Steven Connor

An essay written for the catalogue of Imogen Stidworthy’s exhibition at Matt’s Gallery, London, 25th May-17th July 2011. [pdf version]

The accent of deviation in the living thing
That is its life preserved, the effort to be born
Surviving being born, the event of life (Stevens 1972, 366)

Accents are accidents, wanderings from the straight and narrow of direct speech. They are the sign of what I do not hear and therefore cannot superintend in my speech. They are inessential, incidental, circumstantial. Children enter language by accident, an accident that is however almost impossible to avoid, and so an almost universal contingency. Our entering into language is therefore the necessary accident of what we are. What has to happen is that the voice will be formed in some always utterly particular way, that will never be what had to happen just like that. Our voices, or the accents of our voices, announce that it had to be that it did not have to be this way. And yet, in that I cannot speak with no accent, since to have no accent would be to have no speech at all, this is an essential circumstance, an accident waiting to happen, and having to.

Accent is what, in my speech, gives me away, escapes me, and yet is, for others, the carrier, the leading characteristic of my voice, the annunciation of my enunciation. Accent is the very saying of the thing I may have to say. Accent is what I, or my voice, have picked up on its way, and on its own, by accident, though I am hard put to pick it out from my own voice. Yet this slippage, of that which is never in my grip, seems for that reason to be closer to what I am than I am myself. So what escapes from me also clings tightly to me; I can never rid myself of this cleaving, can as little unpick myself from this acoustic silhouette as from my own shadow. And yet, as with my shadow, I can never wholly meet with it either, can never listen intently enough to get quite in tune with myself, to plant my feet precisely in my own footsteps. So accent is also a kind of haunting of my speech, yes, it is the way I haunt myself, in the vicinity of my accidental self-adherence, in the fact that nobody else will ever be able to have been the exact and specific kind of haphazard that I am, nobody else could ever have followed precisely in the wavering footsteps of the crazy-paving path that I, and with me, almost with me, at my elbow, at my ear, my voice, have taken from the never-to-be straightforwardness of uninflected being.

This exquisitely individuated accident of my accent is also the sound, in my putative propria persona, of others voices. I hear for example, sometimes, in my own voice, vague and failing, a shadow of the sound of my mothers voice, or the sound of the shadow of the rural drawl she so deprecated, that would sometimes, faint and vagrant, slant across it.

There is indeed no accentless voice, just as no day occurs without some kind of weather. Accent is the leaning or reaching of speech towards song – accent is ad-cantum, a literal Latin rendering of Greek προσωδια (prosodia), from προς- towards and ωδη, song. So there are only ever different tints, different accents of accenting. Accent is the tending of speech to song, but speech itself is already this descant, this inclination or angle of incidence of a being towards song. I can break into song, but my voice is already song breaking into, breaking out in me.

If in one sense accent must always be thought of as a drifting, a distraction, a discrepancy, it is also a concordance. Accent arises as the result of a kind of tuning in, the effort to nudge and cluster the adventitious variety of pronunciations towards some single song, some unison. There is a strongly centripetal drive that is constitutive of accents, which bunch ever more closely around their zero degree, the vanishing point of absolute univocity, audible in no one speaker’s voice, from which all the deviations diverge and therefore on which they all also seem to converge. This is not restricted to humans. Hoover, an orphan seal raised by a New England family, learned to vocalise whole ‘sentences’ in a Maine accent (Ralls et. al. 1985). Species living in close proximity or forming social bonds start to sound similar to each other, this kind of vocal convergence having been observed in songbirds, parakeets, hummingbirds, bats, elephants, cetaceans, pinnipeds (seals, sea lions and walruses) and primates (Tyack 2008). An accent is perhaps a collective aversion, a leaning away together, and some accents indeed arise, not from unconscious imitation, but from ‘phonological avoidance’, the vigilant aversion to undesirable sounds (Moore 2011, 42).

If my accent gives my voice its uniqueness, an accent is the approximation that draws my voice towards this average, this orchestrated, coincident divergence a nowhere-existent norm. Not that an accent is fully formed or to the fore even in my own accent, which itself comes in and out of focus, of audibility, and so is a profile of probabilities, an anthology of deviations and derivations from an idiolect that is only a backformation, a derivation from the spread of deviations derived from it. This does not prevent the development of attempts to inhabit or embody this unbodied, ideal speech, this Gaussian curve of normal speech. For all accents are the sound of voice listening out for congruence. Though I am almost entirely tone-deaf when it comes to hearing my own accent, my accent is the sign that my voice itself is all ears for departures from its own impossible norm.

