March 8, 1977
We can consider as correlative in any activity whatsoever, in any production whatsoever, two planes or two dimensions of a plane. We can call one of these two dimensions expression and the other content. Why these two terms, expression and content? Because expression, nothing but the word, has the advantage of not being confused with “form”, and content has the advantage of not being confused with “subject”, theme or object. Why doesn’t expression get confused with form? Because there is a form of expression, but there is also a form of content. Content is not without form.
Seminar Introduction
Following publication of Anti-Oedipus in 1972, Deleuze continues to develop the proliferation of concepts that his collaboration with Guattari had yielded. Throughout the 1970s, Deleuze and Guattari’s interest in expanding these concepts continues, eventually producing the sequel, A Thousand Plateaus. In this group of lectures, Deleuze offers the multi-faceted focus of the semester’s work, mainly to distinguish two forms of segmentarity or two types of multiplicities, in seven different directions: 1) biographical, 2) organization, 3) centralization, 4) signification (signifiance), 5) sociability, 6) subjectivation, and 7) planification.
English Translation
In this longer session, Deleuze refers at length to Dominique Fernandez’s book, La Rose des Tudor (Paris: Julliard, 1976) to discuss the idea of music’s death occurring in 1830 (with Bellini’s and Rossini’s death, with the arrival of Verdi and Wagner). Deleuze first traces what is essential in music by distinguishing between expression and content (and not form and subject or object), drawing terminologically from Louis Hjelmslev (form of expression, form of content), and thus distinguishing two dimensions on the plane of consistency (plane of expression, plane of content). Deleuze proposes to consider music as a particular plane of consistency in terms of the particular understanding of expression and content. Deleuze also proposes his and Guattari’s concept of the “refrain” (ritournelle) as something fundamental for the birth of music. He proposes different ways the refrain emerges, e.g. the child’s refrain in the dark, as a line at once of proliferation and abolition. The musical notion enters into a machinic assemblage, whereas in the refrain, the voice may be territorialized until is the pure voice can be extracted as a properly vocal assemblage. Deleuze then translates Fernandez’s notion of movement beyond the differences of sexes with the musical voice, tracing some national shifts toward the voice’s deterritorialization. Deleuze relates Fernandez’s account of the different solutions for the characteristic vocal roles for the child-voiced soprano, in Italian music the castrati, in English music the countertenor for the child’s soprano. The musical process of deterritorialization, Deleuze argues, is inseparable from a becoming-woman, a becoming-child, even a becoming-molecular, and Deleuze suggests that music also becomes symphonic, that is, a change in assemblage with the voice treated as but one element among many others, i.e. as an instrument. To this change in form of expression comes a change in form of content, i.e. of becomings, and to this corresponds a sound plane of consistency, a machination bearing on the voice for the form of expression, and musical becomings for the form of content, both caught in a movement of deterritorialization. Well ahead of the 1981-85 seminars, Deleuze turns toward the vocal problem at the start of speaking films, notably in the American comedy, and returning to music, Deleuze concludes by saying that true music occurs when the refrain is made to undergo a process of deterritorialization, citing Mozart and Alban Berg. To a student’s question, Deleuze reflects on Nietzsche’s views on music, notably on Wagner, and refers to Boulez’s comments on Wagner’s proliferation of sound form. Deleuze also wonders if virtuosity wasn’t a kind of becoming-musical or deterritorialization, but also that contemporary music (e.g. Luciano Berio, whom Deleuze discusses in terms of the refrain) still maintains vocal virtuosity. Deleuze emphasizes convergent and divergent phenomena, such as inclusion of oriental music, citing Boulez in this regard, and then continues developing his perspectives on the abstract musical machine, distinguishing different types of concrete machines. Deleuze adjudicates something of a debate between Georges Comtesse and Richard Pinhas, pushing back against Comtesse’s additional assemblage of death or castration, since Deleuze (with Boulez) considers silence as a creative element within the musical machine. Unfortunately, the subsequent “long discussion on the death instinct”, indicated in a final WebDeleuze note, is omitted from the transcript.
Gilles Deleuze
A Thousand Plateaus II, 1976-1977
Session 2: March 8, 1977
Transcript: WebDeleuze; transcript modified by Charles J. Stivale
On Music, the Refrain, Haecceities
Translation, Billy Dean Goehring
Deleuze: Last time we talked about a book by Dominique Fernandez. He has a few important things to tell us about music, so I’m taking a step back. It’s unusual because the stuff he normally does is more along the lines of literary criticism based on psychoanalysis. And then, at the same time, he loves music, and so that draws him away from his psychoanalytic concerns. He forwards a claim that runs throughout his book, which is titled La Rose des Tudor.[1] The whole point of the book is this: music dies out around 1830. More specifically, it dies—and tragically, like all good things—it dies with Bellini and Rossini. Tragically, because the circumstances surrounding Bellini’s death are unclear; he either died from a then-unknown illness or from some sinister plot. And with Rossini, it was an abrupt end. A brilliant musician at the height of his career, and he up and quits. He had always had two loves: music and cooking. He was an excellent cook, and he went crazy. I know a lot of people who cut things off at a certain point; it’s a sort of claim you find often enough: “For me, things come to an end at such-and-such point in time.” Philosophy has always been dying: it died with Descartes, it died with Kant, it died with Hegel—it depends on who you ask, but what matters is that it died. And then I know some who say music ends with Gregorian chanting. Sure.
That’s the sort of argument Fernandez is making: music ends with Bellini and Rossini. What makes such a claim possible? That can only mean one thing: something, whether you realize it or not—I’m not going to defend it, because I don’t think he’s right—something that used to be essential to music disappears after Rossini and Bellini, the two last musicians. What is it that, perhaps even indirectly, led to Rossini and Bellini’s demise? What was different about music starting around 1830? It was the emergence of Verdi and Wagner. In other words, Wagner and Verdi made music impossible. Fernandez goes as far as to call them fascists. It isn’t the first time that’s been said of Wagner.
