This Mad Play of Writing
[Ce jeu insensé d’écrire]¹

Joseph Attié¹

attie@easyconnect.fr

Abstract: The author compares, in Lacanian theory, the difference between saying and writing. He demonstrates how the signifier is connected to the speech and related to the truth in contrast to the letter which, in writing, does not have any relation to the truth but to logic, as in the scientific discourse.
Key words: Saying; writing; signifier; letter; knowledge; real.

Resumen: El autor compara, en la teoría lacaniana, la diferencia ente decir y escribir. Demuestra cómo el significante está conectado al lenguaje y es relativo a la verdad en contraste con la letra que, en la escritura, no tiene ninguna relación con la verdad sino con la lógica como en el discurso científico.
Palabras clave: decir; escribir; significante; letra; saber; real.

Preliminary
The question of writing is very dear to me. In fact I started out in life with the idea of writing poetry. That was my idea. A haunting fear had kept me back for a long time. Today I would say that this happened in another time, in another époque. This can be translated into the following formula: there once was a summer where one dreamt.
This is why this was one of the first things I told Lacan the first time I saw him, when I came to ask him for a training analysis. Lacan’s first response (for there was a second one) was: “So you are going to be the first poet analyst”.
My naiveté was huge in those times, and I did not know whether he was paying me a compliment or mocking me. At any rate, deep inside me this was a challenge that only redoubled my demand for analysis, raising it to the second degree.
In effect, I could not wait to entrust my ‘texts’ to Lacan. He took it very seriously, and even with great esteem. At least this is the impression I had, or this is what I wanted to believe.
To finish with my confidences, let us say that I did not relent, and that many years later I asked: “But what am I going to do with all this?” And this is where I received the second response: “You must write it now”, he said to me. My feelings were really hurt. Indeed I believed I had entrusted my ‘writings’ to him, and now he was telling me I had to write them!
But in the meantime, the anecdote started to have consequences. It is in the after-effect that I was struck by this, that I had come to take someone (not just anybody, of course) as the witness of a desire to write by requesting a training analysis. Between demand and desire, there was something very subtle which in those days completely eluded me.
All this took the following turn. Fifteen years later I wrote the first article that really mattered to me, and which I entrusted to the Lettres de l’EFP (Nº 20, 1977). It was entitled Le DIT/L’ECRIT. As epigraph, I placed the three categories of the necessary, the impossible and the contingent as re-elaborated by Lacan on the basis of the Aristotelian categories.
Here then is the beginning of this article, since it is part of the consequences of my request for an analysis, and because things are far from being over, and that, thanks to this conference, I am adding yet another chapter to the story.
This is the beginning of the article: “The affirmation that, when we are in the order of doxa, speech is inscribed in sand, whereas writing is cast in stone for eternity, seems to be taken as read. Yet the analytic situation introduces a paradox in the fact that with speech, though it is fleeting and destined to be effaced as soon as it is uttered, something makes its way, is traced and indelibly inscribed, and in so doing it converges with the privilege which is, in principle, granted to writing. So speech [le dit] seems to constitute another alternative to writing, under the necessary but not always sufficient condition of the analytic situation”.
So today I will try (of course, the question of what today is, is always present for the one that utters it; it is a ‘shifter’, as linguistics tells us, and it should always remains “virginal and living” [vierge et vivace], as the poet says) to follow up this interrogation. So much wasted time, I could say to myself. Because the question has not lost its relevance for whomsoever cares to write. Hence my title of today, which I preferred over the apparently more neutral “Does one know what it is to write?”[sait-on ce que c’est qu’écrire ?], these two formulas being Mallarmé’s. Thus, in the first place, I am going to develop Mallarmé’s conception of writing, as a poet of course. And we shall see how this teaches us as psychoanalysts. For the psychoanalyst also asks this question, but not from the same place.

Mallarméan statement: ‘Does one know what it is to write?’