We discriminate these forms of coincidence with the names of places where they come about – West Country, Yorkshire, Midlands, Cumberland, Ulster – unless, that is, we name those places in the first place from where speech thickens into specific kinds of echoism (the Dutch are the ‘doichies’ that English sailors encountering them thought they were listening to). Sometimes, the places named are as narrow and specific as they can be – as for example the shibboleth, where the tiniest fluctuation in air-flow enacts the perilous constriction of the strait through which the fleeing refugee must pass. But these places are only places in that they provide an opportunity for the entrainment of voices, the propensity, that is, for a large number of matchings and tunings, of circulations and reverberations of speaking and hearing to take place. If you want to learn to understand an unfamiliar accent, you will do best if you imitate it (Adank et. al. 2010). For, when we speak, we do not merely take turns to utter. We echo and reecho each other, blending and overlapping our voices, proposing and proportioning pitches and tempi. We do not only give voice, we give our interlocutors’ voices back to them, that they might hear in ours something of what we have heard in theirs.

There are other kinds of accent than the vocal. Jean-Martin Charcot used to advise his students to listen to the distinctive stuttering cadence of their limping patients, to try to intuit the kind of neurological damage from which they might be suffering (Goetz et. al. 1995, 144). Morse telegraph operators had their own distinctive timbres and rhythms which were clearly audible to other operators and constituted a kind of sonic finger-print; ‘A telegrapher’s Morse’, wrote L.C. Hall, ‘is as distinctive as his face, his tones, or his handwriting and as difficult to counterfeit as his voice or writing’ (Hall 1902, 227). A family loosely constellated around the double-buggy exhibits the same spread of deviation in their shared same-difference resemblings. In fact, there are leanings and variations from the norm everywhere, in every form – ‘Don’t look at me in that tone of voice.’ Accents are matters of bodily disposition too, especially in the secondary shaping of the hands and fingers. A national accent, a French accent, say, will involve an entire choreography of fingers, lips, eyebrows, elbows and shoulders, a multi-channel work of simultaneous translation, that at once scatters and descants speech. Voices circulate along this chorus line, not just between lips, but also in and across hands, those eyeless, tasting, tentative, flirting tongues. When we are trying to flush out a fugitive word, we prod and solicit the air, as though our fingers will help to persuade it out of hiding.

Print is bleached, bony monochrome compared to these swirling, shifting, spectral, ancestral striations. Voice is paradoxically that which will always be too big or too small for phonology, the science of vocal sounds, to formalise. Mladen Dolar suggests that phonology is a kind of organised attempt to cure or kill off the voice by reducing it to organisation (Dolar 2006, 19). Accent is the voice of this voice, its accidental logic, the logic of its accidence. Often, it will seem to hover just beyond hearing or uttering, the subtlest burr or lilt, loll, twang or drawl. Written accents were originally added around 100BC to ancient Greek to help foreigners reproduce its distinctive tones, and seem principally to have marked variations in pitch of individual sounds. But though these specialised uses survive, perhaps most familiarly in French accents, an accent has become a much more generalised thing, one that is diffused across the whole of one’s speech, like the letters arcing across the map that spell out the name of the region. An accent is more like a taste or a fragrance than a form; it is a kind of composite style, or tonality, the thrum of acoustic habit, memory precipitated as melody, like the runnel scored by a watercourse.

Our forgetting of how the predestinate accident of entering into speech occurs encourages us towards some extraordinarily strange beliefs about learning. Once the period of rapid language acquisition has passed, we have to resort to much more self-conscious procedures for gaining and retaining knowledge. We learn things from the outside in, first of all reducing them to a model or blueprint, so as to be able to pre-structure the information we aim to absorb. We do this as a necessary economy, for we have drastically reduced means, no longer having the time that children do, or their capacity to be exclusively distracted. But there are areas where we can still dimly grasp that there is no more efficient way of learning than through the path into and through pure redundancy, that is, through pure wastefulness, repetition, increment, practice. As Michel Serres has wisely intimated, for many of the important forms of learning, the order must be prendre, apprendre, comprendre, that we might translate, poorly, as taking in, learning and understanding (Serres 1999, 104). We inhibit the acquisition of the subtlest forms of knowledge in our desire for shortcuts, for efficiency, to quiet our insatiate craving to know in advance what and why we are learning before we set about learning it. But the most virtuoso things about our various forms of finesse may be those which have been absorbed in the blindest, most mechanical way, in the ‘accidence’. The accidence of a language is the system of its accidental features, its variations of gender, case, number, tense. In the seventeenth century, ‘accidence’ was shorthand for the elements or rudiments of language learning, as for example when John Taylor writes ‘I was well entred (forty Winters since)/ As farre as possum in my Accidence;/And reading but from possum toposset,/There I was mir’d, and could no further get’, and subsequently confesses ‘I barren am of Eloquence,/Nor neuer vnderstood my Accidence’ (Taylor 1630, 57, 106).