What, according to Fernandez, did they get rid of that was so essential to music? This is basically what he tells us—he says there used to be something inherent about music. Let me step in and explain something: we can think of any activity, any form of production, in correlation to two different planes[2] or dimensions. One of these dimensions we could call expression, and we could call the other one content. Why expression and content? Because the word itself, “expression,” has the upshot of not being mistaken for “form,” while the benefit of using “content” is that it won’t be confused with “subject,” “topic,” or “object.”
What keeps us from confusing “expression” with “form”? It’s because there’s both a form of expression and a form of content. Content isn’t without form. But what is it? Adding to everything we talked about previously, I could argue that it isn’t that what we’ve been referring to as the plane of consistency comprises two chunks, but that from one angle it’s a plane of expression and from another angle it’s a plane of content.
Considering the sonorous[3] plane of consistency called music, I could ask what specifically musical expression and content might be, assuming said content isn’t what music is written about or what a voice sings about. Now, Fernandez claims that, as he sees it, music has always been permeated by content he’s particularly close to, which is that of spilling over or overcoming sexual difference. He hasn’t forgotten his psychoanalytic background—though he isn’t an analyst—he argues that music is essentially always a restoration of androgyny. If I attribute that content to music, that implies that I can show that such content really is musical, that it’s essentially musical, by virtue of the music’s form of expression. Now, it’s no secret that music is primarily vocal. We know the extent to which instruments have long been the object of surveillance, notably with the codification of music, the Church’s effort in the codification of music. For the longest time, instruments were kept out, held back; they shouldn’t be allowed to overwhelm the voice.
At what point does a voice become musical? Looking at it in terms of expression, I’d say that, essentially, the musical voice is a deterritorialized voice. But what does that mean? I think there are things that aren’t quite music but are very close to being music. There are types of singing that aren’t quite music; for example, Guattari attaches a lot of importance to a concept we’ve yet to flesh out, that of the refrain.[4] The refrain could perhaps be an integral component in the emergence of music. Little refrains will later be taken up again in music. Non-musical singing: tra-la-la. A frightened child? Perhaps the little refrain originated in what, last year, we called the black hole. A child in a black hole, comforting themselves, tra-la-la. I argue that this kind of humming voice is territorialized; it marks out territory. Which is why, if music goes back to the refrain, one of the most characteristic examples of the refrain’s reprisal is Mozart. [Alban] Berg uses this approach all the time. What topic is the most profoundly musical, and what makes it the most profoundly musical? The death of a child—and not a tragic death, a happy death: “Concerto to the memory of an angel” [1935]. Children and death are everywhere. Why? Why does such proliferation and abolition run throughout music, a line of both proliferation and sonorous abolition?
If the humming voice of the refrain starts out as a territorialized voice—if only in a black hole—music starts with the deterritorialization of the voice. The voice is “machined.” Musical notation fits into a machinic arrangement [agencement]. It forms an arrangement itself, it forms an arrangement in itself, whereas with the refrain, the voice is still territorialized because it’s working with [elle s’agence avec] something else. But once the voice is isolated in its pure state and forms a strictly vocal arrangement, it emerges as a deterritorialized sonorous voice.
What does this deterritorialized voice entail? I’m trying to put what Fernandez says in my own words, when he says that the voice in music is meant to overcome sexual difference. I claim that the sexes, with their particular vocal sonorities, territorialize the voice: ah, this is a woman’s voice, that’s a man’s voice. Deterritorializing the voice: there’s a key moment with musical notation where it’s plain to see. At first, European musical notation essentially had to do with voices. One of the most important considerations here is the twin roles played by both popes in Latin-speaking countries (such as Gregory) and Henry VIII, and the Tudors, when it comes to musical notation. Henry VIII was the one who insisted that each syllable ought to have its own corresponding note. This is not simply, as they say, to ensure that the text is properly understood when sung; it’s a powerful factor for deterritorializing the voice, a key factor. If you make your musical notation account correspond to every syllable, you have a method for deterritorializing the voice.
But notice that we haven’t quite drawn the connection between my claim that, in terms of expression—and as a form of expression—music is defined first as vocal music, and [the idea that] vocal music is the deterritorialization of the voice when it comes to content, as a form of content. When it comes to its form of content, I define music—vocal music, at least—in the same way Fernandez does, not as a return to primordial androgyny, but as the overcoming of sexual difference. Why is the deterritorialized voice, from the point of view of expression, the same as the overcoming of sexual difference from the point of view of content? Looking at it in terms of expression, this deterritorialized voice, once it’s been arranged, having arrived at its own arrangement, arranged on its own, machined on its own—is the child’s voice.
What do I mean by that? It’s true that all music, up to a certain point in time, Fernandez tells us, a sort of sexual subversion permeates music. It’s clear to see with Monteverdi. And whether it’s the Italian/Spanish type of Latin music or whether it’s English music—something like the two ends of the spectrum in the West—what are the key voices in vocal music? The main voices in vocal music are soprano, alto, and what the English call countertenor. The tenor is the one who holds the line, and then there are higher alto, soprano lines. But these are children’s voices, voices intended for children. One of the high points in Fernandez’s text is his indignation that women became sopranos. He’s furious; it’s terrible. That’s only possible once music’s dead. The child is the soprano—not the natural soprano, but the soprano in the musical scheme of things [agencement musical].
All three typical voices, children’s voices: in Italian music—this is shared on either pole—in Italian music, there’s the castrato, the castrated singer, and in English music, which oddly enough, didn’t have castrati (castrati were a Latin development), you have the countertenor. And regarding the child soprano, the castrato and the countertenor are like two different solutions for the same problem. The English countertenor is still around, while there aren’t any castrati anymore—thanks to civilization, Fernandez says, it’s capitalism’s fault, he’s so unhappy. With Verdi and Wagner, music gets appropriated by capitalism.[5]
What makes the English countertenor a deterritorialized voice? It comes down to singing above than one’s own voice. The countertenor is often described as a head voice. It’s about singing beyond your voice, and it’s a real act of deterritorialization, what [Alfred] Deller says is the only way to sing high. It’s a voice that doesn’t use the lungs. It’s a great example of the voice’s deterritorialization because the territoriality of the voice is sex: male voice / female voice.
