“The Problem of Speech in Merleau-Ponty: My View of ‘Speaking Speech’ and ‘Spoken Speech’ in Light of Ontogenesis”

Abstract

The turn away from phenomenology in 20th century French philosophy was in large part due to an increased emphasis on Ferdinand de Saussure’s notion of “linguistic structure”, i.e., that language is the internal system of differences between signs. Thinkers such as Paul Ricoeur and Jean-François Lyotard famously offered a “semiological challenge” to phenomenology. The idea was that phenomenology, especially Merleau-Ponty’s phenomenology, reduces to the sensible world and cannot think linguistic structure. Thus, the argument goes that phenomenology leaves out a basic element of human life: not only can it not think linguistic structure, but it also cannot think about elements, e.g., writing and text, which are its result. This paper takes up this challenge, especially in reference to Merleau-Ponty’s terminology in Phenomenology of Perception of “speaking speech” (parole parlante) and “spoken speech” (parole parlée). I point out that, in retrospect of his later work, Merleau-Ponty very clearly did want to take linguistic structure seriously. This, however, means that we need to reconsider some of the basic themes in his work. Taking inspiration from the recently published “problem of speech” lectures, I reconstruct Merleau-Ponty’s idea that speech is a concrete limit situation from which we get both the idea of a language structure in which there are differences and of an ontological difference between being and beings. This is an internal criticism of both linguistic structure and formal ontology. I begin the paper by noting that, in Merleau-Ponty’s descriptions of the tacit and spoken cogito, also in Phenomenology of Perception, Merleau-Ponty criticizes the notion of a subject to which language refers and highlights the notion of a subject that defies representational and denotational structure. I do not, however, go along with Merleau-Ponty’s own criticism of the tacit ego, which he ultimately declared too subjectivistic. Ultimately, I hope to stress the importance of linguistic structure and writing in Merleau-Ponty’s ontology. This is an ontology of that is fragile and requires symbolization. This paper emphasizes under-developed themes in Merleau-Ponty’s work such as bodily event, difference, symbolization, and the writing of philosophy.

1. Introduction

In the “The Body as Expression and Speech”, the last section of Part I in Phenomenology of Perception, Merleau-Ponty makes the distinction between speaking speech (parole parlante) and spoken speech (parole parlée) [1]. Broadly speaking, the distinction is between an aspect of speech where meaning is constituted (speaking speech) and where already-constituted meaning is relied upon (spoken speech). The distinction follows the logic of institution: we institute meanings even as we feel them to be external to us, dictating our actions and what we do. It is important to emphasize that, according to the logic of institution, it is in fact our actions that inscribe the very meaning we feel external to us. The same is true, Merleau-Ponty thinks, in speech: we speak freely and produce meaning even as we feel that meaning is given to us. He wants to stress that these are not two different aspects of speech. Spoken speech is re-entrenched by speaking speech, its pre-established meanings are re-established each time anew: the child saying her first word, the lover discovering his feeling just as he says it, or the first person who ever spoke ([1], 184fn7).

At first glance, this view of language is at odds with the notion of the “tacit ego”, which Merleau-Ponty famously rejects in his later work. That is because the tacit ego, being an element of experience immediately available to itself and more like Husserl’s eidetic ego, seems to precede language. But, in the “Tacit ego and spoken cogito” section of Phenomenology of Perception, and commenting on the Descartes’ cogito, Merleau-Ponty writes: “The wonder of language is that it makes itself be forgotten: my gaze is drawn along the lines on the paper, from the moment that I am struck by what they signify, I no longer see them. The paper, the letters on the paper, my eyes, and my body are only present as the minimum of production materials necessary for some invisible operation” ([1], p. 422). The idea of a subject to which language refers, the spoken cogito, is itself produced in language. I am a thinking subject, it seems to me, that can be denoted or spoken about: a subject who has thoughts independent of language and who language translates. The notion of a tacit ego here is, however, quite unlike the eidetic, I will point out, because it is precisely the ego that is expressed as a referent in the cogito. For its part, like the lover discovering his feelings as he speaks, it is turned into something denoted. The very appearing of the subject is, in other words, a linguistic phenomenon.