If there is one practice to which we could do with paying more exacting attention, it is practising itself. The ambivalence of the word ‘practice’, which refers both to performing an action and to preparing for performing it, leads to many cheap jokes (‘Would you like an appointment with the practice nurse?’, ‘I’d prefer an experienced one, thanks’). But practice is always indeed preparatory, in that it comes before and after an action that is never quite present or perfected. Processes of coaching, training, drilling, especially when they are remedial, as in the case of a stroke victim relearning speech, or an immigrant learning a language or correcting an accent, are infantilising, making us feel infant, without speech, but because of this, there is something soothingly elementary about it. Is there a more astonishing thing to be discovered in a human life than this, that you can learn things, and in this mysteriously unmysterious way, simply by doing it again and again? If you go over things often and regularly enough, you will retain what you have learned, which is to say, you will forget it, moving it from declarative to procedural memory, melting it down into the oblivion of your being. You will know it only in the sense that you will have become it, you will have the knowledge now about your person. No analytic techniques or cognitive tricks are required, other than the determination to impersonate the speaker, the knower, the one supposed to be able, that you intend to become, until, slowly, suddenly, you may find you are at last them already. Perhaps you can indeed learn while you sleep, but in any case all learning is a little hypnopaedic, in that it requires a kind of lulling of attentiveness that allows learning to pass from message to code, seeping into the unsupervised syntax of your body, as irretrievable as it is unlosable. Rhyme and rhythm are merely the most manifest form of this transaction between the material form of the body and the immaterial form of information, and perhaps there is no knowledge that does not lean towards song: hic,haec, hoc, do, re, mi, thirty days hath September.

This intimates a quiet, huge truth: that the best way by far of producing freedom, delicacy and grace, of body or of mind, and that bridge between them, voice, is through the work of blind, mindless mechanism, urged deep into the body’s habits and manners of speaking and being. This is a much more familiar and intimate kind of unconscious than the Freudian one. For this kind of unconsciousness – my unconsciousness of almost every process not only in what I think of as my body, but also in that industriously preoccupied part of my body, my brain – occurs without my knowing it. Not only is this unawareness no kind of chasm or tragedy, it is the very possibility of my being. I depend on it, I depend on my body’s accents and accidents. My capacity to be me, in excess of the workings of my being, depends on my being able to take myself for granted. Without these orchestrated automatisms, these encrusted accidents, there would be no room, no time, no opportunity for me to emerge at all.

Practice is never more painful than when we are aiming to get something back that has been lost, reversing circumstances in which unwilled or unaware process has had once again to become conscious and manifest. Fluency here becomes task, the spontaneous melody of speech is broken down into its elements. One is forced to reduce oneself to a clumsy mechanism, in order to begin rebuilding the beautiful, incalculable fluency of the subliminal mechanisms, of the thinking machine we are, but do not, dare not, think about lest it fail to work.

How does one consciously learn an accent? The ability to ‘take people off’ (though I always thought of it as forcing them on themselves), was much prized among the boys among whom I grew up. I was not naturally gifted at doing voices, just as I was not a naturally gifted sportsman, but one friend in particular helped me learn how to learn them. The trick, if you want to know, is not to try to capture everything about the voice you wish to mimic or assimilate. Rather, you need to identify one cardinal point, one leading characteristic, of it. If you want to do a caricature of an Indian voice (don’t squirm at this, there are no ways into this kind of thing other than through acts of caricature that we should otherwise find offensive), put the tip of your tongue in the middle of the roof of your mouth and speak. This helps produce the ‘dark’ dental dh sound that is otherwise not usually part of the phonemic repertoire of English speakers. This perhaps assists in switching one’s brain activity from cognitive to visuo-motor skills, or from cognition to performing; Sophie Scott of University College London found that the areas associated with vision and motor control were more active in the brains of skilled impersonators than they were in ordinary people imitating accents (Anon 2008). A similar principle applies to training in sport and dance. Learning to serve a tennis ball by the methodical application of theory involves thinking about too many different things at once to be done consciously. But concentrating on just one aspect of the movement – say, the need to toss the ball high enough that one is drawn up to one’s tip-toes – may be enough on its own to draw the rest of the body, if it has been sufficiently primed by other, rhyming kinds of mechanical practice and memory, into the desired alignment.