Mallarmé carried this question within himself all his life. But it is only in his lecture on Villiers de l’Isle-Adam that he formulated it precisely. Villiers is hardly read these days. Mallarmé devoted to him a friendship of great loyalty. In a way Villiers was is ‘his ideal hero’, side by side with Hamlet, which is saying a lot. Villiers having died of cancer in 1889, it was in February 1890 that Mallarmé was invited to Belgium for a series of lectures on the author. Let us just make a note of ‘the trait’ which interested Mallarmé in this friend in whom he recognised himself. As he had to present him, he immediately suspended what one calls life, biography… “I seek nothing that responds to this term: truly and in the usual sense, did he live?” (Œuvres completes, p. 482)
Having given during his lecture various quotations from Villiers’ books, we can stop before a quotation from the novel Axel. It is a long dialogue between the two lovers Sara and Axel. Sara calls her lover to go and live far away:
- “Over there… Youth, freedom! The vertigo of power! And who knows… all dreams to be realised.
- What is the point in realising them … (Axel replies) they are so beautiful!”
- To live! (he continues), to live? The servants will do that for us.” (Œuvres completes, p. 504-505)
It is possible to say of Mallarmé that he also did not live in the usual sense one gives to the word. He did not idealise life or what one calls happiness, or a form of jouissance, at the expense of something else. He married almost out of duty. He spent his life as a little professor of English, badly noted by his superiors, and a knock-about for his students. Apparently he did not give a hoot. He had one obsession: to write and to be a poet. He ended being a famous man, first, in the small world of Parisian poets from 1885 onwards, and then his reputation grew. Anyone that was anyone, writers, painters, and intellectuals came to his flat on Tuesday evenings - the famous Tuesdays of the rue de Rome, where he lived. And apparently he fascinated everyone. Everything happened as if he was the only one to know what poetry is and what it is to write. His fame was all the more paradoxical that he had really written very little; only a limited number of poems, whereas his idea of the work and of the Book seemed to imply a large number of volumes. Nonetheless, “he seemed to know what he was talking about”. To the extent that Oscar Wilde, André Gide, Paul Valéry, Huysmans, Paul Claudel …were crowding his little salon.
And in the new biography, published by Fayard, that Jean-Luc Steinmetz devotes to him, he speaks of ‘a fifty-year old beginner’. We must recall that he died in 1898, and that he was 56.
Yet this poet marked not only the second half of the 19th century but also our 20th century. To the extent that someone like Jean-Claude Milner, in a little book entitled Mallarmé au tombeau, devoted to him, (published by Verdier), asks himself: “Should we be Mallarmean?”
What is important for us, through Mallarmé, is his conception of writing. And writing is poetry. Mallarmé reduced all literature to poetry. And this for an apparently simple reason: the instrument of poetry is the Verse. There is an entire doctrine of the verse in Mallarmé which amounts to saying that the verse is a proper name. His now well known formula of the function of poetry - “To purify the words of the tribe” [donner un sens plus pur aux mots de la tribu] - reduces poetry, and all true writing, to ‘a process of nomination’. It is all the more important to remember as Mallarmé proscribes all immediate nomination. For him it amounts to losing most of the pleasure: to proceed by allusion, suggestion, to have recourse to a deux à deux, to a metaphorical play that makes the object vanish to make it appear in a new light without one knowing what it is about.
Some examples will better clarify Mallarmé’s practice.
- “Her pure nails on high dedicating their onyx” [ses purs ongles très hauts dédiant leur onyx] is to name the night, bearer of the heavenly constellation.
- The following line seems more easy: “The virginal, living and lovely day” [Le vierge, le vivace et le bel aujourd’hui].4
This verse does not seem not to harbour any difficulty. One would say that this aims to name ‘the day’. But what is the day? It is completed here with two epithets: it is virginal and lively. It would be nice if every ‘day’ were virginal and living, the bearer of an early morning elation. For those that set out to conquer the world in any case. And it is the early morning of the great revolution. But whose day is it? So let’s question this shifter. It is difficult to figure out who it refers to. The sonnet implies that it is the day of the swan. “A swan from past time remembers…” [un cygne d’autrefois se souvient]. Thus the ‘day’ is articulated to ‘the past’ of the swan, that is to say, every time when the swan lacked the occasion of singing “a liveable country”[la région où vivre]. But who says swan says the poet, its symbol in the Antiquity.
