The Strains of the Voice

The Strains of the Voice

Steven Connor

This is the English version of an essay that first appeared in German, in a translation by Holger Wölfle, in Phonorama: Eine Kulturgeschichte der Stimme als Medium, ed. Brigitte Felderer (Berlin: Matthes and Seitz, 2004), pp. 158-72.

What is a voice? It is a straining of the air.

Before Robert Boyle (1658) had demonstrated and measured the ‘spring of the air’, voice had functioned as just this kind of elastic body. For voice is not simply an emission of the body; it is also the imaginary production of a secondary body, a body double: a ‘voice-body’ (Connor 2000, 35-42). This voice-body is not inert image – spectre, wraith, or indolently wreathing smoke. It is tense and braced with a kind of life. Guy Rosolato has spoken of the power of emanation that belongs to the voice (1974, 76), but the seeming naturalness and irrepressibility of the voice’s exuberations should not prevent us noticing that voice is produced through a process that necessarily creates stress, as air is directed under pressure through the larynx and then out through the mouth. As it moves it is modified, bent, detained, accelerated. Everything that is said about the exercise of the voice – by coaches, experts, trainers and voice professionals of all kinds – implies that it should be easy and relaxed, an effortless effect of the breath. The voice must be produced without inordinate stress, which will damage or distort it. Coaches and gurus offer exercises and visualisations designed to free your voice from the cramping, vampiric grip of its bad habits. But there is no voice without strain; without the constraining of sound in general by the particular habits and accidents that, taken collectively, constitute a voice, and the constraining of the body to produce voice. The breath is drawn as a bow is drawn, by applying a force against the resistance of the diaphragm and intercostal muscles. The power of the voice is the release of the kinetic energy stored in these muscles as they return to their resting positions. But the voice’s energy is not simply given out. For there to be voice, there must be a secondary resistance, the impedance or thwarting of this outflow. Where the breath simply escapes, there can be no voice. It is on this basis that phonetics distinguishes between ‘unvoiced’ and ‘voiced’ sounds, like the sibilant ‘s’ and the ‘z’ which comes about through the addition of voice.

Sound, wrote Aristotle, is a kind of pathos, a suffering. The air is battered, stretched, percussed when there is sound. The voice never simply appears, but is expressed, its shape formed out of resistance. What resists the voice? The heaviness, the reluctant inertia of things, the world’s weary wish to hold its peace. The voice must overcome this lethargy deep down things. It is a striving, and a disturbance: it subjects the world to strain.



In English, the word ‘strain’ has a double signification. It can signify the physical stress to which an object or material may be subject, up to, but always short of, the breaking point. But it can also mean a form, kind, pedigree, inherited line, genus, or character, most commonly nowadays in references to the differing ‘strains’ of a virus. The word ‘strain’ also once meant a musical phrase or measure. All of these usages of the English word have in common the idea of a filtering; of that which is refined, selected, or in some other way distinguished from a background. Perhaps the action of filtering brings together the functions of selection and of physical stress. The filter selects by physically forcing a single or simple substance, or sequence, out of a more complex ensemble. We conventionally associate the voice with identity (in French, English and German, a voice can be synonymous with a vote). But this identity is a strain, the product of a pressure, as well as the mark of something that is merely distinguished or distinctive; German Stimme also signifies a tuning, or pitching. My voice is said to be my own, because nobody else can use my voice, or be in the place of my voice, without theft or deception. But perhaps this is so not because the voice is my property, but because it is, to borrow Gerard Manley Hopkins’s term, my ‘instress’, the power by which my coherence (which Hopkins named ‘inscape’) is conveyed or borne out. Extending this to non-vocal objects, as Hopkins does, instress becomes the way in which such objects ‘voice':

each hung bell’s
Bow swung finds tongue to fling out broad its name;
Each mortal thing does one thing and the same:
Deals out that being indoors each one dwells;
Selves – goes itself; myself it speaks and spells,
Crying What I do is me: for that I came. (Hopkins 1970, 90)

When there is voice, the percussions of the air seem to have formed a determinate shape, a style or signature of duress. In a voice, some syntax organises the inchoate roar or rattle of pure noise into a dance of opposed internal stresses. In just the same way, a bridge thrown between two banks can only continue to occupy its space, allowing traffic and communication, because of the patterns of internal stresses that hold it together. What is true of architecture in general, – that only that which can support itself, or hold itself together by internal stresses can stand up – is true of a voice. A voice is a structure of stresses and strains, and is pitched against itself, ‘pitched past pitch of grief’ (Hopkins 1970, 100), as well as standing out against the surrounding silence or noise.