But I could just as easily say that the little refrain is where you speak; I could also say that the diaphragm-lung system is where it emits from. Yet that’s how the countertenor’s voice is described, as if it starts from the head. Deller insists that it has to go through the sinuses, that it’s a sinus voice. Deller’s story is a beautiful one: at the age of sixteen, like all choir schools, he’s told to let his voice rest for two years, and he comes out as pure countertenor. It’s funny, for anyone who’s heard Deller, his voice comes across as artificial and overworked, and at the same time, as a sort of raw musical material, as though, through its artifice, it’s both the most artificial and the most natural. So, the voice comes from the head, goes through the sinuses, without every going through the diaphragm. That’s what makes a countertenor. You can more or less see, even if he doesn’t like the term, what makes it a head voice.
The castrato’s voice is different; it’s a voice that’s just as completely deterritorialized. It’s a voice from deep in the lungs, from the belly, even. Fernandez’s description is rather good. Purcell, the great child musician, also has a great backstory: as a child, he was a soprano, and then was later capable of singing both as a bass and as a countertenor. Hearing Purcell sing was incredible. Twice in his book, Fernandez attempts to pin down the difference between the castrato route and the countertenor approach, the English route. “Here we ought to examine the fundamental difference separating the art of singing in England from the art of singing in Spain. A countertenor’s voice is in their head, giving the impression of almost unreal celestial purity, not devoid of sensuality but with a sensuality burning with the lust it ignites. Sopranos and altos have their voices seated much lower in the chest—you’d almost believe it were in their belly—closer to their genitals, at any rate. We assume that castrati had such an irresistible effect on their listeners not only because their voices were among the most beautiful, but also because they were charged with an intense erotic power. All the sap lacking other outlets in their bodies impregnated the air they expelled from their mouths, with the effect of transforming what’s usually airy and impalpable into a pulpy, gooey substance. [Laughter] Whereas English countertenors ignore the fact that they have or could have a sex, Italian castrati make their song a complete and carnal act of expulsion, symbolizing the sexual act whose painful and voluptuous restlessness their voices betray. The sounds leaving their throat possess an ow, ow consistency—these boys make love with their voices.”
The takeaway here is that these two methods for deterritorializing the voice—the countertenor’s head voice, head-sinus-mouth, without relying on the diaphragm, and the castrato’s voice from deep in the lungs and belly—what does that get us? It’s easy to see here how the voice’s musical arrangement [agencement], the musical process of deterritorializing the voice, effectively amounts to a kind of overcoming of sexual difference. Using our terminology, we might say that music is inseparable from becoming-woman and becoming-child. Becoming-woman as fundamental to music, which isn’t… Why is music so preoccupied with children? My answer would be that, far beyond these themes, these motifs, these subjects, these references, it drills down to the content of music, and what defines properly musical content is a becoming-woman, a becoming-child, a becoming-molecular, etc., etc.
What about this becoming-child? It’s not that music comes down to singing like a child or making your voice sound like a child’s voice; the child might be totally fabricated. You almost need to distinguish between the molar child singing non-musically, the child of the refrain, and the molecular child who is musically situated. And even with a child singing in an English choir, it takes a musical artificial process for the molar child to stop being a molar child in order to become a molecular child. Thus, the child possesses a musically becoming-child. Meaning that the child of music’s becoming, the one that it makes, itself has a deterritorialized child for its content, just as the deterritorialized voice is its expression. It’s not about imitating a singing child; it’s about producing a sonorous child, i.e., deterritorializing the child while deterritorializing the voice.
Here we have the junction between music’s form of content—becoming-woman, becoming-child, becoming-molecular—and music’s form of expression—deterritorializing the voice, among other things, through the interplay of melody and harmony, through polyphony, and even through instrumental accompaniment. But at this level, music is still vocal, essentially, since as a form of expression, it’s defined by the deterritorialization of the voice, in relation to which instruments only play a supporting, accompanying, or concomitant role. And on the other hand we have this becoming-child, this becoming-woman, and as we said last time, the child themselves who needs a becoming-child. Being a child isn’t enough for becoming-child; it involves the entire choir training or the English cathedral, or worse, becoming-child requires the Italian castrato approach.
Bellini and Rossini were the last ones to musically arrange [agencer] the voice using these becomings, becoming-child and becoming-woman. What disappeared at the start of the 19th century was, on the one hand, the castrato tradition—I’m intentionally avoiding saying “castration”; if I call it castration we’ll land right back into psychoanalysis—the machinic arrangement of the castrato doesn’t lack anything. The castrato is within a becoming-woman that no woman has, a becoming-child that no child has. By that same token, he’s caught up in a process of deterritorialization. By necessity, becoming-child is not becoming the sort of child that a child is, but becoming a child qua deterritorialized, which is carried out through a means of expression which is necessarily itself a deterritorialized expression: the deterritorialization of the voice.
Fernandez has some fairly reserved but noteworthy praise for [David] Bowie. He says he has a falsetto voice. But it isn’t a coincidence that pop music came from the English. The Beatles: there ought to be some voices that aren’t far off. Not countertenor, but one of them should have something approaching a countertenor range. It’s odd that the French didn’t take to castrati. It makes sense for the English; they’re puritans. When Gluck was running some opera or other in France, he had to completely rewrite the lead role to have it sung by a tenor. So dramatic. We’ve always tended towards the refrain. So, Fernandez pays this sort of compliment to pop music. But you can see what he’s getting at when he says that music ends with Bellini and Rossini. Again, it amounts to saying: down with Wagner, down with Verdi. It isn’t as good now.
All I want to get out of Fernandez’s book is this: music is inseparable from a becoming-child, a becoming-woman, a becoming-molecular. That’s its form of content, even, just as its form of expression is the deterritorialization of the voice. And the deterritorialization of the voice ranges from the deterritorialized voice of the castrato to the deterritorialized voice of the countertenor. It forms a little block. It comes down to machining the voice, to a vocal sound machine, which entails deterritorializing the voice as far as expression is concerned, which, with regard to content, correlates to becoming-child and becoming-woman, etc.
