series converge toward a paradoxical element that works as their differentiator. The paradoxical element does not belong to either series, but circulates between them and thus needs to be situated at the virtual level. It links the series, reflects them in one another, makes them communicate and thus defines the relations between the elements of each series. The paradoxical element, in itself, has no sense. Or more specifically, it is sense and nonsense at the same time. It is always displaced with respect to itself. It “is absent from its proper place” and “proper identity” (Deleuze 2004b, 67), which constitutes its nonsensical character. What accounts for its sense, however, is the fact that this paradoxical element relates two different series, thus generating a new meaning. As an illustration of this paradoxical instance, we can refer to the stripe motif that returns in every piece of the French artist Daniel Buren. As such, the stripe motif does not mean anything. Each time Buren uses it in a different context – for example on pillars in the Parisian garden of the Palais-Royal, as a garland of flags in a street in New York or on the stairs in front of the Art Institute in Chicago –, he incites a communication between two totally different regimes: in the case of the Palais-Royal, the pillar that carries the motif of deck chair fabric suggests a pragmatism that interacts with the charm of classical architecture, thus revealing the classical sense for balance; in the case of the New York street, the flags carrying the black and white motif imply a strictness that interacts with the festive connotation of the garland, suggesting New York’s controlled folly; in Chicago, the stripes on the stairs refer to safety ribbons, thereby turning the visit of the museum into a risky adventure. What is so special about Buren’s art is that he succeeds in evoking all these meanings by always using the same element. However, the element in itself has no meaning. It is like a chameleon: it borrows its meaning from its context. This context consists of a combination of regimes that are not given, but introduced by the paradoxical element itself. It is, for example, the presence of stripes in the context of stairs that brings up the association with danger and thus with safety ribbons. In another context, for example classical architecture, the stripes will produce other associations. In other words: the context is never just there, but always already influenced by the presence of the paradoxical element.
15In the essay How Do We Recognize Structuralism? – which was endorsed by Louis Althusser, despite its idiosyncratic nature – Deleuze enumerates again the aforementioned ideas. The first characteristic of structuralism, according to Deleuze, is its discovery of the order of the symbolic, as distinct from the real and the imaginary. This is what Deleuze will later call “the virtual”. The second criterion states that a structure does not refer to a pre-existent reality, or to a conceptual meaning, but to the way in which actualities differ from one another. Thus, it has a “positional” or “topological” sense (Deleuze 2004c, 174-5). The third criterion is where Deleuze refines his analysis of the composition of virtual structures. A structure does not consist only of differential relations that determine the symbolic elements, but also of singularities that correspond to the value of these differential relations. The differential relations between phonemes, for example, determine the singularities of a language, around which the sonorities and significations characteristic of that language are built. In the fourth criterion, Deleuze explains how the determination of virtual structures differs from the determination of the actual objects and relations in which they incarnate themselves. The virtual is differentiated, whereas the actual is differenciated.7 The fifth and sixth criteria sum up ideas that were already put forward in our discussion of The Logic of Sense: the symbolic elements of a structure are organized into series and, from there, they relate to other series. These series are constructed around a paradoxical element, called here an object x, that has neither identity nor place. Or, more specifically, it “has no identity except in order to lack this identity [pour manquer à cette identité], and has no place except in order to be displaced in relation to all places” (Deleuze 2004c, 188). Just like the empty square in a game, the object x must remain empty for the structure to evolve (Ibid., 191). Deleuze points out, further, that the emptiness of the object x is not negative. It is not empty in the sense of non-being but in the sense of being problematic, of being-problem or -question, all of which signal Deleuze’s attempts to theorize a positive or excessive emptiness. In Merleau-Ponty we can find an idea that is similar to Deleuze’s recognition of the importance of “the empty square” for the movement within the structure: it is because, in original speaking and thinking, we can never fully express what we mean, because there is always a divergence between sense and sign, that we continue speaking, that we do not stop innovating language. Merleau-Ponty, like Deleuze, conceives this divergence in terms of something that is absent from its proper place – the “ground of meaning is decentering” (Merleau-Ponty 1973b, 45 note) – and he associates it with excess rather than with lack: “What we mean ( . . . ) is only the excess of what we live over what has already been said” (Merleau-Ponty 1964, 83).