Commentators have long pointed out that, in stressing the reciprocity between speaking and spoken speech, Merleau-Ponty in fact undermines the importance already constituted meanings in language. For his part, Merleau-Ponty ultimately wanted to undermine the very idea of a tacit ego, in part, because he thought that even this element of subjectivity could be subsumed into a spoken about cogito and is therefore too representational. The apparent inability of phenomenology to accommodate the linguistic structure did indeed lead to the departure from phenomenology, especially in later twentieth century French philosophy. For quite some time, though, Merleau-Ponty commentators have also responded on his behalf and discussed the ways in which he does accommodate linguistic structure [2]1. This is not my focus here. Rather, I would like to go a little further into the reciprocity of speech and suggest that Merleau-Ponty’s “speaking speech” and “spoken speech” show that the tacit element of subjectivity can never be denotational or representational but rather ontological in a unique way. Indeed, Merleau-Ponty was looking for ways to outline the connection between language structure and ontology until his death. More importantly, the reciprocity of speech, I will argue, indicates a special ontology in which some concrete element of embodied life sublimates itself and makes itself forgotten. Instead of betraying an eidetic moment, this is a less rational, more perilous subject with a structure of misapprehension or that cannot guarantee itself. We see this more clearly in Merleau-Ponty’s speech or passivity lectures, but ultimately, I think, speech is a matter of “ontogenesis”: the very genesis of the problem of an exterior world to which I can refer, or of an internal world to which I can refer. Ultimately, I want to show, the wonder of language, that it makes itself forgotten, highlights an element of the body and of being that cannot be spoken about but needs to be spoken, and even written, into expression, that paradoxically needs discovering just as we say it.

2. Some Orienting Comments

I have long thought the dual criticisms, that Merleau-Ponty can think representational art but not non-figurative art and that he cannot think linguistic structure and the “figural” or “figurative” aspect of language, should be approached at the same time: simply by pointing out that Merleau-Ponty radicalizes a basic phenomenological insight, that perception is not representational and that it intermingles the imagination and the sensible. We might even say that, for Merleau-Ponty, perception is indissociable from the figurative and is figurative. On this view, there is nothing like truly “representational” art because, however implicitly, art concerns a figurative aspect of the sensible world and our place in it [3], and this figurative aspect of the sensible world cannot be ignored in language. This more hidden sense of the figurative, I have argued, betrays an ontology not often discussed in the Merleau-Ponty literature: the very ontogenesis according to which our interior life (including linguistic structure) and exterior life is constellated [4]. In other words, “language” needs to be freed from its too-subjectivistic implications and broadened. All manners of phenomena are included within language: unconscious wishes and phantasies, as well as existent-sense objects. More specifically, language needs to be shown to belong neither to the inside nor to the outside, nor even to both at once, but rather to some ontological oneirism from which an entire referential system emerges.

Merleau-Ponty uses the phrase “languagely” or a “languagely ensemble [un ensemble langagier]”, and often speaks of the “pre-linguistic” aspect of experience [5]. In the Preface to Phenomenology of Perception, and responding to Jean Wahl’s declaration that “Husserl separates essences from existence”, he says that this separation only takes “place in the realm of language” whereas in experience “essences are what appear in the silence of originary consciousness” and here we are confronted “not only what these words mean, but also what these things mean, that is, the core of primary signification around which acts of naming and of expression are organized” [6]. Although phases like these lead one to think that Merleau-Ponty’s ontology eclipses linguistic structure and focuses instead on the “silence of originary consciousness”, once language and the sensible world are put back in touch with one another this eclipse cannot in fact be maintained. It is true that, where for Husserl phenomenology reduces to eidos, which is both ground and the idea of ground, for Merleau-Ponty it goes to the silence of originary consciousness. But this silence is not at all self-sustaining, as eidos is meant to be for Husserl. Instead, it reinvigorates the problem of some concrete element that is always inside of the referential structure of intentions and significations. What Merleau-Ponty calls the “primary signification around which acts of naming and expression are organized” is at once an internal criticism of both language and ground. This, I think, reinvigorates speech at the ontological level because it means giving an account of a moment interior to language that, at the same time, does not simply refer to but is inside being: a concrete limit interior to both language and ontology. In this sense, I think, speech is for Merleau-Ponty generative of the problem of ontological difference between being and beings. This being is not foundational or merely archeologically available, I will ultimately argue, because the very idea of an archeological or preceding structure is generated from within this concrete moment. Here we encounter the depths of ontogenesis and differentiation. It will paradoxically demand alternate modes of expression, not just one direct way back towards it. One possible way may be the expressionistic writing style that Merleau-Ponty himself employs throughout his work and appreciates in various authors.