Indeed, at first glance, Verdi and Wagner take us back to a sort of immense molar reterritorialization in our language, i.e., however sublime their voices may be, Wagnerian singers are men with men’s voices, women with women’s voices. It’s the return of sexual difference. They snuff out music’s becoming. You can see why Fernandez blames that on capitalism—he claims that capitalism can’t sustain sexual difference, there’s the division of labor. In other words, instead of the voice being machined into musical arrangements, the deterritorialization of the voice becoming-child, it sort of gets put through the ringer: the binary machine—the woman’s voice responds to the man’s voice, the man’s voice responds to the woman’s voice. Tristan and Iseult. You know that in old opera characters like Caesar were sung by castrati. By no means were castrati intended for mincing about or for flourishes; an all-powerful Caesar or Alexander was supposed to overcome sexual difference, to the point where there’s a becoming-woman to the warrior. Achilles was sung by a castrato. Indeed, Achilles has a becoming-woman.
It ought to be forbidden to talk about what one doesn’t like. There ought to be an absolute ban on it. We always write for/about the things we love. Truly, any literature that isn’t about love is shit. Fernandez is discreet; he says very little about Verdi and Wagner, but I think something else is going on: music is becoming symphonic. It doesn’t stop being vocal when it needs to be, but it starts to become symphonic. A low point in Fernandez’s writing is when he claims that the instrumental development is what forces voices to become male and female again, putting them through this kind of binary machine. Actually, countertenors are screwed when it comes to symphonic ensembles. He seems to be saying that instrumental or symphonic music makes too much noise, too much noise for any of these subtle becomings to be perceptible. We might picture it completely differently.
What happened once the voice was deposed like this? What happens when the musical machine is no longer mainly vocal, when the instrument is no longer merely an accompaniment for the voice, and music is instead instrumental and symphonic? I think what really changes is the musical machine, or the layout [agencement] of music. It’s no longer a matter of arranging the voice, but of handling the voice, and that seems like the main revolution in music. It’s a question of handling the voice as one element among others, as having its own specificity as one element among others in the instrumental machine. Flutes or violins are no longer there to accompany the process of deterritorializing the voice or to make this process possible. The voice itself has become an instrument, no more or less than a violin. The voice is put on equal footing with instruments, so that it no longer holds the key to musical arrangement. The whole arrangement is turned upside down. I’d say it’s a genuine transformation.
No longer is it about finding or inventing a voice machine; it’s about raising the status of the voice to that of an element in a symphonic machine. It’s completely different. Unsurprisingly, Fernandez is right, only from a rather narrow point of view: Verdi and Wagner usher in a reterritorialization of the voice, and that continues with Berg (“Lulu”). But that has to be the case, since what makes the voice a musical element is no longer that it’s a voice as such. Such that, if we consider as a voice, we revert back to its pseudo-natural determinations: male voice or female voice. We revert back to its binary machine since its place in the musical machine is no longer as a voice. As a voice, it effectively then falls back onto sexual difference, but that’s no longer what makes it musical.
The huge advantage of symphonic instrumental music is that, instead of proceeding by via a simple sonic machination of the voice, it carries out a generalized sonic machination that no longer treats the voice as anything other than an instrument on par with any other. So much so that, again, when you consider voices as voices, they revert to their natural or territorial male-female determinations. But at the same time, that isn’t what makes them musical; they’re musical through an entirely different lens, based on how they’re related to the instruments whose equals they are, in their overall machination, where, ultimately, there’s no longer any difference in kind between the sound of a flute and a voice’s timbre. We’ve transitioned to a new type of arrangement. I’d almost argue that music’s form of expression has changed—instead of a machination of the voice, you have a symphonic machination, an instrumental machination in which the voice is just one element among others.
But then its form of content changes too, and you get a change in their becomings. The form of content is still becoming, but you’ll find it impossible to recapture what was so essential about purely vocal music, namely, its becoming-woman and becoming-child. What you’ll find is that it opens onto other becomings. Before, musical becoming ended with becoming-woman and becoming-child; they were mainly becomings that more or less fell off at a borderline, that of becoming-animal, especially becoming-bird. Becoming is a constant throughline, i.e., musically manufacturing a deterritorialized bird. To put it plainly, deterritorializing the bird means snatching it out of its environment. Music doesn’t reproduce bird songs; it produces a deterritorialized bird song, like with Mozart’s bird, which I’m always bringing up.
Now, this new instrumental or symphonic music may have lost its grip on becoming-child or becoming-woman—things have changed—but has instead opened onto other becomings, as if animal-becomings were sort of unleashed, uniquely sonorous, uniquely musical animal-becomings, elementary power-becomings, elementary becomings—Wagner [and] the idea of continuous melodies, which is a sort of form of expression whose corresponding form of content is the unleashing of elements, of elementary sonorous becomings.
Ultimately, opening onto something which, in my opinion, simply does not exist in vocal music, but which the voice can take up in this new music, in this new musical arrangement: molecular becomings, unprecedented molecular becomings. I’m thinking about the singers in Schoenberg. It’s already there in Debussy, in all modern music. [Luciano] Berio. In “Visage,” it’s clear that the face is only involved insofar as it’s undone. There’s the whole field of electronic music, where things are headed toward molecular becomings only possible with the Verdi, Wagner revolution. So, I’d argue that the form of musical expression is changing and that, as a result, its form of content opens up different types, different genres of becoming.
As for a very general definition of the sonorous plane of consistency or musical machine, how would we define it? I’d say that if we look at it in terms of expression, its form of expression always consists in machination, either directly machining the voice or a symphonic machination integrating the voice as an instrument, and then looking at this same sonorous plane of consistency from the content side of things, there are always uniquely musical becomings that never boil down to imitation or reproduction, all the becomings whose developments we’ve been following, and there’s the theme as form and form of content: both are caught up in a process of deterritorialization.
I’m wondering if the same thing goes for cinema. Someone was working on this last year, on sound films. Couldn’t you argue that in the early days of sound film, the voice wasn’t as individualized? It wasn’t really an individuating factor. Take American comedy, for example. It’s as though the voice’s individual characteristics were superseded. Sound film only adopted the voice in order to overcome the voice’s individual characteristics. Ultimately, with the birth of sound film came an individualization by face or by type, and the voice, as a determining factor in so-called sound film, goes beyond particular or even specific determinations. It’s only later that stars get recognized by their voices—Marlene Dietrich and Greta Garbo.