16With respect to Deleuze’s interest in the immanentism of structuralism, we can refer to the statement in Difference and Repetition that structuralism is the only valuable genetic model of explanation: “‘structuralism” seems to us the only means by which a genetic method can achieve its ambitions” (Deleuze 2004a, 231). How does Deleuze understand this genetic method and what is the link with immanence? According to Deleuze, a genetic method is first of all characterized by the fact it conceives the condition as being immanent to what it conditions because the condition is situated between the empirical. As structuralism situates the origin of the sense of a sign in how the sign differs from other signs, the origin does not lie outside of the empirical but between the empirical. Secondly, a genetic method does not understand the condition in terms of what it conditions. Since there is no resemblance between a differential relation and a linguistic meaning, structuralism cannot be accused of committing this error. There is a difference between condition and conditioned, but this difference does not imply a difference in being, a fundamental difference. It is a difference that does not break the immanence of being.
17In The Logic of Sense (11th series) Deleuze writes that structuralism’s philosophical value consists in offering an alternative to philosophical theories that have attributed to sense a hidden transcendent character by considering it, for example, God’s new avatar or by situating it in the depths of the human being (Deleuze 2004b, 83). These transcendent conceptions of sense assume that sense is some sort of origin or principle. Structuralism, on the contrary, shows that sense is not to be found in some kind of depth or height, but is produced on the surface. Sense, understood as a result and as a by-product (as in optical effects), is a surface effect of the linguistic elements. Moreover, with what Deleuze describes as the paradoxical element within structures, structuralism paves the way for an understanding of genesis that is moved by paradoxes, not by some supreme identity or essence.
18Although the majority of the references to structuralism in the early Deleuze is positive, there are also some critical remarks to be found. The first one corresponds to what Merleau-Ponty already noticed: Saussure’s focus on the study of langue is contradictory to his immanentism. Deleuze explains that the focus on langue runs counter to Saussure’s idea – or to Deleuze’s interpretation of his idea – that sense is produced (and not constituted) and is, thus, essentially a practice.8 In other words, the fact that structuralism chooses, at the expense of the speech act, the a-temporal forms of language as its object of study indicates that the dynamic nature of structuralism’s notion of “structure” is rather limited. One might say that, in Saussure’s structuralism as interpreted by Deleuze, the event in the structure is limited to elements – and relations between elements – that have already been determined as belonging to this structure. It is a closed system. Merleau-Ponty shares this observation, although he does not present it under the form of a criticism on Saussure:
Saussure may show that each act of expression becomes significant only as a modulation of a general system of expression and only insofar as it is differentiated from other linguistic gestures. The marvel is that before Saussure we did not know anything about this, and that we forget it again each time we speak.
19Merleau-Ponty goes on to argue that this proves that each partial act of expression
(...) is not limited to expending an expressive power accumulated in the language, but recreates both the power and the language by making us verify in the obviousness of given and received meaning the power that speaking subjects have of going beyond signs toward their meaning. Signs do not simply evoke other signs for us and so on without end, and language is not like a prison we are locked into or a guide we must blindly follow; for what these linguistic gestures mean and gain us such complete access to that we seem to have no further need of them to refer to it finally appears at the intersection of all of them. (Merleau-Ponty 1964, 181)
20Contrary to Saussure, Merleau-Ponty is thus stressing the fact that, although the meaning of signs is just the product of the diacritical play between them, signs nevertheless refer to the world and thus open up the closed system of language. James Schmidt captures the paradox nicely: “Self-reference makes reference to the world possible” (Schmidt 1985, 131). It is remarkable that, in this context, Merleau-Ponty mentions the same (rather obscure) linguist Deleuze refers to: Gustave Guillaume (1883-1960) (see Merleau-Ponty 1964, 86). In his approach, Guillaume substitutes the principle of differential position for that of distinctive opposition (cf. Deleuze 2004a, 256).9
21This brings us to the second criticism Deleuze reserves for Saussure: his linguistic structuralism employs a binary and negative logic. It tends to think the differential relations between phonemes as relations of opposition (Ibid., 255). Thinking in terms of oppositions is, as we know from Difference and Repetition, a form of identity-thinking Deleuze rejects.