3. The “Problem of Linguistic Structure”

Saussure famously distinguishes la parole, the actual and concrete acts of speech, from la langue, the linguistic system of phonological rules on the basis of which there is a construction of words in language through distinctive, oppositive, or relative characteristics of their component sounds. For Saussure, la parole is “accidental” while la langue is “essential” ([1], p. lxxix). The linguistic system can be further dissected: it bears a diachrony and a synchrony. While diachronic linguistics studies the history of changes within a language (e.g., from Latin to French, Italian and Spanish, or the relationships between German and Sanskrit), a synchronic linguistics is the science of the given state of a linguistic system. Saussure points out, again this is well known, that, in the synchronic linguistic structure, every linguistic value is positive with respect to its content and negative with respect to what it is not. This is the model of meaning within the linguistic system known as “diacritical” (although Saussure doesn’t ever use that term). The diacritical model of meaning—taken from the Greek diareisis, or “meaning by divergence”—emphasizes what Saussure calls the first principle of linguistics, the arbitrariness of the sign: the sign does not mean anything in itself, it does not have a positive value in itself, but it only has value in oppositive or differential meaning to another sign or other signs [7]. A system of signs is, then, the behaviour of identities through their differences. Saussure’s revolution is his insistence in focusing on this system. To use a word, a speaker need not know its historical variations or meanings. The usage of a word rather indicates not only the difference between speech and the linguistic system but the central importance of synchrony within the system. It is only this, for Saussure, which carries meaning within language.

Paul Ricoeur argues that language breaks radically from phenomenology and ontology. He calls this the “semiological challenge” to phenomenology, and especially to Merleau-Ponty. He argues this by pointing to metaphor. In Aristotle’s classic definition of metaphor, for example, disparate terms are brought together so that they suddenly appear close; metaphor forms a kinship between heterogeneous ideas to make a totally new sense. This means that metaphor implies a transfer of sense from an everyday and direct use of a word or words (flame referring to fire) to a novel or indirect purely linguistic use of words (flame referring to love). Ricoeur even extends this observation: a language system involves not only the behaviour between individual signs but a behaviour onto sentences and metaphorical statements. Here, language involves a totally new and innovative use of the imagination—non-logical, virtual—and destroys the consistent or lexical meanings of the coherent use of the imagination in direct language. Actually, as Ricoeur also points out, the term “figure” (linguistic figure, figure of speech, figure of language, and so on) is already a metaphor we use when we talk about metaphors. It suggests that the new use of imagination in metaphor has its own purely virtual terrain and shape, distinct from the actual and sensible terrain and shape in which we exist2 [7]. Moreover, the “figure” is a spatial metaphor that indicates we have no real idea what we mean by metaphor in language; so, when we use the metaphor of “figure”, this is already the destruction of a logical space, the expression of a notion impossible to express, and the formation of the circumference of a purely linguistic space opposite to a sensible space. Jean-François Lyotard’s Discours, figure—a book highly influenced by Ricoeur—especially emphasizes this purely virtual, and not spatial, sense of the figural and puts it directly at odds with Merleau-Ponty’s understanding of perception. There is an opposition between the visible and a structural linguistics, and the strongest account for this, Lyotard says, is the text.

We can surely wonder whether he can take seriously the idea of a linguistic system in which signs behave only in relation to one another—that is, independently of the sensible world and purely virtual with its own terrain and figuration—especially since he wants to return words to the sensible world. In fact, Merleau-Ponty seems to get Saussure wrong when, in the first paragraphs of “Indirect Language and the Voices of Silence”, and introducing Saussure, he seems to plainly confuse la langue with la parole. There, Merleau-Ponty describes how the unique speech apparatuses in the mouth are already caught in a system of signs that make up the language [8]. The apparently strange result is that, even when discussing Saussure, Merleau-Ponty seems to be saying not more than what he said in his “The Body as Expression and Speech” chapter. Namely, that on the one hand speech is active, instituting and fixing meaning, while on the other hand it is passive, is instituted and an intersubjective or even cultural object. Apparently, even in the system of signs, Merleau-Ponty emphasizes the primordial acts of meaning-giving and movement. Of course, for his part, Saussure is very clearly uninterested in this question of an “ontogenesis” of language in speech, and actually uses the term “ontogenesis” to describe the very speech act he excludes from his linguistics ([6], S, pp. 39–45). But “ontogenesis”—and Merleau-Ponty repeats the word ([7], CGL, p. 9)—is precisely what he says needs to be considered when it comes to language.