To follow J.-C. Milner’s exegesis, this ‘day’ refers to two other ‘days’, those of two other poets, Victor Hugo and Baudelaire, which have nothing to do with the today of Mallarmé. Hugo’s moment “is to expect it all from a day to come, which is eternally virginal and living”. The Baudelairian moment “is to expect it all from such a day, but to know that it will never happen in this world” (J.-C. Milner, Mallarmé au tombeau, p. 57). The Mallarméan moment consists in “pronouncing that there is no today; that there is no day that is distinct from another; that boredom [ennui] is structural and that death pins us down (ibid., p. 58). To the extent that Milner suggests we entitle this poem (which does not have a title) Death of the Swan or The Tomb of Stéphane Mallarmé.
So this is a nomination. But the real question remains: Does one know what it is to write? It is the first sentence of the lecture on Villiers on which we have to dwell.
1 - “An ancient and very vague, but envious, practice whose sense lies in the mystery of the heart” [une ancienne et très vague mais jalouse pratique, dont gît le sens au mystère du cœur] (Mondor, OC, p. 481). So at first sight it would seem that we are not very enlightened as to what writing is! At the very least, it is ‘a practice’ whose sense lies ‘in the mystery of the heart’. We are dealing therefore with a praxis that supposes a ‘know-how’ of the poet, Lacan would say. A know-how with language. And an entire science of language. One can add, in principle, that the analytic practice also supposes a ‘know-how’ with language. But the analytic practice is not, strictly speaking, a writing. In a sense, something is written for the subject of speech, for the analysand. It remains to know what type of writing is at stake. The analyst is content with taking note [prendre acte], punctuating, interpreting what operates in this writing.
2 - “Whoever accomplishes it (this practice) integrally, withdraws from the scene.” [Qui l’accomplit intégralement se retranche, ibid.]. Such a proposition can be heard by psychoanalysts. The subject who advances a signifier disappears immediately afterwards. More precisely, it is divided by the signifier. Let’s not shy away from saying that so is the poet, but not in the same way. Here we have the key of what draws closer and distinguishes the poet and the analyst, and which will be the core of this presentation.
Let’s take up again the formula “whoever accomplishes it, integrally, withdraws from the scene.” This sends us back to the status of the metaphor in Mallarmé, which is quite central. In effect, Mallarmé’s entire art resides in the suggestion, in the allusion of a signifier to another.
We have numerous echoes of this conception of writing in Mallarmé. To Degas, the painter, who asked him for advice, for a few good ideas to write a sonnet, Mallarmé gave the following reply: “It is not with ideas that one makes sonnets, Degas, but with words”.
As a result it is necessary “to leave the initiative to words, which implies the elocutionary disappearance of the poet” [céder l’initiative aux mots, ce qui implique la disparition élocutoire du poète]. Just like in the free association, which implies a forced disappearance of the subject of enunciation.
3 - Integrally, withdraws from the scene … “It is this mad play of writing (Mondor, OC, p. 481) I will not dwell on this formula. With the questions that have to be raised, namely why and on what account writing is a play, and why is it mad? Let’s allow Mallarmé to pursue this sentence and nearly describe what is at stake in this play.
4 - “To appropriate, in accordance with a doubt,… some duty to recreate everything…” [s’arroger, en vertu d’un doute, … quelque devoir de tout recréer, Ibid.] ‘To recreate everything’, this is no trifling matter. But to recreate everything ‘in accordance with a doubt’ constitutes a method. It is this doubt that we have to underline. It is central for Mallarmé, one could say as much as it is for Descartes. Let’s go to what is most simple, that is to say what may be the most telling. In effect, the doubt is what most tortures the Faun. The Nymphs he wanted to catch, was it a dream, the fruit of his awakened libido, or the very reality the proof of whose existence he will search for a long time?