22When Deleuze starts to collaborate with Félix Guattari, his criticism of structuralism becomes more outspoken and sharp. We can sum up Deleuze and Guattari's stance to structuralism in one sentence: they replace the notion of “structure” with the notion of “machine”. The kind of structuralist theory they have in view determines the kind of machine they bring to bear: desiring machines are alternatives to Lacanian structures; signifying machines replace Saussure’s linguistic structures, economic machines are brought in when political-economical structures are at stake, and so on. What does this mean?
23Let me explain by focusing on Deleuze and Guattari’s notion of the signifying machine. Deleuze and Guattari believe language is not a representational or expressive tool manipulated by a subject, but a force that influences people’s behaviour and thoughts, a force that shapes reality. This conception is clearly inspired by Austin’s speech act theory. By revealing language’s performative capacity (doing something by saying it: I swear by pronouncing the words “I swear” [Deleuze & Guattari 2004, 86]), as well as its illocutionary potential (doing by speaking; for example: I ask a question by saying “Is …?” [Ibid., 86]), Austin shows that there are actions proper to language itself, actions that can only be executed in and through language. Deleuze and Guattari pursue this idea further when they claim that language does not serve an informational or communicational aim, but is a way of acting on others and on the world. For example, when a journalist writes an article in the newspaper, he is not describing the situation in a specific part of the world; he is telling his readers how to think of that part of the world and of the world in general, what to expect in the future, what to fear, etc. Hence, Deleuze and Guattari name language’s elementary unity the “order-word” (Ibid., 84). This order-word does not presuppose meanings that are given, but the inverse: the discourse of different significations presupposes the discourse of the order-word or the field of dominant and dominated (or subjected – assujetti) forces (Ibid., 88).
24In other words: language is a machine because it acts. Moreover, Deleuze and Guattari do not believe that language originates in a subject, in a speaking subject. In the words of the journalist resonates the culture in which she was raised, the encounters she has gathered, the gender identity she has constructed for herself, etc. Through this one particular voice, many voices are speaking. Which is why Deleuze and Guattari write that “There is no individual enunciation. There is not even a subject of enunciation” (Ibid., 88). Enunciation always has a social or collective character, which is, therefore, not an extrinsic property but an intrinsic one. It is not as if enunciation can be determined by its social context, it is always already shaped by this context.
25Because of these two elements – the active and collective nature of language – Deleuze and Guattari replace the notion of the speaking subject with the one of “collective assemblage”: “a subject is never the condition of possibility of language, or the cause of the statement: there is no subject, only collective assemblages of enunciation” (Ibid., 144). According to Deleuze and Guattari, an assemblage has two aspects or dimensions: on the one hand, there is the concrete dimension of the assemblage – called “the machinic assemblage of bodies” or “content” – and on the other hand, we have the abstract dimension of enunciation, the “collective assemblage of enunciation” or “expression’. Hence, content “is not a signified nor expression a signifier; rather, both are variables of the assemblage” (Ibid., 101). The machinic assemblage of bodies is an assemblage of things and bodily modifications, whose form is determined by the “woof of [interacting] bodies” (Ibid., 95). The enunciative assemblage is an assemblage of acts and statements, of incorporeal transformations, whose form is determined by the “warp of expressed” (exprimés) that attribute themselves to bodies. According to Deleuze and Guattari, content and expression are ontologically dependent of one another: the content is not the referent or cause of the expression, and thus it is neither primary nor transcendent to the expression. On the contrary, content and expression shape one another. Deleuze and Guattari speak of an intervention: the expression affects the body by inserting itself in the body and vice versa.