What is meant by ontogenesis? Why does it implicate language? Ontogenesis and speech, I want to say, go hand in hand. This indicates the possibility of a critique of both ontology and language that goes beyond Merleau-Ponty’s account of Saussure in terms of speech. Merleau-Ponty declares in one of the first published working notes that his ontology starts anew, “from behind [the] point” of philosophy”, and this renewal, he remarks, is prior to both Husserlian and Heideggerian analysis3 [9]. Before Husserl, because Merleau-Ponty is not thinking about intentionality, or signification, in merely referential terms. This is because the reduction to intentional consciousness cannot reduce to eidos, to both what is and what it is about, but rather this reduction is always open or opened. Before Heidegger, because this openness precedes the difference between being and beings, that is, the already too formal idea of some identity preceding difference. Merleau-Ponty calls this an “Ursprungsklarung” ([9], VI, pp. 166–165), and this primordial clearing is a concrete, but still ontological, site of differentiation. This is helpful, I think, when it comes to ontogenesis and speech. A primordial clearing is Merleau-Ponty’s attempt to account for the concrete site of some internal delimitation to formal ontology; and, because it is the very site of differentiation, is also the concrete site of some internal delimitation to the linguistic structure as well. Speech sublimates speech. It is at once bodily and sublimates its own embodied element, and this very splitting is the ontogenesis of the inside and the outside.

4. Language and Body: Reintroducing Speech

Without focusing too much on the development of this thought, let me point out Merleau-Ponty’s arguments: he thinks that differences in the synchronic linguistic system in fact turn on diachronic changes, and he thinks diachronic changes turn on changes in speech patterns. He thinks, in other words, that Saussure’s exclusion of la parole from la langue is untenable [9]. In one sense, Merleau-Ponty is adopting what linguists sometimes call a “sound symbolism” where the sonorous character of the language (intonation, rhythm, pacing, etc.), and the concrete actions of sounding words out, is not extrinsic to sense and meaning. Agamben writes the following about this problem in the context of the dictation of poetry:

When literary criticism and aesthetics finally came to formulate the problem of the relation between lived experience and poeticized experience with regard to the work of art, the terrain on which the problem could have been correctly posed had already been covered over and forever altered [10].

In other words, as soon as the distinction is made between speaking and structure, we are unable to deal with their connection. This will require a set of assumption and have consequences. In Merleau-Ponty Between Philosophy and Symbolism, I gave the example of the historical problem of the epic language in Homeric poetry being affected by the historical changes in the Greek language going back to the Mycenean Age. Words that were pronounced in the common Attic dialect as one sound were in the epic dialect often resolved by separating them into two sounds to make the poem’s metrical requirements work. This resolution is technically called the “diaeresis of the diphthong”—i.e., splitting of the vowel sound—which functions to make the meter of Homeric poetry work. The argument is as follows: lost pronunciations or changed meters in the language may in fact lead to both a lost sense and an alteration in the syntactical rules of the language. It may in fact mark an alteration in both the practical and formal elements of the language. And, in the reverse, the recitation in which a single vowel is split into two sounds would also be a new formal construction of sense and syntax. A further way to understand Merleau-Ponty’s demand to return to “ontogenesis” is political. He joins the Marxist linguists, like Valentin Volosinov and Mikhail Bakhtin, who refuse to put pure structure above the concrete usage of language: speech acts perturb the system.

Merleau-Ponty’s ultimately point is that the exclusion of speech from the linguistic system relies on a too-theoretical division between language and existence, and the supposition of a subjectivity that has access to, or takes a detour through, a purely virtual system. He writes in The Visible and the Invisible, for example:

Ideas that are too much possessed are no longer ideas; I no longer think anything when I speak of them, as if it were essential to the essence that it be for tomorrow, as if it were only a talking thread in the fabric of the words. A discussion is not an exchange of a confrontation of ideas, as if each formed his own, showed them to the others, looked at theirs and returned to correct them with his own… Someone speaks and immediately the others are now but certain divergencies by relation to his words, and he himself specifies his divergencies in relation to them [9].

And elsewhere, also in The Visible and the Invisible:

The speaking-listening duality remains at the heart of the I, its negativity is but the hollow between speaking and hearing, the point where their equivalence is formed—the body-negative or language-negative duality is the—the body, language, as alter ego … ([9], p. 246).

