Talk given at Noisetheorynoise, University of Middlesex, March 2004. See Ben Watson’s strange interpretation of the text below at the Radical Philosophy site.

Paul Hegarty, Voice as Noise

I: Nature and the Lost Good Voice
Can we separate speech from language? Speech from voice? Voice from soul/subject or body? What would we be doing? What would we be hearing? Would we be listening? Would we be still listening? What kind of judgement are we making in enacting any split, and what kind of judgement results from the split – where is the ‘good’? good speech, language, voice, communication, absence, interaction, absence of all of the above? What is a bad voice? For Steven Connor, ‘the bad voice is the voice of the self become other: the good voice is the voice of the other become self’ (Dumbstruck, 32).
When we think of ‘voice’, I think we necessarily raise the question of human voice, of voice as human. Even before speech, and even before (devant and avant) the human, a voice carries an organising principle: it is identical with itself and therefore sets up a location for itself – i.e. it identifies a given animal. Speech implies a heightening of this, and writing its crystallisation. Writing has been seen as the bad, and where else to start but with Rousseau, after Derrida. Derrida claims that writing, for Rousseau, represents a decline in the authenticity of communication, and that in fact he needs the supplement offered by writing to make his claims about speech. The play of speech and writing illustrates the presence of subjects, who have access to themselves and believe themselves self-present. This self-presence, however relies on an unending process of never quite being yourself, never quite communing internally. In short, all speech, including of your ‘self’ to your ‘self’ is caught up in writing, in organised systems, and therefore always already part of ‘bad’ writing.
For Rousseau, humanity develops in parallel with the development of language, but simplest is not necessarily best. Rousseau does not praise a formless language, but the earliest formations of it – where it is musical, beautiful. Language is aesthetic first, and so is humanity. Gestures can suffice for communication, but speech is already a supplement, a useful one. ‘So we speak to the eyes much better than to the ears’ [‘Ainsi on parle aux yeux bien mieux qu’aux oreilles’, but words move us (Essai sur les origines des langues, 46)]. Early languages are to do with sentiment (51), with reason and abstraction coming later (54). But rather than the story of origins, what I am interested in is the aesthetic aspect, which dominates what occurs in early language. He argues that we have lost the ‘langue sonore’ (60) – sonorous, or pleasant sounding language. Rousseau’s early humans are not more in touch with themselves, but more in touch with each other. The communication is as much to do with reception as transmission, and the ‘purity’ of first language(s) is an empty one, devoid of meaning. This emptiness is seen as good by Rousseau, and we can take this to be a privileging of voice rather than of speech – it is still to do with identity, but an uncertain, unknowing one that extends beyond individualism.
Kant has other ideas about voice and voices. In the Critique of Judgement, Kant has some notable difficulties with music (and abstraction, for example), and some problems with the wrong kind of voice – one that would seek to imitate nature. Natural music, in the form of birdsong, is fine, and supplies the necessary suggestion of purpose to let us fool ourselves into treating it aesthetically, but this pleasure can be disrupted:
Even birdsong, which we cannot bring under any rule of music, seems to contain more freedom and hence to offer more to taste than human song even when this human song is performed according to all the rules of the art of music, because we tire much sooner of human song if it is repeated often and for long periods […] when birdsong is imitated precisely by a human being (as is sometimes done with the nightingale’s warble) it strikes our ear as quite tasteless. (94)
Birdsong is nature with the suggestion of art; a human imitating a bird is art pretending to be nature, which disqualifies the art and negates the natural. Perhaps predictably, it is not the imitation itself that is the problem, but our awareness of it that creates the problem. Writing of an inn-keeper whose idea of fun is getting a ‘rogueish youngster’ to copy the nightingale, he states that initially people are happy to hear the song, ‘but as soon as one realises that it was all deception, no one will long endure listening to this song that before he had considered so charming’ (169). Also, the ‘rogueish youngster’ needs mechanical assistance to carry out the deception (‘a reed or a rush in his mouth’). The key difference between Kant and Rousseau actually highlights a similarity of purpose: Kant is talking about now, but Rousseau some mythical ‘dawn of time’. However, both concur that human speech has been degraded. In Kant, it is only the human distance from nature (withheld in the trick) that creates the problem, and if a monkey copied a nightingale, there might be less of a problem. Then, in turn, the problem becomes a way of defining the human.