So doubt, which is, par excellence, Igitur. It is a Latin ‘therefore’. J.-C. Milner relies on Breal and Bailly, authors of Mots latins (1885), in order to explain Igitur. “Igitur was first a particle of time meaning ‘then’. From then it went on to mean ‘consequently, therefore’ (J.-C. Milner, Mallarmé au tombeau, p. 72). And he himself adds: “Therefore Igitur designates nothing in this world save the position - temporal and logical - of a subject that utters Igitur (Ibid. p. 72). To which it is necessary to add that Igitur is what defines the Mallarméan cogito, which could be said as follows: ‘I die therefore I am’. Thus it constitutes the turn in Mallarmé’s life and work.
5 - … to appropriate in accordance with a doubt a duty to recreate everything…, ‘with reminiscences’ [avec des réminiscences], this is a term with a Freudian reference - it is the hysteric who suffers from reminiscences. For Mallarmé, reminiscences have a different function and it is “to attest to the fact that one is where one should be (because, allow me to express this apprehension, there remains an uncertainty)” [pour avérer qu’on est bien là où l’on doit l’être parce que, permettez-moi d’exprimer cette appréhension, demeure une incertitude, Mondor, OC, p. 481]. Again doubt, because the duty is to “to endow our sojourn with authenticity” [de douer d’authenticité notre séjour]. Our sojourn on earth - to endow it with, to found it on poetical speech (we will come back to this poetical speech). And the sentence continues: “One by one, each one of our prides, rouse them in their anteriority and then see” [Un à un, chacun de nos orgueils, les susciter, dans leur antériorité et voir]. First dimension of the labour of writing.
6 - And this is how the paragraph ends, on the point that matters the most to me. “Otherwise, if that was not it, a ‘summons’ to the world “to make its fear equal to rich, ciphered postulates, as its ‘law’, on paper blemished by such audacity - I truly believe that there would be a deception, ‘almost a suicide’”. (“Autrement, si ce n’était pas cela, une ‘sommation’ au monde qu’il égale sa hantise à de riches postulats chiffrés, en tant que sa loi, sur le papier blême de tant d’audace - je crois, vraiment, qu’il y aurait duperie, ‘à presque le suicide.’” ibid.)
Apart from reminiscences, from the prides one has to rouse in order to see, a second dimension of writing is here at stake: it concerns a ‘summons’, an injunction, an ultimatum [mise en demeure] addressed to the world. Which complicates the sense: what can the world be ordered to do? That it should make “‘its [his] fear” equal, ‘its [his]’ referring as much to the poet, to the one who writes, as to the world itself. That the world therefore be the equivalent of its own fear. Let’s note the reversal - the poet is not to make the world equal to his formulas. But the poet’s speech is to be equal to the laws of the world through “‘the rich, ciphered postulates’ as its law”. The law of the world and of the poet. Precisely, the true task of writing is to produce these ciphered postulates which are those of the poet and of the world. We would say that it is for the poet to produce ‘his own mathemes’ which are those of the world. One can also say that each verse, as proper name, should come up to this standard. It should suffice to evoke, after these ciphered postulates, the formula “A throw of the dice will never abolish chance ” [un coup de dés jamais n’abolira le hasard] . The law of the world and the law of the subject. It would be an interesting challenge to demonstrate this on the basis of any of Mallarmé’s verse.
If this was not writing, “there would be a deception, almost a suicide”.
Between the world and the fear of the subject, it is important to bring out the articulation between ‘the Real and truth’. This is what constitutes the very axis and gist of this remark on writing.
In opposition to the world, where the poet has to found ‘his being-there’, as the philosopher says, there is nature. And “Nature takes place, the poet tells us, one will only add cities, railways and many inventions forming our material.” [La nature a lieu, on n’y ajoutera que des cités, les voies ferrées et plusieurs inventions formant notre matériel in La musique et les Lettres, OC, p. 647]. One may attempt to represent them - it is not a writing. “To narrate, teach, even describe - this works even though it would perhaps suffice, to exchange human thought, silently to take or place a coin in the hand of another …” (Narrer, enseigner, même décrire, cela va et encore qu’à chacun suffirait peut-être, pour échanger la pensée humaine, de prendre ou de mettre dans la main d’autrui en silence, une pièce de monnaie, in Mondor, OC, p. 857). This remark was often taken up by Lacan to evoke empty speech.