It is not as if thought is monological, Merleau-Ponty says, trapped inside of me, and in need of translation in language. In fact, he thinks, there is no thought that refuses language, and no language that refuses discourse. Yet discourse requires something even further: my interlocutor and I can agree to speak “about something” but only because of a situation in which we find ourselves and not have not simply agreed upon. There is in this sense a lack of agreement between speaking and hearing. When my interlocutor speaks, for example, strictly speaking I do not hear her but the resonance of her voice inside my ear. What I hear is my own hearing. There is, in other words, an internal disarticulation between hearing and speaking, and the entire interlocutionary situation is shot through with a silence, to use Merleau-Ponty’s word, which, for its part is not spoken or heard but makes hearing and speaking possible. Its internal delimitation is precisely what we discover in language: when I speak, the tongue, larynx and vocal cords disappear and, in retrospect, feel unrelated to the language that is entirely mine; when I write, my hands, arms and fingers fall into the background and opposite to the free ability to call up an entire linguistic system.

This is the production of an inside and outside that, because it does not have a place, is not technically anything that language could be “about”. Rather, in language, my own deviation from myself, my own lack of placement, itself dissipates. I displace the fact that I am displaced. Even when I say this language is mine, Merleau-Ponty will argue, something has already transpired and there is something about which I can speak. In the speech course, for example, he makes clear that this transpiration is not at all formal or having a formal structure. There, he contends with the idea of a “Sprach-form” [11], the idea that there is some form, like grammatical or lexical structures, that internally guides language. Saussure calls such ideas “tempting”, but ultimately should be avoided. Then, in a discussion of Goldstein’s analysis of “inner-speech”, Merleau-Ponty points out that inner-speech “is not the “empiricist sense of images” that “must precede emissions” and “not only in the sense of: talking to oneself…” ([11], p. 124)4. He means that, instead of representing subjectivity, the very idea of a subjectivity that can be represented appears in language and linguistic use, and thus the apparent need to posit preceding grammatical or lexical structures. This is why Merleau-Ponty says he wants his study of speech to go “beyond that of naming”, and that there is a “link of the mind and the verb” ([11], p. 128)5. He wants to analyze the linguistic moment according to which we become the inside and there is an outside.

5. Speech and Ontogenesis

In relation to a “link between mind and verb”, Merleau-Ponty points out that there is a “dedifferentiation” of inner speech, for example, “staccato replacing legato” ([11], p. 127)6. This may make no difference for us in interior monologue. Where inner speech is apparently dedifferentiated, though, it bears a relation to outer speech which of course cannot be dedifferentiated or stripped of difference because then it would lose its meaning to both us and the listener. Interestingly, in the passivity course, Merleau-Ponty borrows Blanchot’s phrase “unspeaking speech” [12] to refer to a special silence in dream language that is entirely interior and bears no obvious relation to speech performed either on the inside while awake or spoken on the outside to others. One can read a relation between this “unspeaking speech” and Merleau-Ponty’s analysis of the sleeping body, also in the passivity course, which he says cannot be “dedifferentiated”7. [9,12] The sleeping body’s refusal to be “dedifferentiated” implies that it is not totally suffused into sensible things or pure passivity. It also means that it cannot be equated with the pure activity of consciousness where it presents dream images to itself.

A connection between speech and sleep is in fact made wholly explicit by Merleau-Ponty when discussing “total apraxia and dreamless sleep”. We experience a sort of “apraxia”, he points out, “in cases of sudden awakening”. There are times when we wake up so suddenly that we “unable to act” and even “the mouth is sealed”. Thus, Merleau-Ponty says, “the system of speech = [a] particular fragile superstructure of the body schema, a subtleness of being in the world”. Here he claims that, when explicit speech is aware and articulate, it has “[a] power [pouvoir] of divergence in general” and is “consciousness of anything”. “Man wakes up [se lève] and man speaks”, he says. “The conversion of movement into expression [is] to be studied through prelinguistic forms of expression. That’s why”, Merleau-Ponty remarks, “I’d like to study movement in painting and cinema…” [13].