In the voice, the air is both constrained and stretched out – as, indeed, it is in the cycle of pressure within each individual sound wave. ‘So I at each sad strain, will strain a tear’ says Shakespeare’s Lucrece as she listens to the mournful song of Philomel (l. 1131). Lucrece proposes to press a knife to her heart as she listens, matching the thorn against which the nightingale traditionally presses its breast to elicit its song: ‘These means, as frets upon an instrument,/Shall tune our heart-strings to true languishment’ (ll. 1140-1). English retains the idea of stretching in the word ‘tenor’, used to mean the burden or gist of an utterance. The tenor of an utterance is that which, diffused through it, holds it together through time. The idea of holding is there too in what is called the tenor voice, conventionally said to occupy the range from the octave below middle C to the A above it, and so-named because it was the part to which the melody is usually assigned, and which therefore ‘holds’ the tune. The word ‘tune’ itself is a modulation or screwing up of the word ‘tone’, which has as its origin the Greek tenein, to stretch, which lies behind the whole family of words signifying modalities of sound and stress: tone, tenor, tune, tension, tendency, tenderness, tenuity, and intent.

Speaking is an action; while the voice-body it precipitates is a quasi-object. And yet the opposite might also be thought to be true. Speech is articulated, broken or segmented into separate objects. Voice is what reaches across and between these segments, maintaining their continuity. Voice is the tenor and the tensor of speech, what binds or holds it together. More than this, even, we might say that voice is the tone of being. There are many ways of maintaining the tenor of the voice. The child, enraptured by its new acquisition of the power of voice, will babble, barely pausing for breath. The garrulous aged will similarly keep silence at bay through an unpausing onslaught of voice. There must be no chinks in the continuity through which silence or formless sound could rush. When the mind or body is subject to unbearable stress, the voice may be drawn out into howls and moans, which take the strain all the better for being drawn out, for being, as we say, ‘inarticulate’, unbroken.

What happens, in fact, when the voice breaks, when it is wrenched apart by stammer, terror, hilarity or grief? The smooth and regularly successive cycles of the breath are breached, by sobs and hiccoughs, say, but the voice survives, in the form, precisely, of the periodic rhythms of the broken voice, that holds together convulsion, holds together in its convulsion. Paradoxically, the voice at breaking point, the voice that appears to give way, is also a tensor, a taut, cohering cuirass.

Though itself formed from tension, voice can also hold or contain stresses which might otherwise tear us apart. Didier Anzieu’s ‘acoustic envelope’ or ‘audio-phonic skin’ is an imaginary integument provided by the enclosing, protective touch of the mother’s voice (Anzieu 1989, 10). It is an ideal surface tension, neither too slack to maintain coherence and tone, nor too tight to resile from shock and accident.


Hanging On Its Words

Where does the voice come from? The history of ventriloquism provides many examples of voices produced from surprising or illegitimate parts of the body. The ventriloquial voice speaks from the belly, from the sternum, from the armpit, from the genitals, from the nose, from a second throat, or alternative vocal apparatus hidden within or alongside the usual one. But where does the voice come from in ‘normal’ speech? From the tongue (the tip of the tongue?) The mouth? The throat? The larynx? The lungs? The diaphragm? None and all of these things. The voice depends upon an entire vocal apparatus, a series of way-stations, the elements of which are held in tension as long as there is the action of the voice. What joins and holds them together is the straining passage of the air through them. Air joints and articulates the puppet that is voice.