7 - Thus for Mallarmé writing is a process of nomination. It is what he calls the pure Notion (Epouser la notion is an unfinished poem of Mallarmé published in 1992 by Fata Morgana). To make an object exist as absent. And beyond the Notion this amounts to producing ‘ciphered postulates’. These premises lead to the idea that ‘writing is an act’. For the poet the act par excellence is ‘a way to be-in-the-world’. It was in another lecture, this time delivered in Oxford and Cambridge, on the state of literature, that he announced to the English this incredible event: ‘the verse was tangled with’ [On a touché au vers]. Governments change: ‘the prosody always remains intact’ [Toujours la prosodie reste intacte, in Mondor, OC, p. 624].
That one tangled with the verse is an event more important than the French Revolution, the Empire, the Third Republic. One tangled with the verse in the second half of the XIX century because one started to write poetry in prose, and even to write free verse. “Happy find”, Mallarmé comments, “it is an individual modulation, for every soul is a rhythmic knot” (Heureuse trouvaille, c’est une modulation individuelle parce que toute âme est un nœud rythmique, in Mondor, OC, p. 644).
And this is to say that rhythm is at the foundation of all writing. It is not separable from style, of which, as we know, Lacan gave a true doctrine. Rhythm and style are at the heart of all true writing.
And it is in these times of disruption of prosody that “the act of writing scrutinised itself to its origin” (Ibid., p. 645). One must never forget: writing is an act. An act which sends us back to the fundamental question formulated as follows by Mallarmé: “Namely, if there is cause to write” (A savoir s’il y a lieu d’écrire, Ibid.). If there is cause to write beyond nature, as Mallarmé already noted it. Nature is self-sufficient.
More precisely: “What good is there in the marvel of transposing a fact of nature in its vibrating near-disappearance according to the play of speech, except to cause the pure Notion to emanate without the discomfort of a close and concrete recall” (A quoi bon la merveille de transposer un fait de nature en sa presque disparition vibratoire selon le jeu de la parole, cependant, si ce n’est pour qu’en émane, sans la gêne d’un proche ou concret rappel, la notion pure, in OC, p. 857). The concept should emerge from speech, like a flower, distinct from any bouquet.
There is here an effect of redoubling which touches on the very essence of the act, as Lacan said of the analytic act. The analysand can only found his analytical act because he encountered an analyst who had already authorised himself through his own act. “… the true original analysis can only be the second one, for it constitutes the repetition which makes an act of the first.” (Scilicet No 1, p. 24). The act of writing leads Mallarmé to the question of knowing whether there is cause to write, the original question. This is a question to which Blanchot did not cease to return, hence his formula: “In order to write, one must already have written”.
8 - Mallarmé pushes this Notion he has of writing even further; he articulates it with speech - and this is of great interest to us. Hence his doctrine of the double state of speech. On the one hand, ‘speech is raw or immediate’ [brute ou immédiate], empty speech, Lacan would say, current discourse, universal report, Mallarmé specifies. On the other hand, ‘speech is essential’, and it is poetic speech. And for him the function of writing is to ‘isolate’ this poetic speech.
By the same token we re-find our point of departure, namely the distinction between speech and writing, and their possible articulation.

Lacanian statements: ‘A mountain between saying and writing’
We know that Lacan’s work is divided between the written and the spoken. On one side, we have of course Ecrits, and on the other, the Seminars. There is a problematic which is present very early on in Lacan. Thus, The agency of the letter in the unconscious, 1957 - same date as the Object Relations Seminar - is a text situated by Lacan ‘between writing and speech’ (Ecrits, p. 493). Nonetheless, he will end up making a cutting remark on the question: “There is more than a nuance, there is a mountain between speech and writing” (Conference at Columbia University, Scilicet No 6/7, 1975, p. 43).