But in the same passages, Merleau-Ponty also indicates that consciousness of anything is “not consciousness” [13]. To wake up and speak, in other words, speech “sublimates” the moment of apraxia when the body is sleeping and silent, and when it may be said to be within some specific, rather than general, divergence. Indeed, in the passivity course, and referring to the resistance to dedifferentiation, Merleau-Ponty describes sleep as “being in the divergence” ([12], p. 148) and even “the internal possibility” of ontology [12]. He means that the event of sleep concerns some element of embodiment that interrupts and intrudes into both waking and dreaming life, i.e., an aspect of embodiment caught up and not more than the difference according to which there is an inside and outside. Here, I think a tension needs to be made explicit: rather than only the study of so-called prelinguistic forms of expression, Merleau-Ponty also indicates a study of the very mechanisms of sublimation in speech itself, where it goes from a “subtleness of being in the world” to speaking when we wake up and to others or to inner monologue. This becomes more explicit in his work. In The Visible and the Invisible, for example, Merleau-Ponty criticizes Lacan’s formulation that “the unconscious is like a language”. He calls this formulation a “regional problem” and says the deeper problem is that “being is itself structured like a language” ([9], p. 126). It cannot be missed that Merleau-Ponty calls the écart “diacritical” ([9], p. 224). That being is itself structured like a language means that being is not more than the differences that produces identities. The speech act can be located within the virtual system of signs behaving differentially and, simultaneously, within the very being that behaves differentially. The concrete embodied activities of speech involve some ontological texture that takes a route and runs through interior and exterior life, through language, dreams, figurations as well as things, sensation, and bodies.

An inter-connection between being and speech means we are no longer concerned with language at the level of reference. Although Merleau-Ponty moves towards the prelinguistic forms of expression—painting and cinema—to arrive at the concrete site of expression he says is sublimated by language, he also opens the possibility for a phenomenology of speech that is at once both linguistic and prelinguistic and defiant of reference. This demands a radical account of the meaning that gets sublimated by language and language itself. Here, we are also no longer concerned only with what Saussure meant by the “ontogenesis” of language, but instead pointing to some element of bodily existence from which language and ontogenesis are one and the same problem. I suggest, for example, that speech is the very site of “coherent deformation”. There is a moment in speech that deforms, and what it deforms is an element of embodiment that, since it is in the very differentiation to which it would otherwise refer, in fact defies reference: the place where we find the mouth is sealed. Here, speech reverses things: it is the coherence of explicit spoken speech that sublimates, and it is deformation or incoherence that is the internal delimitation of this speech. At the very site of speech, in other words, there is the emergence of the structure of reference and denotation—“in here are words” and “out there are things”—but this site is technically not anything that language could be about. It is in this context that I have argued for an ontology of symbolization or symbolic formation. It is a symbolic formation, I stress, because the genesis of interior and exterior precedes the structure of reference but is not itself a referent [14]. Though phenomenology typically concerns itself with some meaning that precedes us, Merleau-Ponty’s phenomenology goes in the direction of a meaning that emerges in the very speaking of it. This phenomenology cannot only be archeological but eventive and productive. I do not think we have a philosophy about speech in Merleau-Ponty but a phenomenology that is lateral to, or within, the very “problem of speech”.

6. A Speaking Philosophy—Towards Writing

“[T]hings call for speech”, Merleau-Ponty writes in lecture on Proust from the speech course, and “[t]he interior of things is speech” [11,15]8. This subtle change from prelinguistic forms of expression to speech being on the inside of things was already intimated above. Nevertheless, it allows for a remarkable thesis which places literary art at the very site of sublimation between the sealed mouth and the speaking mouth, and lateral to both painting and cinema.