Nor is the production of the voice limited to the physical production of sound or those parts of the body that are capable of producing sound. For the voice also induces and is taken up into the movements of the body. The face is part of the voice’s apparatus, as are the hands. The shaping of the air effected by the mouth, hands and shoulders marks out the lineaments of the voice-body (which is to be distinguished from the voice in the body). When one clicks one’s fingers for emphasis, claps one’s hands, or slaps one’s thigh, the work of gesture is being taken over into sound, and voice has migrated into the fingers. Strange, that the ‘unnatural’ or artificial production of voice through the classic ventriloquist’s dummy should also link hand, fingers, face, mouth and voice into a single assemblage, though in a different configuration than in ordinary voice production. The ventriloquist’s relaying of voice through the speaking figure is a kind of anagram of the organisation of elements in ordinary vocalisation.

The voice is the body’s syntax; its uprightness, its tone, tension and extension. If it is true of human beings that language enables us to be where we are not, and prevents us from ever being anywhere but beside ourselves, then it is the voice that stretches us out between here and elsewhere. One cannot be fully ‘here’ unless one is silent; one cannot vocalise without being ‘there’ as well as here, without being drawn out into the ambivalence of being here and there at once.

Samuel Beckett gives us in his play Not I the image of a speaking mouth, which speaks of ‘the whole being…hanging on its words’ (Beckett 1986, 379) In what sense does one hang on another’s words, or on one’s own? One waits for them, straining to hear, held back until the words come, longing (and when one longs one is elongated, of course). Or one clings by one’s fingertips, drawn out tight by one’s own strait weight.



Voice is not just a force of outgoing. It is also often a reaching. Voice is in fact for babies the first grasping, a way of fetching and carrying objects of desire when no other way exists; the baby in the cot brings the world to it by means of voice. The child who has learned how to coordinate eye and hand to grip and carry does not forget that voice can reach further than the fist or fingers. Perhaps thereafter the voice is identified principally as a mode of extended reach, as a way of stretching towards what one cannot physically reach. The voice reaches out and brings back, like the frog’s tongue. The fact that, in Latin languages, languages themselves are known as ‘tongues’ allows the sense of spoken languages as protrusions, tastings, attentions, elongations, elinguations.

Freud’s account of the fort/da game in Beyond the Pleasure Principle foregrounds the two modes or states of the voice, excursive and recursive. In staging disappearance, naming what is away and what is there, ‘fort’ and ‘da’, the voice characterises, not just the two conditions of its play-object (the cotton-reel as symbol of the mother), but also its own two conditions, as excursive and recursive (Freud 1955, 14-17). In one phase of the game, the voice is despatched, sent out, expressed, uncoiled – and Freud characterises the sound made by the child to accompany and enact it not as the end-stopped ‘fort’, but as its open, yearning approximation in ‘o-o-o’.. But in the next phase of the game, it is reeled back in, brought back to itself, on the very imaginary elastic employed for its propulsion. The voice is carried in the tension of the cotton, which becomes transformed into a stretched telephone wire, a way of telephoning oneself, calling oneself back. Perhaps the fort/da game could be thought of as figuring the hearing-speaking circuit. What the voice gives out, the ear reins back in.

The more complex its utterance, the more the voice may appear to be reaching beyond itself, with the risk that it may be unable to be retrieved. As we come more and more to live in the voice’s mode of reach, language proves to be able to reach towards things that can never be retrieved, since they exist only in the reaching for them. ‘Words, after speech, reach/Into the silence’, wrote T.S. Eliot in Burnt Norton (1986, 8). If language is defined by its units and the rules of their combination (lexis, grammar), then the action of voice seems to be signalled in words that testify to the magically soliciting reach of language: invocation; evocation; provocation; revocation.

Of course the voice conceived in this way, as the stream, thread, cable, or wire, is tenuous. But one might also say that it is strengthened by this concentration into the form of the line. The voice is as feeble and as steely as the spider’s thread.


Though we delight in and are captivated by full, rich voluminous voices, voices that are full of room and boom and body, the default condition of the voice is as something drawn out or elongated. Voice is one of the principal ‘extensions’ of man, in Marshall McLuhan’s phrase (1964), but extension includes tension, attenuation, as well as enlargement of scope. Voice arises, bubbles up, comes out, is ejaculated, emitted, ejected. But this emission takes the form of a continuous stream, which we feel remains in contact with its source, as in a jet from a fountain or a geyser. Early graphic depictions of voice represent it not, as contemporary comics, by a speech bubble, a formed and enclosed cloud of utterance hanging complete in the air, but as a scroll, uncurling from the lips. (But even the speech-bubble seems to suggest that the action of the voice may be a kind of lasso or noose.) The habit of thinking and speaking of ventriloquism as a ‘throwing of the voice’ participates in this tensile conception. The thrown or projectile voice is not despatched, sent out once and for all, as a cannonball or javelin is, but maintains an unbroken arc through the air. It forms a line, like the line the angler casts, or as though the words one hears were a kite out at the end of a string lodged in the throat of the speaker.