Now, there’s no voice in American comedy, and yet sound film has one use that’s fantastic, but voices aren’t distributed using binary machines or using individuating machines. What’s so great about Bogart’s voice? It’s that his voice isn’t individuated at all; it’s a completely linear voice. What made Bogart’s voice so successful was that it was blank; it’s the countertenor of film. It’s a blank voice that’s rather rhythmic, one that, strictly speaking, doesn’t go through the lungs. It’s a linear voice that comes from the mouth.[6] When music is vocal, voices don’t show up as individualized voices, as gendered, male/female voices; they serve as form of expression for a becoming, becoming-woman, becoming-child. In the same way, sound film began to use the voice as a form of expression for becoming.
We’d also have to define the fixed plane [or “still frame”]: just as there’s a sonorous plane of composition bound up with the musical machine and all this machine’s becomings, and the musical machine’s becomings are what run through the sonorous plane of consistency—well, the musical machine could just as well be called a sonorous still frame. But “fixed” means absolute speed as much as it means absolute slowness or rest; it’s the absolute of motion and rest. And the becomings inscribed on this plane are relative motion, relative speeds and slownesses, right, in the same way that cinematographic still frames could also be described as absolute motion and absolute rest: cinematic forms of expression and the role of the voice in sound film are inscribed on this plane, as well as the corresponding becomings based on changes in the forms of expression with new forms of content. I’d love to hear what you think.
The child’s game of playing fort-da with his spool isn’t at all what psychoanalysis thinks it is. It has nothing to do with a differential opposition between signifying elements. It’s something else entirely: it’s a little refrain. The little refrain of territoriality. By no means is the spool game a binary machine; there are so many in-betweens. It’s not a phonological opposition; it’s a refrain.[7] Real music starts when you take your little refrain and deterritorialize it, putting the refrain through a process of deterritorialization. Which is what Mozart is constantly doing. It’s what Alban Berg’s “Concerto in Memory of an Angel” is: a deterritorialized child. A dying child, and the conditions for bringing about the child’s deterritorialization.
I’d like for us to take what we’ve done thus far and take an overall review of it, recentering on a vary particular type of plane: the sonorous or musical plane of consistency, and the musical arrangements that get drawn onto this plane of consistency.
Student: What does Nietzsche think of Wagner?
Deleuze: It’s a funny story. It’s impossible to just read it as literature, however beautiful it may be. It’s well known that Nietzsche himself made music, and everyone agrees that, with a few exceptions, it wasn’t very good. This isn’t a famous remark. Nietzsche put all his musicality into his writing; that’s Nietzsche the musician. What’s interesting is how much and how often his music resembles Schubert, or Schumann. I’m begging you: go and listen to Nietzsche’s tunes at nightclubs. What’s Nietzsche’s problem with Wagner? He says his music is watery, that you can’t dance to it at all, that it isn’t music but morality. He says it’s full of insufferable characters: Lohengrin, Parsifal.
What is he implying, more or less? There’s a certain way of understanding the plane / plan where the forms you find are always still developing, however developed they may be, and subjects are still forming. Bringing it back to music, I’d argue that Wagner completely innovates the domain of musical forms; as innovative as it may be, it still has an aspect of form development. Boulez was one of the first to highlight his proliferating form; he credits Wagner for how he’s continuously developing form, which was relatively new. But no matter how novel his approach to development was, it’s still a development of sound forms. Hence there’s a necessary correlation between the development of sound forms and how subjects are formed. Lohengrin, Parsifal—Wagnerian characters are those in apprenticeship, the notorious German theme of education. There’s still something Goethean about Wagner. The plane of organization is defined by the two coordinates of the development of sound forms and the formation or training of musical subjects. Nietzsche contributes to an entirely different approach to the plane.
When I said that there were only two things the plane of consistency understood: it no longer registers developing forms; all it knows are speed and slowness, motion and rest. It only knows the speeds and slownesses between particles, between molecules. It no longer recognizes forms in the process of developing. It only knows speed differentials between elements. It doesn’t know anything about the development of form. I’d add that, consequently, the same goes for the training of a subject; sentimental education is over. Wagner is still all for sentimental education. The Wagnerian protagonist says, “Teach me to fear.” Not so, for Nietzsche. There are only haecceities, i.e., combinations of intensities, intensive compounds.
A haecceity isn’t a person; it isn’t a subject. When I think about Nietzsche, this is right where I think he’s coming from. What’s so great about Ecce Homo? It’s not much of a stretch to say that Nietzsche is someone who makes it a point to tell us that there’s nothing but speeds and slownesses.[8] They all pay homage to Goethe, but they’re really sneaky. Hölderlin and Kleist pay homage to Goethe, but that’s still just their pure hatred. Nietzsche doesn’t say, “be fast”—he himself wasn’t very fast. You can be rather fast while moving slowly; again, it comes down to the differential relationship between speed and slowness. You can be very fast without moving; you can go on incredibly fast trips while standing in place, and then be back before you’ve left. Ecce Homo is incredible; it’s one of the most beautiful books ever—how Nietzsche talks about seasons, climates, diets. He’s always saying: I’m not a person, don’t treat me like a person, I’m not a subject, don’t try to shape me. That’s what he says to Wagner; he says that it’s music for Wagner. He’s not interested in sentimental education. What he’s interested in are haecceities and compounds of intensities, and he views himself as a collection of haecceities.
I think that this is what Nietzsche achieves in his writing, the disappearance of any apprenticeship or education in favor of spreading haecceities. When he says that Bizet’s music is much better than Wagner, he means that there’s something that stands out in Bizet’s work that is later more successfully pulled off by Ravel. And this something is the liberation of musical speeds and slownesses, i.e., what after Boulez became known as the discovery of non-pulsed time, as opposed to the pulsed time of developing forms and forming subjects—a floating time, a floating line.