Jean-Claude Milner ruminated over this problematic in his work L’œuvre claire. The entire first chapter of this book speaks solely about this. Without lingering over the manner in which he treats the question, I note a little remark of his which entails momentous consequences: “The truth speaks, it does not write” (L’œuvre claire, p. 30). This is a formula which says a lot to every analysand. And every reader of Lacan remembers his famous prosopopeia: ‘I, the truth, speak’ (The Freudian Thing, 1955).
I shall oppose to this two other moments in the teaching of Lacan: Seminar XX: Encore, 1972-73, and Seminar XXIII: Le Sinthome, 1975-76. I chose the first one because there is an entire lesson entitled The Function of the Written, to be understood in the analytic discourse, and the second one, Le Sinthome, because the writing of reference is that of Joyce: literary writing.
To specify the function of the written in the analytic discourse, Lacan begins by evoking the little letters he has produced (Seminar XX, chap. 3).
1 - The (a) which is an object.
2 - The (A) which is a locus, a place, the locus of the (A).
3 - He evokes in passing Bourbaki’s set theory, where one finds ‘logical signs’. “One of them designates the function ‘place’ and is written with a little square”. This simple square is of the order of writing for Lacan.
4 - In the same chapter Lacan puts forth the Saussurian algorithm S/s as writing. The bar has a great importance, since there is nothing to understand in it, even when one utilises it in logic as negation.
5 - Finally, he introduces F.
These three letters, (a), A and F “do not have the same function but are of the order of writing”. So, in an important statement, Lacan tells us: “writing is in no way of the same register as the signifier” (Ibid., p. 31). In this perspective, writing is of the order of the scientific discourse. And to be clear, elsewhere Lacan, as an example of a writing, gives that of Einstein’s formula of relativity.
Let’s make things more precise. In the Letter to the Italians (April 1974), we have the following remark by Lacan: “There is knowledge in the Real” (Ornicar? No 25, p. 8). He adds that “it is up to the scientist to accommodate this knowledge”. What we must stress here is that the knowledge in question has nothing to do with truth. Truth and knowledge are totally distinct here.
“As to analysis, it accommodates another knowledge”, it is a knowledge that can only take account of and be extracted from the truth of the subject. At the same time this knowledge must take account of the knowledge in the Real. Here we have a mark of Lacan’s ambition always to maintain a scientific aim for psychoanalysis. This does not mean however that scientific writing be the same as analytic writing.
In effect, the knowledge in the real can be produced by the scientist “through the semblant of making oneself the subject of this knowledge” [du semblant de s’en faire le sujet]. Whereas the knowledge produced by the analytic discourse results from having unveiled the subject’s own division to him/her.
Let’s note in passing that this is why Lacan can say that the subject in analysis is the subject of science.
Let’s see now what Lacan tells us about the signifier not being of the same register as writing. “The signifier, in the auditory sense of the term, has no relation with what it signifies” (Encore, p. 31). We have here a new, and very significant, definition of the signifier. We know the first definition of the signifier as what represents the subject for another signifier. And from one signifier to another, a part of truth of the subject is unveiled (not to mention the object that falls).
Incontestably, truth speaks. And here we are in the circle of signifiers. To say that signifiers have no relation with what it signifies is to underline ‘the real dimension of the signifier’. The signifier is henceforth two-faced: on the one hand, it is what pertains to the symbolic, part of truth, on the other hand, it is what pertains to the real and can be raised to the rank of a knowledge, to the rank of a matheme of the subject. It is in this Seminar that Lacan said that the signifier is the cause of jouissance. From then on it is the letter in the signifier which begins to function; our aim would be to elevate the signifier to the dignity of the letter; and the signifier can be said to be what is in labour in writing.
We are here in what Lacan expects from psychoanalysis, more precisely from what results from the process of the pass.
Jacques-Alain Miller notes Lacan’s turn in the 70s between ‘the unconscious as a ‘letting the truth speak’ to the unconscious ‘as knowledge’ (see Course Banquet, 1989-90, published in Documents de travail sur la passe), that is to say what can become writing. But here we need to distinguish three types of writing (a distinction noted by Jacques-Alain Miller):
- Scientific writing.
- Psychoanalytic writing, which has a scientific aim but is not produced from the same place.
- Literary writing.