Commenting on a paper titled “La notion de verb”, written by Jean Fourquet just a few years earlier in 1950, he is especially interested in the historical analysis of the special verb that is “being” [11,16]9. Merleau-Ponty takes the point that ontology itself is a linguistic tradition, and that it is impossible for ontology to ignore its language. He criticizes logical positivism and German Idealism equally for being philosophies that stay at the level of “naming”. In general, he thinks, the Western tradition has been unable to think the language internal to its thought. He, of course, does not say that the West has no philosophy of language or the notion that philosophy is a language game. He simply points out that, where these philosophies are concerned, there is always an implicit adoption of the representational model of language, so that the relation between the subject that thinks and what it thinks about remains unproblematized. The philosophy that is more properly aware of its own ontological situation will, however, be sensitive to the history of the language of being, and even interrogate the very use of the verb. This is the link Merleau-Ponty signals between mind and verb. It points to a profound methodological issue: to be sensitive to the language of being and its contingency means that being does not simply await re-discovery—something already there to which we can return—but something produced and even internal to our very expressions of it. Being is not just a referent awaiting clarification but demanding creativity. Later, but still in reference to the verb, Merleau-Ponty gives the somewhat surprising distinction between speech and literary writing. In the latter, he says, a reference to both the reader and the author is lost. It is also worth pointing out that the authors in whom Merleau-Ponty is especially interested—chief among them Proust and Stendhal—are authors of autobiographical fiction that pose questions concerning identity: who is being written about (Henri Brulard or Marcel) and who is doing the writing? The idea of an identity behind the writing, Merleau-Ponty says, borrowing from Paul Valéry, is a “false illusion (impostures)” [17]. The double eclipse of reader and author is not a drawback so much as it goes to the “thing which is also a hundred things” [11]10. Instead of going to a thing that precedes experience, literary writing goes to both the thing and experience in their possibility of being otherwise and reverie-like—in Merleau-Ponty’s terminology, a lateral relation between experience and being, because experience is not only the power of a general divergence but internal to the specific divergences between things.

This ontology, for Merleau-Ponty, is the more elemental one: being in its primordiality is paradoxical because it does not simply wait for expression but calls for expression, is being in the very expression of it. He famously writes in a working note to The Visible and the Invisible that “[b]eing is what requires creation of us to experience it. Make an analysis of literature in this sense: as inscription of Being” ([9], pp. 247–48). The director of his own thesis, which became The Phenomenology of Perception, already famously accused Merleau-Ponty as having a philosophy that ends in a novel. That is, with thinking literature as the critical limit of phenomenology. That is an accusation, I think, which more properly belongs to both Heidegger and Sartre, for whom poetry and literature, respectively, disclose the truth of the phenomenon or of language; and they did of course write poetry or literature. It is in fact a philosophical point that Merleau-Ponty did not write either under his own name. He does not oppose language to being and he does not think there is a special kind of language that discloses being. For this reason, too, he anticipates having to guard against a more extreme reading of metaphor, e.g., the one that Derrida gives when, commenting on Paul Valéry, he notes that philosophy is regulated by the law of pure cognition, is the desire to hear oneself speak in monologues, and needs to break from this tendency so it can explore its own surpluses [18].

Here is an issue uniquely Merleau-Ponty’s: “I will have to disclose a non-explicated horizon: that of the language I am using to describe all that—And what co-determines its final meaning”. In place of pure phenomenological description, Merleau-Ponty is asking: what language would disclose the fact that language is itself ontogenetic? What language could possibly reveal the ontogenesis of language that no language could ever be about? For Merleau-Ponty, there is, in fact can never be, no one special language. “Inarticulate cry”, “voice of no one”: these are phrases Merleau-Ponty takes from Blanchot and Valéry. He takes “chiasm” from Valéry too, and “flesh of the world” from Claude Simon. “Coherent deformation” is from André Malraux. It is worth considering if he would be stuck at the level of lived body without these phrases, and if these phrases do not phenomenalize an otherwise impossible phenomenon.

The accusation that Merleau-Ponty cannot think language structure and the figurative goes along with the accusation that his phenomenology returns to the sensible world in its bruteness. I point out that this bruteness is both in front of and inside consciousness—entirely through the axis between existence and consciousness—and therefore never free of the figurative and literary. One can see how posing this bruteness and literary writing as one and the same problem conveys a thesis of diacritics: being does not require pure description but neither does it require pure expression, since being is not itself something to which return so much as it is threaded throughout both existence and consciousness and is therefore always differential. Thus, for Merleau-Ponty, phenomenology cannot be only archeological, and being cannot simply be simply pre-linguistic. To return to being, phenomenology instead becomes paradoxically eventive and concerns a being indissociable from the linguistic structure that expresses it. It is important to stress further, and contrary to the challenges we started with, indissociable from the text of philosophy or the writing of ontology. What Merleau-Ponty eventually calls the “endo-ontology”—doing ontology from inside being—is not ultimately meant to be self-circumscribing. A phenomenology concerned with the writing of it becomes instead a reflection that shows the ontological fragility of reflection itself: a philosophy concerned with the fragility of being itself, even a philosophy concerned, then, with its own fragility, written in such a way that many things can be taken from it, that could be said in other ways, or that refuses to let its metaphors become “dead”.