Or like a telephone wire. The telephone actualised the fantasy of the wired voice more palpably and suggestively than any other device. The wired voice of the telephone is much more than a technical prosthesis for the voice; it is an image of the voice itself, considered as prosthesis. In the telephone, the voice is thinned to a filament, a living, thrilling nerve. The fact that wires carried voices, or electrical impulses that were to be converted into sound suggested that this device had succeeded in compressing sound into a line. Thus we continue to urge each other to ‘stay in touch’. The language of language is full of images of threading and filiation. Somebody who detains you with his speech was once said to be ‘buttonholing’, you, recalling the practice of physically crooking a finger in the buttonhole of one’s interlocutor. ‘I’ve lost my thread’, we may say, when the stream of spoken language fails.

The ‘voice wars’ which raged among French anatomists during the eighteenth century about whether the voice was a wind, reed, or stringed instrument was eventually resolved during the 1740s by Antoine Ferrein’s description of the action of the vocal cords (Connor 2000, 199-200)This description confirmed the phenomenological intuitions that the voice was a kind of ‘blown string’, an extruded tautening of the air.

Voice was not able to be transmitted through wires until Robert Hooke invented the device that survives in the children’s string telephone today and anticipated Bell’s electrical telephone. Hooke stretched a silken cord between two diaphragms, and was able to transmit the sound of voices for distances of up to 150 metres; the sound would even go round corners, provided the angles were carefully constructed and the line kept taut. But even before this, there had been an apprehension of the relations between sound, especially the sound of the human voice, and the principles of constraint. Musical instruments, like pipes, trombones and trumpets, amplified and transformed the voice by forcing it through narrow channels, in what was often understood as a doubling of the voice’s own methods of sound production. The use of pipes and speaking tubes to convey voices long distances established an understanding of the energy ratio between power and constraint: the voice that is put under pressure can reach further.

It is not just amplification or increase of reach that involves force. One of the most curious but long-lived beliefs about sound was that it, if it could be accelerated, it could also be trapped. Giambattista della Porta’s Natural Magick of 1584 devoted a chapter to the topic of ‘Whether Material Statues May Speak By Any Artificial Way’. The chapter begins with the report that, in ancient times, ‘there was a colossus of Brass, placed on a mighty high pillar, which in violent tempests of wind from the nether parts, received a great blast, that was carried from the mouth to a trumpet, that it blew strongly, or else sounded some other instrument, which I believe to have been easy, because I have seen the like’ (Porta 1658, 385). Porta goes on to discourse upon the properties of sound upon which both musical instruments and speaking statues depend: ‘We see that the voice or a sound, will be conveighed entire through the Air, and that not in an instant, but by degrees in time’ (Porta 1658, 385). The fact that sounds are preserved and conveyed ‘entire without interruption, unless they break upon some place’ (Porta 1658, 385) makes it possible to convey sounds great distances: ‘if any man shall make leaden Pipes exceeding long, two or three hundred paces long (as I have tried), and shall speak in them some or many words, they will be carried true through those Pipes and be heard at the other end, as they came from the speaker’s mouth’ (Porta 1658, 385-6). This principle of the preservation of sound convinced Porta, as it did others, that cunningly engineered pipeworks might be able also to trap sound, not just conveying it through space, but holding it up in time, thereby adding a phonographic to a telephonic principle:

wherefore if that voice goes with time, & hold entire, if an man as the words are spoken shall stop the end of the pipe, and he that is at the other end shall do the like, the voice may be intercepted in the middle, and be shut up as in a prison; and when the mouth is opened, the voice will come forth, as out of his mouth that spake it; but because such long Pipes cannot be made without trouble, they may be bent up and down like a Trumpet, that a long Pipe may be kept in a small place; and when the mouth is open, the words may be understood. I am now upon trial of it. (Porta 1658, 386)