26Content and expression are thus two ends of the same axis of the assemblage. However, the assemblage can also be considered according to how it links with its condition of reality. The scheme of the assemblage thus also comprises of a vertical axis according to which it deterritorializes when it goes up the vertical axis, or territorializes when it goes down the vertical axis, in the direction of the concrete bodies and expressions. The vertical axis connects the assemblage, via the abstract machine, to the plane of consistency or immanence, which Stephen Zepke defines as a material flow, a “genetic flux of unformed matter/energy” (Zepke 2005, 118). The plane of consistency is the assemblage’s condition of reality because its flux of unformed matter and energy allows the abstract machine to extract matter-functions from it. These matter-functions precede content and expression in the sense that they only contain traits of expression (‘tensors” or functions operating as differentials, such as clear-obscure, line-colour, or closed-open form) (Ibid., 121) and traits of content (degrees of intensity, resistance, conductivity, heat and speed). When these traits of expression and content stratify or territorialize, we end up with actual assemblages of bodies and enunciations.10
27What are, from a Deleuze-Guattarian perspective, the shortcomings of Saussurean linguistics when seen from the scheme presented above? To begin with, the object of Saussurean linguistics is limited to the right side of the X axis, that is, to the collective assemblage of enunciation. Saussure does not pay enough attention to how the latter presupposes the machinic assemblage of bodies, to how the expression inserts itself into the body and vice versa. Deleuze and Guattari describe structuralism’s fixation on the assemblage of enunciation as an “imperialism of the signifier” (Deleuze & Guattari 2004, 73). In addition, it can also be said that Saussure has neglected the vertical axis – the deeper ground of the assemblage, that is, the abstract machine – and the plane of consistency as well. Deleuze and Guattari write, citing Bakhtin, that “there must be “an extra something” that “remains outside of the scope of the entire set of linguistic categories and definitions”, even though it is still entirely within the purview of the theory of enunciation or language” (Ibid., 91). In other words, Saussure has not gone far enough in the process of abstraction, of distancing himself from concrete things, concepts and words (Ibid., 100). Because this plane of consistency is one of ultimate differences that can only be expressed in rhizomatic traits, structuralism’s failure to reach this plane is mirrored in the fact that it brings to the fore linear and tree-like structures that are constructed around constants and universals (Ibid., 102). Differentiating its binary system and giving more weight to the signified will not help to bring the vertical axis to the surface because it will not alter the fundamentally closed and autonomous nature of its structures.
28According to Zepke, Deleuze and Guattari’s criticism of structuralism ultimately suggests that they think it incapable of giving up the notion of “subject” (Zepke 2005, 120). The fact that structuralism refrains from constructing something like the Y axis indicates that, in the end, it cannot think through an a-subjectival origin of language. Like Zepke, Anne Sauvagnargues also describes structuralism and the philosophy of Deleuze and Guattari as being in total opposition. She deduces from their text, “Memories of a Naturalist” (Deleuze & Guattari 2004, 258-61), that Deleuze and Guattari consider structuralism to be anchored to a representational or mimetic conception of language (Sauvagnargues 2005, 225-6). Beyond the fact that Deleuze and Guattari never said anything to bear out such strong claims – I think both Zepke’s and Sauvagnargues” presentations of Deleuze and Guattari’s criticism of structuralism are disproportional to what Deleuze and Guattari clearly inherited from structuralism. As we have shown, it is absolutely clear that structuralism made very significant headway in dethroning the subject. Moreover, the way it achieved this – its focus on the generative power and the non-positivist presence of differences – is very much in line with Deleuze and Guattari’s machines. It makes no sense, then, to situate structuralism in opposition to Deleuze and Guattari’s philosophy. It is true Deleuze and Guattari identify some shortcomings in structuralism; in particular, structuralism fails to develop the impersonal ontology that would allow it to definitely give up the notion of subject. It is still haunted by the remnants of this notion. A good example here is the notion of identity, which is presupposed in the understanding of difference as opposition. Deleuze and Guattari can discard this by proposing that a structure is a self-organizing process. They replace structuralism’s structures by machines, by pre-individual grounds of individuation. Just because structuralism failed to see through to the end the consequences of the ideas it was on to – something Deleuze and Guattari do: they identify structuralism’s limitations and try to overcome them – it does not mean that it needs to be situated in the “enemy” camp. Doing so would be to blame structuralism for not being Deleuze-Guattarian.