Noise can be annoying in its own right, when human and, particularly when aspiring to godliness: with regard to the public singing of hymns, Kant writes that ‘by such a noisy [Kant’s italics] (and precisely because of this usually pharisaical) worship they impose great hardship on the public’ (200). Any kind of human noise disturbs Kant, it seems, and in his ‘Speculative Beginning of Human History’, he has the following complaint about:
[This impulse in] children and thoughtless persons who disturb the thinking members of the commonwealth by rattling, shouting, whistling, singing, and other noisy amusements (the same occurs in religious devotions). For I can see no other motive for this except that they want to make their existence known far and wide. (Perpetual Peace, 50n)
This, though is a bad, or at least an impoverished version of acceptable communication. God has given us the instinct or ‘impulse’ to communicate, and within human bounds it is desirable to imitate the sounds of language (50). Kant might seem now to be very similar to Rousseau, but Kant is very clear that good voice happens elsewhere and that human society has always created a noisiness (the noise comes from the same ‘impulse’ that instinct has created for us to use in a good way), a bad voice that impinges on reason and morality. Where Rousseau has a sonorous ideal environment, Kant’s tends to silence.
It is Rousseau we see more often in the performance of voice as something outside or beyond language, where another language comes in – that of feeling, expression, nature; but it is Kant that is closer to an interesting reading and critique of that approach. Noise, in the form of the human voice outdoing itself and nature, playing tricks or inflicting itself, undermining the holiness of religion, or just rattling, shouting and whistling, is about undermining order and nature. Noise is not supposed to take us somewhere, let alone somewhere better, amiably ecstatic. It has to annoy us, disturb us, not be itself, not be anything, possibly.

II: Passing Through the Body
The last century saw many attempts to mobilise the unwanted sounds described as noise. In terms of voices, we can hear Marinetti’s onomatopoeic poems, and Schwitters’ Ursonate. Sound poetry starts to use sounds of the voice in an ‘improper’ way. John Cage, meanwhile, incorporated sounds other than verbal he was to make during a lecture (45’ for a Speaker (1954)). What unites these, and other, more improvised uses of ‘meaningless’ or ‘contentless’ singing, speaking, or voicing, is that the noisy voice is being put to use, and its noisiness reduced. Just as musique concrète sought to bring noise in, to find a new resource for composition, so often experimental music in many genres has sought something other than meaning in voice, and merely displaced the location of meaning, to something like expression.
How then can voice be made noise? At the most direct level, we can be the ‘rattling shouting’ children or rogues; or we can contribute to the world of background, ‘ambient’ noise – the irony being that Kant’s worst noises are now mainstream and expected. Is noise to submit to noise? Possibly. Can this be used, in a use that would challenge its own purpose – i.e. fail, but well? We have to not associate noise with ‘any sound that doesn’t fit’, or that doesn’t carry meaning. In terms of the human voice, we need to find expression that undoes itself as expression, that stays on in the debris of itself.
Part of how it can do this, of how, in Bataillean terms, it can bring itself low, is to stress the material production of voice, and also the failure or difficulty in that production. Voice has a claim to transcendence that has not gone away since Kant and Rousseau. It represents the supposed interior life or existence of the speaker, and therefore has a link to the ideal, to contemplation, rationality, self-reflection, and acknowledgement of the other. A voice that eluded these, whilst suggesting them, might offer the other to the Ideal Voice, as long as the bodily processes took over from ways of thinking ‘the body’ as idealised location of identity (as insisted on by a lot of ‘performance’, performance theory etc, even though it this an older, less interesting position than even so-called ‘mind-body dualism’).
One such voice is that which courses through Artaud, unwitting discoverer of his own ‘body without organs’. His is the body without self, and also a body dismantling itself. One way Artaud is able to observe this process is through the mutation of voice: ‘neither my cry nor my fever are mine’ (vol I, 80), and this in the context of his perilous embodiment: ‘with each of my tongue’s vibrations I retrace all the paths of my thought through my flesh’ (165). This flesh is not a new body, a surrogate self, but living as ‘apprehension’, an apprehension caught in the circuit of breath and exhaled sound. In the course of learning to work this strange situation, he stresses the expelling of breath, where he would be ‘expelling not air but the very capacity to make sound’ (vol II, 113).