Ventriloquism centres around these two strainings of the voice, in extension and compression. The traditional ventriloquist performance has the dummy or speaking figure operated by the ventriloquist’s fingers. The voice is out at the finger’s ends, the words hovering not on the tip of the tongue, but at the fingertips. In the so-called ‘distant voice’, the voice may be in the air, on the roof, up a chimney, in a basement or on the end of the telephone line. But the knowledge that the voice really belongs to the one who affects only to hear it keeps us aware of a kind of filament or cord stretching between the apparent and real locations of the voice. The distant voice is also often enclosed, in boxes, suitcases, cupboards, or even underground. When it is thus enclosed, the voice is imagined as coiled upon itself with the kinetic tension of compression. In Edgar Allan Poe’s story ‘Thou Art The Man’, a corpse is stiffened with whalebone and then pushed down into its coffin, so that when the lid is taken off it springs upright, so that, given a voice by a nearby ventriloquist, it may dramatically denounce its murderer (Poe 1850, II. 418-432). The voice is either stretched, or compressed – but is always elastic, always under pressure.

This means too that there is always a possibility of pain in the voice. Children learn to use their voices first of all to cry, with fear, hurt, or need, before they are able to use language for pleasure. When we cry, with pain or loss, it is as though we used the voice to rid ourselves of an unbearable burden of tension. But the sound that we release can hold the pain, even in its sundering from us. Indeed, the sundering is itself a pain, as though the dissociated voice were haunted, like a phantom limb longing for its tenant body. Philomela, her tongue torn out by her ravisher Tereus, eventually takes the form of a nightingale, the rarely-visible bird whose song takes the place of its physical presence, and will lengthen out our agony into beauty.

Myths, legends and stories have often imagined the condition of death as being like a voice separated from its body, for example, in the twittering shades who come to drink themselves back into bodily substance from the ditch of blood provided for them by Ulysses in Book XI of the Odyssey. The voices of the dead conjoin the two kinds of strain I have been evoking here. They are attenuated, ‘drawn out’ from the body into the most evanescent condition. But they are also, like Echo in Ovid’s fable, shrivelled down to bones and voice, or like the voices interred in the cave, or grave, or constricted in the tight coils of the gramophone record or the reel of tape, pent in their impalpability. Their contemporary image is to be found in the ‘electric voice phenomena’ allegedly captured on tape by Friedrich Jürgensen in 1959, while recording birdsong. Mere flitters of utterance, the voices seem to be trying to keep themselves entire amid the blizzard of white noise that threatens to obliterate them. We too, apparitions of the voice, hang by a thread, on what continually escapes us.


Anzieu, Didier (1989). The Skin Ego. Trans Chris Turner. New Haven: Yale University Press.

Beckett, Samuel (1986). Complete Dramatic Works. London: Faber and Faber.

Boyle, Robert (1660). New Experiments Physico-Mechanicall, Touching the Spring of the Air, and its Effects (Made for the Most Part, in a New Pneumatical Engine). Oxford: for Tho. Robinson.

Connor, Steven (2000). Dumbstruck: A Cultural History of Ventriloquism. Oxford: Oxford University Press.

Eliot, T.S. (1986) Four Quartets. London: Faber and Faber.

Freud, Sigmund (1955). Beyond the Pleasure Principle. Trans. James Strachey. In The Standard Edition of the Complete Psychological Works of Sigmund Freud. London: Hogarth Press. Vol 18, 1-64

Hopkins, Gerard Manley (1970). The Poems of Gerard Manley Hopkins. 4th edn. Ed. W.H. Gardner and N.H. MacKenzie. London: Oxford University Press.

MacLuhan, Marshall (1964). Understanding Media: The Extensions of Man. London: Routledge and Kegan Paul.

Poe, Edgar Allan (1850). The Works of the Late Edgar Allan Poe. 2 Vols. New York: J.S. Redfield.

Porta, Giambattista della (1658). Natural Magick by John Baptista Porta, a Neapolitane; in twenty books … wherein are set forth all the riches and delights of the natural sciences. London: for Thomas Young and Samuel Speed.

Rosolato, Guy (1974). ‘La voix: entre corps et langage.’ Revue française de psychanalyse, 38, 75-94.