This goes against the grain (even though the laying down of grain is precisely what I am getting at here). We are schooled into a preference for alert attentiveness to all aspects of a situation, for flexibility, sensitivity, for what we soothingly like to call holistic understanding, and shun and abominate anything that we feel may thin our profile of possibilities, reducing us to the condition of a mere, mindless machine. But the mnemonic trio prendre, apprendre, comprendre should remind us that we do not need to have what we would learn in mind from the beginning, for mind is the end, not the beginning of learning. And we are anyway never more locked or mechanistic than in thinking about what a machine might be. Play, for instance, is machinery through and through. Spontaneity, invention and grace are not the opposite of machinery, they are its consummation. If you want to know what machinery can do, and what nothing but machinery has the patience, indifference and equanimity to accomplish, try writing the equations that will generate the aurora-shimmers of a flock of starlings, the stippling of a meadow with poppy and cornflower, and whatever is fickle, freckled, who knows how.

Patience is not a particular attitude, it is simply the declining or deferral of impatience, the holding back from the decision that would sidetrack time into space. Patience is play, the patience to keep things in play, which is to say, able to be indefinitely resumed. The instinct of play is what children have in place of patience, and patience is what adults have to compensate for the loss of play. The growth into adulthood is nothing more than the patient relearning of the patience of the child, a patient acceptance that this adult form of play will involve more rigorous and painful schooling than children need. So perhaps patience is the great, utterly unconcealed secret of all thinking, learning, and invention. I think there can be nobody who has ever grown into greatness or happiness who did not have one person in their lives who patiently showed them the omnipotence of persistence. W. R. Bion developed an entire psychoanalytic wisdom around the principle of patience, arguing that only ‘an infant capable of tolerating frustration can permit itself to have a sense of reality’ and seeing knowledge as the result of ‘tolerance of doubt and tolerance of a sense of infinity’ (Bion 1962, 36, 93). Bion used the term ‘patience’ to correspond to the condition of what Keats called ‘negative capability’, that is, ‘when a man is capable of being in uncertainties, mysteries, doubts, without any irritable reaching after fact and reason’ (Keats 1958, I.193; Bion 1970, 123).

There is ordeal and ardour in this undertaking, for patience will always mean a certain acceptance of suffering, being willing to be a patient, and learning above all how to be obediently patient with oneself. Children used to have knowledge beaten into them, but the language of music shows that we cannot reduce this violence to nothing if there is still to be the possibility of learning: we all need at the very least to be able to beat time, to be able to follow a score, to take the strain of learning.

The agony of the hiatus, the aching inhabitation of the space where words will not come, is to be found in Samuel Beckett’s last poem ‘Comment dire’, written in French in 1988 following a fall which left him for a period with difficulties in speech and memory. Beckett translated the poem into English the following year, shortly before his death. The following year, 1990, the Hungarian composer Gyrgy Kurtg used the work for his own last composition, which he wrote for the singer Ildik Monyk, who had been in an accident seven years before and lost the power to speak. Singing the words about the struggle to find words seems actually to have helped her to find her way back to her own voice.

folly for to see what –
glimpse –
seem to glimpse –
need to seem to glimpse –
afaint afar away over there what –
folly for to need to seem to glimpse afaint afar away over there what –
what –
what is the word –

what is the word (Beckett 2002, 115)

The ideal, magic entirety of voice, that ordinarily holds and draws the plainsong of our being together, is here slowed down, broken up and sounded out in a discontinuous sequence. Speaking here consists of patiently piecing together the dispersed apparatus, calling back to mind the saving amnesia that accident has scattered.