In so doing, he builds a subjectivity that always comes apart, that fails, that is only as expelled. The most obvious way this occurs is in glossolalia, vocal sound seemingly without form, without conscious control, and in screaming – most importantly in taking your voice outside of its ‘natural’ pitch (does a natural pitch exist for me, or in reception of me, or in me as reception of me etc). It is not, then, just the loss of words and meaning that counts, but the alteration that creates noise, creates a way of being other than rational, controlled, anchored in ‘me’. There is something very hackneyed in thinking screaming is more authentic than speaking or singing, or alternatively that it is good for avoiding claims of authenticity, so with Artaud what gets his screaming out of that bind is that it occurs in contrast to and in the context of discursive speech. This speech is itself broken by its rhythm, which undermines the claims of the content (such as the Americans, who are taking over the world, being resisted through microbes of God…). Screaming and percussion offer interludes in Pour en finir avec le jugement de Dieu, parodying a play, or musical narrative. Within the scripted elements, Artaud, Blin, ……. Continually drift to the form of their vocal performance, which exceeds both content and form of the discourse, while emphasizing its effect. There is, in other words, a use-value in screaming and shouting as if possessed if the script talks of shit and God, God and shit, shit as God, we are all shit….but instead of seeing a tidy relation between form and content – expressive form – we should take the vocalising in the context of Artaud’s belief in exhalation, voice, breath as weakening the power of the self and the self’s utterance, and take it as a bad use-value, unproductive.
What Artaud expels is a voice which is different to the voice of reason and subjectivity, as even Derrida acknowledges: ‘breath [le souffle] is not the same thing as the voice, not the voice of language, discourse, the verb and the word, in any case’ (‘Forcener le subjectile’, 85). This is a material voice that realises a low immateriality (like Bataille’s low materialism) as opposed to the highness of the cultured speaking or singing voice. Artaud realises Lucretius’s idea of the voice as material: ‘when atoms of voice in greater numbers than usual have begun to squeeze out through the narrow outlet, the doorway of the overcrowded mouth gets scraped. Undoubtedly, if voices and words have this power of causing pain, they must consist of corporeal particles’ (On the Nature of the Universe, 146). And while we’re with the low, the same process could be heard, felt, seen, in reverse, in the dematerialisation of Artaud’s shit, Artaud’s insides.
Barthes might offer us a more literal perspective: for him the voice is something added to breath – there is a ‘grain’ which is the product of internal percussions and channellings, of the physical form of the person emitting the sounds (‘L’écoute’, 226). This voice has nothing to do with talent, training or even correct physique, but with the lowness of voice, as ultimately uncontrollable. This perspective takes away the possibility of expression being controlled the subject individual and democratises the listening judgement to be made. In addition, this grain creates the individuality of the voice. There is a simple way that this applies to a singer you might like, but, more interesting is his assertion that this individualisation further minimises the singer/speaker’s role in producing ‘their’ voice, because he claims this ‘individuality’ is not a personal one – it is not someone’s property (‘Le grain de la voix’, 238). Artaud is of course ahead of Barthes here – and aware that ‘grain’ is not another term for ‘timbre’. Intonation is part of it, but only as a means of attacking intonation (as carrier of significance, of emotion, expression in general). The voice(s) of Pour en finir lose their individualised grain, as the ‘grain’ takes over, and crosses from one participant to the next, regardless of gender.
The concept of grain might seem to encourage a consideration of sexual difference, or presumptions about it, but when the voice is becoming noise, it is also losing gender – for example in the loss of unified pitch, or ‘natural’ pitch. Pour en finir becomes a realisation of Artaud as not Artaud, as not God, as not shit, only to become all those things again, only ever failing to fully become. This noise can only occur by making the noise cross from one level to another, from content to form and vice-versa, from shit to God, from Artaud to Blin to…… Grain in Artaud is an undoing, a bringing to be as expulsion rather than an expulsion realising voice.