Richard Pinhas: Still, at one point Nietzsche’s preference for Bizet is troubling, a difference that completely vanishes with Ecce Homo, where he completely turns around on Wagner, ultimately saying, “I love him,” for a while; during his big falling-out, there’s a sort of complaint he leverages at Wagner, and he frames Bizet as the positive creator of his age. It’s a problem with the melodic line: Wagner supposedly messes up the melodic line, and what he likes about Bizet is how predominant his melodic line is. At the same time, he accuses Wagner of being a rhetorician and theater artist; his terms, and they’re precisely the terms able to describe subjectivity and the creation of the subject. But it isn’t clear to me how Bizet went any further than Wagner; it isn’t obvious. The issues with melodic lines are ambiguous in Nietzsche, and in certain respects, despite how much I love and admire him, it might be a step backwards compared to the innovative benchmark found in Wagner. It remains to be seen.
What I found so interesting about what Gilles is working out is how quickly he divides things up. There are lines of transition, or even a wide variation of planes, when it comes to musical becoming in general. At one point, he posed the question: why haven’t we been able to hold onto countertenor or castrato voice? They’re disappearing. There’s an easy answer: at a certain point in musical becoming there’s absolutely no need for it, i.e., once a musical plane of composition or musical plane of consistency opens onto or is oriented toward a new method of sound production or creation, a method that applies as much to writing as it does to the materials or arrangements used.
I’ll use a concrete example: what would it mean today, what would be the point, what use would there be for the sort of virtuosos we used to train in the past, when it comes to playing the music of today’s composers? There’s no place for that anymore. Musical composition no longer demands that level of performance. While virtuosity was a necessary factor in composition a century ago, today that element has completely disappeared. So, we’re simultaneously witnessing the creation of new forms, new arrangements, new developments, new materials—so much is happening all at once. We’re witnessing the loss—not even explicit rejection, but the fatigue or exhaustion of certain former components, virtuosity in this case. Really, you might say that there’s nothing left for us to do with virtuosity.
Deleuze: Could we say—this might go against what you’re trying to say—that virtuosity was a deterritorializing technique specifically tied, not to all musical becomings, but to music’s becoming-woman and becoming-child? Throughout its history, very particular types of animal-becoming have always been a part of music.
Claire Parnet: We can assume that the most deterritorialized becomings always happen through the voice. Berio.
Deleuze: Berio is a striking example. It’d go back to the idea that the virtuoso is disappearing, as Richard brought up music’s machinic evolution, and that, from then on, the problem of musical becoming is much more about molecular becoming. It’s easy to see how, in electronic or synthesizer music, virtuosos are dispossessed, in a way. Which doesn’t stop music as modern as Berio’s, which uses all these processes, from holding onto virtuosos and maintaining a vocal virtuosity.
Pinhas: I see that as the persistence of a code, an archaic code; it’s part of what makes Berio’s composition innovative. Still, it’s a funny way of treating this voice.
Deleuze: I’d agree with you there, because Berio inserts all sorts of refrains into his work. I’d defined the refrain by its contrast with music, deterritorialized voices or instruments. The refrain is sonic territorialization, as opposed to music as such, which is the process of deterritorialization. Yet, just as there are becoming-women, becoming-children, becoming-animals, there are becoming-people: hence the significance of all these folk themes in music. Little folk tunes fall under the refrain; little tunes from any given region, which a musician literally takes, snatches, and—what’s more—transforms both its expression and content, sometimes leaving a phrase intact—the extent of its transformation can vary. Now, with Berio, he makes use of popular folks songs from every country; if necessary, he inscribes them in a multiple language, and in that way, he has a sort of vocal virtuosity. I’m interested in the little refrains of children or women, and in the deterritorializing machine that takes it and puts the voice or instrument, the folk song, through a special treatment, to the point where Verdi gets associated with the Italian revolution. That explains the connections. Verdi becomes the genius of nascent Italy.
Pinhas: From what you’ve said, I’d lay out four basic periods—there aren’t any real divides, strictly speaking, but there are variations and transformations, translations leading to new planes of musical composition. The first, not in time, but in relation to what we’ve discussed, ends at Rossini; the second stops when Debussy and Ravel come in; the third, coincidently, more or less lies with the aftermath of World War II; the fourth would be the musical forms found today, both in commercial pop music and in terms of what’s thought of as avant-garde work. Allegedly contemporary music.
For the first and second periods, we find extremely close connections when it comes to content figures, between becoming-animals and becoming-children and becoming-women—in the first case especially, a becoming-child and a becoming-woman, the same in the second case, with another dimension of reformation unique to examples like those found in Wagner. And from Debussy and Ravel onwards, on the one hand we have molecular becomings and a particular relationship to becomings we haven’t defined yet, to “terrestrial” materials. When Ravel titles a piece, “La Mer,” on the one hand there are molecular becomings, on the other, a particular ratio of elements.
Then, there’s music today, which for me is primarily molecular, abstract. In the first two categories, or series, it’s legitimate; all you can do is bring in an analysis referring to figures of content and figures of expression, and with that, let’s say that my question from a few weeks ago has been fully answered, as I have the impression that, starting with Ravel and Debussy, the figure of content gives way to something that, of course, could also be called a figure of content, but which would be much closer to singular type of arrangement which would replace these figures of content, at least so far as any analysis is concerned, and which would split the figure of expression into a figure of expression proper and lines of effectuation, whether material effectuations, writing effectuations, performance effectuations, or what compound affects are found. It’s not mutually exclusive. It doesn’t invalidate figures of content or figures of expression; it merely develops them. It seems to me that in today’s music, especially with English and American composers, there’s practically no content left; instead, what we’re witnessing is a sort of generalization of molecular becomings.
Deleuze: But becoming is content like any other—molecular content.
Pinhas: Yes, but so long as it’s general, it’s not really something that you can get at in terms of analysis. But obviously, it is [content]. What I’m focusing on is how singular arrangements are a form permitting us to develop the term “figure of content.” And I see how, when it comes to contemporary music, what’s going to happen—and you can basically pick it out in any given country or any given trend in composition—is the assertion of extremely differentiated and elaborated times. For example: of course, we have two basic categories—pulsed time and non-pulsed time—but within these categories, or alongside them, we find that English music and some American music—I’m thinking of La Monte Young and sometimes Steve Reich—is a kind of music that references or builds a metallic time of execution and affection, hence composition, that we get a non-pulsed metallic time; on the other hand, some Americans like Philip Glass use pulsed metallic time, as well as other forms of time that’d fall under that same heading; while the Germans use a time that’s just as abstract as the others, but a mechanical type with very precise rhythmic inscriptions.