29Despite differences in their articulation of their respective positions towards Saussure’s linguistic structuralism – an articulation that, in Merleau-Ponty’s case in particular, is not very reliable – it is clear that Deleuze and Merleau-Ponty are interested in the same ideas Saussure brought to the fore. They are both inspired by his discovery of the genetic power of difference, by his differential theory of the individuation of sense. They adopt Saussure’s idea that the origin of sense cannot be situated in an actual identity that precedes the sign and is external to it. On the contrary, the meaning of a sign is determined by its difference from the meaning of related signs. Thus, one need not search for the origin of meaning outside of language – in some sort of depth – but inside it (immanence) or, as Deleuze says, on its surface. Nor need one look for the origin of meaning in some kind of identity because its origin – insofar as one can still preserve the notion of “origin” – is relational or positional (difference). The implication of this differential conception of the genesis of sense is that sense cannot be understood as a positive entity but rather as something that is characterized by a fundamental emptiness or paradox, as well as by a fundamental dynamics.
30Both Deleuze and Merleau-Ponty recognize the tension between Saussure’s immanent and differential conception of language and his subordination of parole to langue. Merleau-Ponty does not criticize Saussure with so many words, but the fact that he proposes to complement the diachronic study of langue with a synchronic study of parole indicates that he realizes that there is a danger that Saussure’s preoccupation with langue will turn into a search for the logical a prioris of language, that is, into a search for structures that are divorced from the individual speech acts that shape them. Merleau-Ponty’s idea of a universality that is not prior to facts, to the singular or historical event, but is simultaneous with them, is an attempt to bring langue and parole together. The early Deleuze will reverse the subordination of parole to langue by arguing that linguistic structures are events and are constituted by the individual speech acts, which nevertheless presuppose these structures. The later Deleuze abandons the notions of langue and parole altogether. Austin’s insight that language is not passive but is itself a force affecting bodies leads Deleuze and Guattari to the idea of an impersonal machine that not only conditions enunciations and bodies, but directly affects them, in the Spinozaen sense of the word. Although the machine possesses the same immanent and differential characteristics as the structure, it cannot be considered a structure exactly because of its affective nature, because it is a force rather than a neutral condition. Hence, for Deleuze and Guattari, an affect has nothing to do with the emotion experienced by a subject, but with the encounter of forces as a result of which changes occur in relations and intensities, creating new meanings, desires and bodies. Thus, affects not only hit subjects, but also abstract entities, and they do not imply a final state but rather a transitory configuration, a becoming.
31The fact that Deleuze and Guattari drop the notion of structure indicates a difference with Merleau-Ponty, who seems to be more loyal to Saussure. The same notion we introduced to determine Deleuze and Guattari’s position to Saussurean linguistics can be applied to the difference between Merleau-Ponty’s and Deleuze’s reception of structuralism: while their respective conceptions of language share the same structuralist framework, Deleuze and Guattari’s alternative to overcoming the limitations of the structuralist framework – the impersonal ontology of forces underlying language – has more affective power.11 With his idea that self-reference makes reference to the world possible, Merleau-Ponty likewise opens up and ontologizes Saussure’s closed language system. As is clear from the experiential origin of his theories, Merleau-Ponty’s diacritical ontology not only involves signifiers and signifieds, but also concrete bodies. Moreover, Merleau-Ponty attributes ontological power to the condition of signifiers, signifieds and bodies, or the flesh: it is not only a condition of possibility but also of reality. The flesh is not so much what makes signifiers, signifieds and bodies possible but what brings them into actual being.
32However, the flesh does not possess the force, the brute energy, and the (implied) destabilizing power of Deleuze’s machines. Merleau-Ponty’s flesh only seems to be generative in a positive way, whereas Deleuze’s machines can also cause things to fall apart. They can also generate conflicts. Hence, Merleau-Ponty’s differential conception of the ground and of individuation does not go as far as turning the ground into something that can also unground (effondement, Deleuze 2004a, 80), as is the case in Deleuze. However, in order to decide whether this is a fundamental difference between Merleau-Ponty’s and Deleuze’s philosophies, we should take a look at all the resonances and divergences between both thinkers.