The unintelligible language of the foreigner has often been given the name of pure vocal noise, the noise of an improper name, barbarian, berber, hottentot, piccanniny, mumbo-jumbo, gobbledygook, jabberwocky. But unintelligible noise can also provide a place to hide, in the secret languages or cryptolects of many groups. Canting, deriving, like the word accent, from Latin cantare, to sing, was the name for the secret languages of beggars, thieves, prostitutes and other stigmatised groups. There are many such languages, which disguise their intent in the purest racket and blather, such as the quinqui language associated with quinquilleros, tinkers named after the clanking ironmongery that was their trade (named by others, their preferred name for themselves beingmercheros, merchants), or Polari (from parlare to talk), the underground language of British gay men in the 1950s (Baker 2002), and Gayle, its South African equivalent (Cage 2003). In canting and other kinds of argot, the voice imitates the vacuous accents of its own articulation. In the many examples of backslangs that disguise the speaker’s intent by packing out words with the noise of extra syllables, the hurling, slurring hurly-burly of word-work is designed to be all that meets the ear. Here, voice is immersed and dissimulated in its own noisiness, the rattling staccato of the articulatory engine amplified to the point where sense is accessible only to the ear that can listen away, winnowing it of its cacophonous accretions. The disturbing thing about cant and slang languages (and, for their users, the consoling, sheltering thing), is that they seem to be made up of nothing but language itself, voice, or accent apparently separated from any purport. But, if cant comes to mean empty, formalistic jargon, it also hints at enchantment, incantation. Jargon, the mindless chattering or warbling of birds, vox et praeterea nihil, the uttering of what is utterly empty of anything but utterance itself, is thus close both to hocus-pocus and to poetry (Tiffany 2005, 88-90). Elementariness and accent seem to be opposites, but they are not always, and there are surprising shortcuts from one to the other. What knits together the elemental and the accentual is articulation, the word that itself joins together the sundering and the sounding out of speech.


Adank, Pattie, Peter Hagoort and Harold Bekkering (2010). ‘Imitation Improves Language Comprehension.’ Psychological Science, 21, 1903-1909.

Anon (2008). ‘Voice Impersonators Use Brain to Become Characters.’ New Scientist, 26 November, 16.

Baker, Paul (2002). Polari: The Lost Language of Gay Men. London: Routledge.

Beckett, Samuel (2002). Poems: 1930–1989. London: Calder.

Bion, W.R. (1962). Learning From Experience. London: Tavistock.

———- (1970). Attention and Interpretation: A Scientific Approach to Insight in Psycho-Analysis and Groups. London: Tavistock.

Cage, Ken (2003). Gayle – The Language of Kinks and Queens: A History and Dictionary of Gay Language in South Africa. Houghton, South Africa: Jacana Media.

Dolar, Mladen (2006). A Voice and Nothing More. Cambridge, MA: MIT Press.

Goetz, Christopher G., Michel Bonduelle and Toby Gelfand (1995). Charcot: Constructing Neurology. New York and Oxford: Oxford University Press.

Hall, L.C. (1902). ‘Telegraph Talk and Talkers: Human Character and Emotions An Old Telegrapher Reads on the Wire.’ McClure’s Magazine, 18, 227-231.

Keats, John (1958). The Letters of John Keats, 1814-1821. Ed Hyder E. Rollins. 2 Vols. Cambridge MA: Harvard University Press.

Moore, Robert. (2011). ‘ If I Actually Talked Like That, I’d Pull a Gun On Myself.: Accent, Avoidance, and Moral Panic in Irish English.’ Anthropological Quarterly, 84, 41-64.

Ralls, Katherine, Patricia Fiorelli and Sheri Gish (1985). ‘Vocalizations and Vocal Mimicry in Captive Harbor Seals, Phoca vitulina.’ Canadian Journal of Zoology, 63,1050-56.

Serres, Michel (1999). Variations sur le corps. Paris: Le Pommier.

Stevens, Wallace (1972). The Palm at the End of the Mind: Selected Poems and a Play. Ed. Holly Stevens. New York: Vintage.

Taylor, John (1630). All the Workes of Iohn Taylor the Water-Poet. London: J. B. for James Boler.

Tiffany, Daniel (2005). ‘Fugitive Lyric: The Rhymes of the Canting Crew.’ PMLA, 120, 82-96.

Tyack, Peter L. (2008). ‘Convergence of Calls as Animals Form Social Bonds, Active Compensation for Noisy Communication Channels, and the Evolution of Vocal Learning in Mammals.’ Journal of Comparative Psychology, 122, 319-331.