The turn from the unconscious as truth to the unconscious as knowledge is “precisely contemporary with the valorisation of the matheme, and with an emphasis which is no longer on speaking but on writing. This writing is not literary writing. It rightly has to do with the logical form of scientific knowledge. For psychoanalysis it is an adoption of a logic of knowledge” (Cf. J.-A. Miller).
Lacan said it very clearly in evoking [his text] Lituraterre in the Seminar Encore, and in advancing an additional precision: “Writing is a trace where an effect of language is read” (Ibid., p. 110). We would say that this is an articulation of letters, and this is why writing requires a reading. Thousand and one different readings. In the writing of the matheme of fantasy (S <> a), we have two letters and a possible or impossible articulation between them. Indeed everything happens as if speech were to be taken as an articulation between two writings. The writing which is unconscious, which pushes to the reading, and this is what constitutes the work of ‘deciphering’ (there is jouissance in that); and the writing which is perhaps made possible through speech as knowledge to be written, which constitutes a work of ‘ciphering’ (there is also jouissance there). Literary writing is of the order of this type of ciphering, which passes directly in the real. The writing of Mallarmé, that of Joyce, in particular that of Finnegan’s Wake, is the paradigm of this type of writing.
Now, this is what Lacan tells us of his own mathemes: “none of it would stand if I did not prop it up with a speech [dire] which is that of language [langue]” (Encore, p. 110). This amounts to saying that speech and writing belong to a very precise and sophisticated dialectic. This dialectic operates mostly in the analytic discourse and in literature. It does not operate in scientific discourse, since the latter forecloses the subject.
On the contrary, what is written in the analytic cure is what is written for a subject. Here too, Seminar XX is the place where Lacan specifies what this operation of writing consists in. For this purpose he refers to Aristotle’s modalities, giving them his own interpretation. These modalities are:
1 - The necessary as ‘that which does not cease to write itself”. This is what in the symptom and in the fantasy does not cease to repeat itself, that is to say, to try to write itself.
2 - The impossible: “and it is what does not cease not to write itself”. This is what in terms of a logic of knowledge sends us back to the sexual rapport. Or the name of the father for the psychotic.
3 - The contingent: “that which ceases not to write itself”. Something ceases repeating itself for the subject. Hence one of the theses of this Seminar, namely “that the apparent necessity of the phallic function turns out to be mere contingency” (p. 87). Through the analytic operation the phallus “ceased not to write itself” (p. 87). It is the inscription of castration which is in play here.
We can note here that for the psychotic the phallic function is of the order of the impossible, it does not cease not to write itself, whereas for the pervert the phallic function is of the order of the necessary: it does not cease to write itself.
What other writing is enabled by the analytic operation? The one that derives from the contingency of the phallic function. The identifications of the subject can also cease not to write themselves. The subject can cross the plane of identifications, just as he can traverse his fantasy. In certain AE testimonies, one can isolate the point to which the entire neurosis and the entire cure can be reduced. Writing is thus reduced to the writing of two signifiers knotted to an object.
The dialectic of speech and writing thereby finds a perfect illustration.
Having interrogated psychoanalysis and science in their relation to writing, we must return to the relation between psychoanalysis and poetry. So we must return, on the one hand, to Mallarmé, and on the other, to the Seminar Le Sinthome, namely the writing of Joyce.

Four moments in the Seminar Le Sinthome
In this Seminar we find first of all a thesis of Lacan’s which has become a classic.
In a ‘first moment’, Lacan tells us “that, historically, it is through little letters, through little bits of writing that one enters in the real, namely ‘that one ceases to imagine’” (Le Sinthome, lesson of 13 January 1976). It is an important remark because one can say that the neurotic ceases to imagine in passing to the symbolic. We can also say that the neurotic begins by imaginarising the symbolic. We just have to follow the interpretation of any dream to measure what happens, particularly if the dream is a nightmare which openly contains its part of jouissance as Real.
What is Joyce’s objective in this general perspective between R, S and I? Well, Stephen the hero, who is Joyce, apparently proceeds like the neurotic, “he wants to decipher his own enigma” (Ibid.). However, unlike the neurotic, he proceeds through writing, not through speech.