In France, I’m thinking of a group called Magma; you see a comeback with a time of war, which isn’t at all some sort of despotic hierarchy of sounds; it’s a time that, in the context of its execution, has a totally innovative aspect to it, and you find tons of other times: live times, instant times show up in these sorts of music. On the other hand, in pop music, what we see is a sort of holdover, a sort of return to something that leaves me rather perplexed—something that plainly counts as figures of content, something that stands in for a signified, but which isn’t really a signified. The best term to describe it would be an abstract icon. An abstract icon would be something that doesn’t represent anything, but which acts and functions as an element of representation. So, we’re finding something like that again.
Deleuze: A quick note, Richard. In these voices, in this sort of vocal machinery, in pop music—Fernandez isn’t wrong when he says that there’s also a voice that goes beyond the gender binary machine. It’s not just Bowie; it’s also the Stones, and Pink Stuff [Pink Floyd, perhaps]. Would you agree with that?
Pinhas: Yes, except that it doesn’t really seem pertinent since referring any voice to the problem of sexual difference is such a despicable move.
Deleuze: Now you’re pulling my leg. You can’t be serious. If we say the binary machine, man/woman, is a territoriality of the voice, the milieux, the sexes, the types of refrains and the parts of the body involved, the lungs, the throat, the diaphragm—it’s all mixed together. That’s what I’m calling the territorialized voice, with the refrain as its musical form. I claim that music begins by deterritorializing. So, as I see it, music’s constitutive process of deterritorialization—you’re right to say that [sexual difference] has nothing to do with music, since it’s only with the deterritorializing process that music begins. There is no music except through the deterritorialization of the voice.
So, [in the two] techniques for going about the process of deterritorializing the voice—castrato, countertenor—the two aren’t at all identical, they aren’t the same parts of the body, they aren’t the same milieux. So, there are methods for deterritorializing the voice which are integral to vocal music. And then there are deterritorializing processes that are strictly instrumental, which make the voice into just another instrument. I’d argue that all becomings first happen through the voice. Looking at how this arrangement unfolds: what I’m emphasizing is how the artificial/natural dichotomy gets replaced by the difference between territoriality/deterritorialization, because, ultimately, there isn’t anything natural or anything artificial.
Student: [Inaudible remarks about anachronisms]
Deleuze: Completely. Every deterritorializing process also produces more or less artificial reterritorializations. When instrumental music, when instruments take precedence over the voice, the voice itself becomes a reterritorializing factor, whereas before it was essentially caught up in a movement of deterritorialization—it was even an agent of deterritorialization.
Student: Is Bob Dylan really [an instance of] deterritorialization?
Deleuze: Yes, yes. What is Dylan’s voice, musically speaking? It’s kind of a blank voice. It’s very odd. It’s increasingly nasal.
Pinhas: What you were saying earlier about the use of archaisms is pretty important because, once you use an anachronistic element and fold it into an innovative perspective, what you end up with is even more powerful. And at a certain level, the use of binary structures, which began in contemporary jazz with Miles Davis, is the advent of American neo-binarism. It picks up one of the most territorialized elements used today, the drums; it’s what cuts musical times in seconds or thirds, using conventional standards. And what does he do with this most territorialized element? He invents, or reinvents, proliferating compound beats, to the point that, ultimately, with the help of this “old” or highly coded event, he creates a sort of quasi-absolute deterritorializing line at the level of rhythmic structures.
Deleuze: I think there are convergent and overlapping phenomena. Steve Reich says what all he owes to Eastern cultures, but that doesn’t mean he didn’t come to them after a convergent process that went through Eastern music. I’ll quote Boulez: “Tempo is due to a written numerical relationship, but it’s completely modified and carried by a rate of unspooling. Accounting for this phenomenon, it was much easier to get extremely complex relationships using intrinsically simpler relationships, adding modifications in speed to these numerical relationships. If you incorporate it into a fairly simple rhythmic structure”—with regard to form—“accumulations of little notes”—this is already there in Mozart—“the accumulation of little notes making it possible to produce complex relationships between speed and slowness based on formally rather simple relationships, at every turn, you get a broken tempo. Thus, there’s music that can do without pulsation altogether, a floating music, where the writing itself makes it impossible for the instrumentalist to keep in line with a pulsed time. The little notes, the ornamentation, the multiplication of dynamic differences.”[9] There are some critics who talk about these little notes in Mozart as “blocks.” We should also look for these little blocks in Debussy, which literally break down the form’s development, and against the backdrop of a relatively simple form, they generate extremely complex relationships between speed and slowness. Which is precisely what Richard was saying.
Pinhas: Yeah, basically?
Deleuze: Basically? Basically? Alright.
Pinhas: I’m basically saying, not with regard to what you’re saying or your interpretation of Boulez, but with regard to Boulez’s text itself, which is always ambiguous—it’s often right, but it’s ambiguous.
Deleuze: Ambiguous? I’d like to hear what you think about this business with the voice in sound film. The parallel I see? If we accept the idea of a musical machine—musical machine meaning what takes place on the sonorous plane of consistency—if we can describe the musical machine abstractly as sonorous deterritorialization, then I might say it’s the abstract machine of music. The abstract machine is the collection of processes for sonorous deterritorialization. We could very well understand this machine’s mutations such that its different elements completely change gear.
So while history figures into it, I’d say that if I look at concrete musical machines, there is a history there. As far as my abstract machine goes, which I described as a sonorous plane of consistency, I’d claim that this abstract machine is necessarily actualized in concrete machines. The first type of concrete machine: deterritorialization brought to bear on the voice; voices are no longer male or female. Deterritorialization of the voice by way of these sub-machines: the castrato machine, the countertenor machine, etc., all these different designs [agencements]. So, that’s how I define the first concrete machine effectuating my abstract machine.
Then, I think, we come to a different concrete machine. Granted, these concrete machines can be dated. I can say that such-and-such arrangement [agencement] happens here, while such-and-such sub-arrangement happens over there. The castrato machine takes shape in Italy at a certain point in time and then ends at a certain point in time. It’s a fact. Then I turn to another arrangement: sonorous deterritorialization continues, but it no longer operates on the voice; it’s an instrumental or symphonic deterritorialization that takes the voice and makes it part of the machine. It’s no longer a question of machining the voice; it’s about making the human voice a part of the machine. At which point, I’d argue that something in the machine has changed.