III: The Expelled Voice Within
Joan La Barbara wants to intervene in the breath and grain of the voice, the voice as some pure expression, but unable to attain expression. In ‘Hear What I Feel’, she tries to communicate a pure emotional reaction, through her voice. She sits blindfold for an hour, touching nothing with her hands, then is led into the performance space. She is made to touch six substances in ‘small glass dishes’. We are to hear what she feels, and presumably feel her feeling – which is not quite the same as experiencing the feeling, so this is the first interesting divergence form the purity she ostensibly aspires to. She shuns habitual codings for those feelings, and seeks an extra-musical vocabulary, again, with the notion that this is more natural, closer to something. These sounds are a translation, a replacement, and offer interference for the listener wishing to access the profound being of La Barbara. For the listener, if not the live audience, the substances themselves are noise rather than the source of a chain of reactions (and to produce the sound expression of them requires they be noise for the performer). Above, or beneath all, the sounds are not the triumph of expression, but its restraining. The recourse to notionally primal sounds brings bad communication, real noise, beyond the ‘noises’ being brought into artistic creation, expression etc.
La Barbara is both the good Rousseau voice and the bad Kant voice: returning to an animalised humanity raises the voice, but also makes it a simulacrum. The natural voice, the voice as user of sound is always this. The voice become noisy is something else – the simulation failing, becoming noise, becoming minor, or weak, through its obviousness. Here the ‘birds in La Barbara’s head’ make the voice doubly artificial – like the musical voice of early humans, like the birdsong we’re not supposed to imitate. The attempt at attaining a pure voice catches itself, brings its own limit, and somewhere around this limit is noise and a voice whose striving results in blockage.
La Barbara’s conception of the voice also undermines its mission: she claims that ‘voice is the original instrument’ (the title of an album and a recent collection), and that she is returning (it) to its original state. If voice is an instrument, though, it is there to be used, and therefore something else takes primacy – i.e. the user of this instrument, the human subject. But it is this insistence that brings La Barbara’s interesting failure, when the instrument resists, when it becomes an instrument beyond control of the singer/emitter/speaker, or even when, still under control, it can be heard as the resistance to meaning, the physical noise of producing sound as expression undoing that expression, dis-figuring it.
Like La Barbara, Diamanda Galas experiments with a ‘classically trained’ voice. What that voice then does is it attacks the concept of training, or the proper use to which it should be put. Her songs, even at their most narrative, are split by sharply contrasting notes and registers (glissandi and choking, gobbling sounds). Galas seems to be working in the realms of catharsis, of purposeful, if not pure, voice, but she too is nearest to noise when that voice is pushed to where it is not working, where something else is (a kind of es gibt, il y a of the voice). This something else is akin to possession, but by no-one, nothing.
Noise is where it is most obvious, but the noise is not the obvious bit – i.e. screaming, shouting, swearing. The same can be said of Whitehouse, with William Bennett’s recent ‘old lady’ voice as he recites tales of abuse, exploitation, rape and murder (and echoing Rob Halford’s 1980s work…). Galas’s ‘Wild Women with Steak Knives’ also brings such a collapsing of force just where it seems to be at its height, just where the subject has managed to enlist the dual force of voice and knife. It stutters, it fades, it starts to concede to reason, but still occurs, and all it does is occur, almost or only immanently – the voice occurring is the voice outside the subject, but not idealised – it is ejected, and in the ejection, expulsion, excretion, rather than in the person who would be the ejector.

IV: No Voice, ‘Nothing that is Without Soul Utters Voice’
The synthesized voice is wrong, an other immanence – clearly not natural, it defies our attempts to identify the who and the what of the emitter. It raises the question of whether there is a human speaker. Barthes, unsurprisingly unaware of synths and vocoders, tells us that ‘if we encounter a neutral, empty voice, it terrorises us’ [‘si parfois ce neutre, ce blanc de la voix advient, c’est pour nous une grande terreur’ (‘La musique, la voix, la langue’, 247)]. This voice is a threat as our listening fails. The synthesized voice cannot aspire to authenticity, and distances a possible catharsis – so Masonna might be screaming his head off, but it is out of phase with what we hear (through fx) – our sight and sound conflict. But I want to end by looking at the failure that is Neil Young’s Trans tour. The audience is expecting the gritty authenticity and virtuosity of his electric work, and instead gets this: (extract ….). Here we can see/hear the neutral voice acting as noise – the discrepancy between expectation and product, the absence of a ‘real’ voice, which also has to be seen operating, and the drama of alienation already incorporated and disposed of, replaced by alienation. This literal weakening of the voice, of expression, of authenticity, is the mobilisation of failure – actual, risible failure, but as interactive failure rather than just failing. This voice puts the performer at stake much more ‘objectively’ than active noise-making, and brings noise to the formless as process, as weak, malleable form you don’t want to experience, that cannot bring itself to properly be.