In a ‘second moment’ of this Seminar, Lacan wonders whether Joyce saw himself as a redeemer. To answer this, he adds “We are reduced to feeling because he did not tell us this. He wrote it”, and this is the whole difference, “it is that when one writes one can touch to the real but not to the true.” (Ibid., 10 February 1976).
And this is the major gap which Lacan hollows out between the Real and the true. And we understand well that there is a gap to the extent that the true passes through speech. Through writing, on the contrary, it is the real side of the signifier which operates and not the symbolic side.
In a ‘third moment’, answering a question that had been posed to him, Lacan advances: “Since the real is deprived of sense, I am not sure whether the sense of this real could be clarified in being taken for anything less than a symptom.” (Ibid.)
So the stake of this Real for the subject passes through his symptom. This is the best access he can have to this real. And here we have the major thesis of this Seminar: Joyce’s writing allowed him to keep R, S and I together, in other words to constitute his symptom; since for him it is the imaginary, namely his relation with his own body, that fails.
Thus writing becomes the writing of the symptom, which in the meantime had become sinthome.
Hence in a ‘fourth moment’ Lacan introduces a specification of writing in the literary sense of the term. This is an observation: writing for a writer or a poet ‘is a doing [un faire]’ (Ibid., 11 May 1976) in the sense of a know-how [savoir faire] which is not given to the neurotic. Hence a new remark on the gap between the Real and the true: “The signifier, namely that which is modulated in the voice, has nothing to do with writing” (Ibid.)
Here Lacan did not place the accent on the Real side of the signifier. He did not do that because for him writing in this context is the Borromean knot, “what changes the sense of writing” (Ibid.) And he performs here a kind of inversion which could be surprising. For the Borromean knot “shows that there is something to which one can hook signifiers.” (Ibid.) In effect, the knot is first of all a knot in the psychical [mental]. And one can always hook signifiers to this knot (knot of R, S and I, which is a writing) as: dit-mension ------- mension du dit [of the said] that Lacan extends with “mensionge”.3 Lacan goes on to say that “what this indicates is that the said is in no way inevitably the true.” (Ibid.)
So we are once more finding the gap between the Real and the true; between the dimension specific to the poet-writer, which is writing, and the dimension of the neurotic whose speech can pass through the lie.
This is what Jacques-Alain Miller points out in the first lesson of his 1998 Course (18 November 1998): “The R does not agree with truth”. It obeys laws that have nothing to do with the laws that govern truth. And “to situate the unconscious in relation to the Real is entirely distinct from situating it in relation to the A”.

Return to Mallarmé
Those not familiar with Mallarmé’s biography could misjudge the importance of speech for him. All his lectures and his writings in prose are nourished with this speech. The faithful of la rue de Rome all testify to this. He knows how to make his speech into writing. This is to be distinguished from his poetry properly speaking, which is a writing that can be called ‘mathematical’.
The question of speech and writing will always remain relevant to every analyst. Indeed there is often a necessity to write for an analysand taken by the analyst’s desire: to testify, to transmit, to elaborate a theory. And everyone can see this: writing in the analytic literature (in the strict sense of the term ‘writing’) is really not that great.
That’s because writing necessitates resorting to poetry and to the discourse of science. In short, in order to write one must at the same time be a poet and a man of science. I have thus come to a simple conclusion, which everyone can intuitively guess.
I am left with the comfort of having clarified the question for myself.
Translated by Bogdan Wolf and Véronique Voruz

I. Quotes from Mallarmé have been left in French in the text (in brackets) at the suggestion of the author.
II. The translation of Plusieurs Sonnet II, as well as of other poems, can be found in Stéphane Mallarmé: Selected Poems, trans. C. F. MacIntyre, University of California Press, 1957 [TN].
III. Since le monde is masculine in French, sa [its] should be understood as ‘his’ and so applies both to the poet and the world [TN].
iv. And so echoing with mensonge [lie] [TN].

¹Member of: l’École de la Cause freudienne, New Lacanian School and Escola Brasileira de Psicanálise.