So, I can’t help but bring in, if not history, at least dates, just like proper nouns. A proper noun is an indicator for a concrete arrangement. I treat any noun I might use to point to a concrete arrangement as a proper noun, including dates. And from one concrete arrangement to the next, every mode you can conceive: you can conceptualize mode by proliferating. Here’s where I’d bring up real-life rhizomes. History would only play a strictly secondary role; I’m not trying to bring us back to a historical perspective. What I need are concrete coordinates for concrete arrangements, concrete coordinates like: proper nouns, dates, places, whatever sort of haecceity designating concrete arrangements which, with all the same degree of perfection—depending on how much perfection they’re capable of, at least—all effectuate the same abstract machine. But that doesn’t mean the methods for sonorous deterritorialization aren’t very different, depending on whether they apply primarily to instruments, to forms, and so on.
Student: [Inaudible]
Gilles Deleuze: There wouldn’t be a cut-off for when abstraction begins or ends in music; I’m not on board with the idea of abstract music.
Georges Comtesse: We ought to divide the vocal machine further, into a machine deeper than the vocal machine, on that entails the machine of silence. If it weren’t for this machine of silence, Boulez wouldn’t be able to say that there’s a musical process within silence, a process of abolition, of destruction, and that in music, we always cherish the object we want to destroy. That’s the machine of silence.
Pinhas: What you’re saying is pretty serious. You’re picking back up on the subject of “noises,” turning musicians into carriers for the death drive, major factors in contemporary distress, the great figure of death, at the same time as repetition is becoming a phenomenon of its own in the case of stock “noises.” So, on the one hand, there’s a misunderstanding, and on the other, it rots away at everything attached to music, a decay that crystallizes precisely around the aspect of abolition you’re referring to. But what’s the deal with silence—the most theoretically elevated form of music, even? Ultimately, it’s—my example is Cage, Boulez also references Cage—silence comes from the environment. Absolute silence does not exist.
Comtesse: The intensive silence of a musician has nothing to do with their environment. It’s zero degrees.
Pinhas: I’d be happy to buy your “zero degrees” as soon as you give me a definition for it. I’m not so sure, and nobody in the history of music has been able to define what silence is, apart from Cage, who uses silence to refer to an environment that allows for ambient noise. I can’t see what this zero-level silence, this absolute silence, corresponds to besides just a dimension of abolition, a new term for death tacked onto music. The problem facing musicians today is not at all one of subjectivity, nor is it their relationship to silence. It’s about allocating sonorous matter, it’s about speeds and slownesses, it’s about metallic time. Never has it been about death, about representation, or about silence.
Deleuze: I’d like to add something because my heart is bursting with joy. I feel like Richard has his finger on something: in all of your comments—and you know I find them really interesting—I always tell you there’s something I can’t quite figure out; you’re always confronting me with another machine, another arrangement. In every intervention you make, no matter the subject, you’re basically saying, “You forgot about an arrangement.” Today, you’re saying, “You forgot about the silence machine,” which is different from the refrain or the deterritorialization of the voice, and you’re always adding one more to the pile. Richard’s saying that, by always adding more, meaning that in the best sense, aren’t you flinging us back to something that acts as a death drive? Or a castration machine?
I sometimes have a similar feeling. When you say all that, when you say that I’ve forgotten a silence machine, silence especially being something I wouldn’t make into a machine. For me, it goes without saying that silence is a creative element, among the most creative components of the musical machine; there is absolutely no silence outside the musical machine. In the act of deterritoriality, you have the refrain, with noise and the environment; in the musical machine, you have all sorts of elements in various proportions, and one outcome of these deterritorializing processes is silence.
As to the question whether silence can or cannot be defined—personally, my position is different from both yours and Richard’s; I’d claim that silence is perfectly capable of being defined, but it can only be defined within the musical machine. In Boulez’s text, the tendency towards abolition is wholly a component of the musical machine, and any other kind of trend towards abolition would be completely different, wouldn’t have anything to do with the sort of abolition specific to sonorous abolition. So, for Boulez, abolition is fully integral to the musical machine. For you, we no longer get an arrangement or a machine; we end up with a death drive, or the equivalent of a death drive. That seems to be what Richard is saying.
Pinhas: It’s the worst thing you could say about music.
Deleuze: There is no death drive. There are machines whose components drift towards abolition. If you extract all these abolitions, which comprise different machines—as I see it, if you isolate pure abolition for the sake of a dedicated machine, that messes everything up.
Note from WebDeleuze: A lengthy discussion about the death drive. [End of text]
Notes
[1] La Rose des Tudor (Paris: Julliard, 1976). See ATP, p. 303.
[2] While plan can mean “plan,” “plane,” or in the context of cinema, “frame” or “shot,” (among other things), it appears here as “plane.”
[3] “Sonorous” (sonore) is less comfortable than “audio” or “sound,” but the translator uses the former to follow precedent and to avoid ambiguity over “sound” as an adjective.
[4] Following precedent, the French ritournelle appears here as “refrain,” though the reader might understand it as a “ritornello,” a tune, a little ditty. On the “refrain”, see plateau 11, “1848. Of the Refrain”.
[5] Note from WebDeleuze: Deleuze’s mention of Wagner is an implicit reference to Boulez.
[6] On sound film in American comedy, see Lecture 17 of the Cinema 4 Seminar (March 26, 1985). On Bogart in particular, see Lecture 19 of Cinema 4 (April 23, 1985).
[7] See Deleuze’s mention of the “fort-da” game in the Thousand Plateaus 8 Seminar.
[8] Regarding haecceities in Nietzsche et al, see ATP, pp. 328-330.
[9] Text by Boulez; in light of the discussion in ATP, pp. 330-331, esp. p. 542 no. 53. The text cited might be Par volonté et par hazard (Paris: Seuil, 1975).
For archival purposes, the original transcript was prepared by WebDeleuze in the early 2000s, with the updated transcript prepared in March 2023. The translation was completed in July 2023, with additions and a new description completed in September 2023.