Whisper Music

Steven Connor

A lecture given at the Take a Deep Breath symposium, Tate Modern, 15 February 2007 and at Giving Voice, Centre for Performance Research, Aberystwyth, 28 March 2008.


All voice is shaped breath. But there are two phases to this shaping. There is, of course, the articulation of the breath, the application of stops and delays, and the chivvying of the voice into particular channels, gates and pathways. Thus is the breath filtered and whittled into diction. But, prior to any of this, there is the primary process that phonologists call ‘voicing’, which has already transformed the breath into voice. Though this too results from a constriction, in the forcing of air across the vibrating folds of the larynx, it seems like a charging, or enrichment, as though voice were fuel-injected into the breath by the larynx, as a breeze is infused with the perfumes of the tangled bank.

Voice is the sound made by ensouled creatures, says Aristotle: ‘Voice is a kind of sound characteristic of what has soul in it; nothing that is without soul utters voice’ (De Anima 2.8; Aristotle 1992, 32). Shakespeare could still refer in Julius Caesar to ‘the voice of the lion’, but, for us, animals, though animated, no longer seem sufficiently possessed of anima for voice to be allowed to them. Animals have cries, songs, even vociferations, but it nowadays feels awkwardly catachrestic to credit them with ‘voice’. Aristotle then makes an observation, almost as an afterthought, that will be of considerable moment for me here. Although, he says, only the ensouled can give voice, that is, give soul to sound or sound out the soul, ‘not every sound made by an animal [i.e. an ensouled being] is voice’ (De Anima 2.8; Aristotle 1993, 33). The example he gives is the cough, a sound made by the voice that yet has nothing to do with utterance, since its function is merely physical, entirely, as we might say, animal. The cough is an accidental trespass of unensouled into ensouled sound, a mindless or involuntary convulsion in the continuity of the voice, a rending of the vocal thread. Though the cough may have elements of voice in it, its function, as a mere ‘impact of the breath’ (De Anima 2.8; Aristotle 1993, 33), is to expel irritant matter, not to express thought or feeling.

The cough is perhaps little more than the overtaking of the voice by the unvoiced breath, by a breath that has nothing more than a hydraulic function. Though the cough is often richly voiced, in the phonological sense, its effect is not of the breath regulated or tuned by voice, but rather of the breath barging its way intransigently into voice; the cough is voice coerced by breath, not breath tuned and tutored into voice. The cough is closely twinned with and often implicated in the laugh, and indeed, we might say that laughter is the orchestration of the reflex action involved in the cough. Only humans laugh, we humans have made up our minds about that, even though that very laughter has often been viewed, especially in Hellenic and Christian traditions, as bestial or diabolic. There need be no serious contradiction in this. Only humans can laugh, we seem to think, because only humans have the capacity of being ambushed by the animal they dream they no longer are.

And yet the cough is far from inexpressive, and far from inhuman. How disconcerting it is, for example, to hear an animal – a sheep, cat or dog –  cough. The cough is not an inhuman sound, but the sound of the human being overtaken by something else, or even some other creature (a frog, we are wont to say) or form of creaturedness. And for this reason, coughs can become overlaid with meaning and import, to the point of becoming veritable vocal signatures. Thus, there is an entire thesaurus, a prosody of coughs, from the tussis nervosa of the timorous, to the wine waiter’s  discreetly imperious ‘ahem’. And then there are the incipient, exploratory or aborted coughs, along with the plethora of ways of clearings of the throat, of which no actor has been more the master than Michael Hordern, who could draw harrumphing symphonies out of his slow, growling trawls of the mucous membrane.

The cough is neither the only nor even perhaps the most conspicuous of these incursions of the raw, errant or otherwise unvoiced air into the economy of voice. Let us convene for the moment only the following crew of creole quasi-locutions: the lisp, the gasp, the sigh, the rasp, the whistle, the hiss, the brrr, the purr, the snore, the sniffle, the crepitus, the croak. 

In all of these, the meaning comes from the involuntary nature of the sound, a sound not subdued or wholly suffused by the operations of voice. In such sounds, the air is not expressed, pressed out into audibility, impressed into audible shapes and postures, but seems rather to be escaping, as though through a rent or gash. The lisp is no taut suspiration, but a leak or flatulent collapse.

The early anatomists of voice had two competing theories for the structure and function of the larynx. One saw the voice as wind instrument, as reed, flute or organ pipe. The other saw the voice as stringed instrument. Eventually, the explication of the function of the vocal cords meant that the dispute ended in an honourable draw (Connor 2000, 199-200). The voice was a wind instrument in that it employed air, but a stringed instrument in that its sound was produced by the modulation of strings or cords. The voice could therefore be seen as a kind of Aeolian harp, string sounded by the inner breeze of the breath. These two anatomical alternatives were accompanied by imaginary or ideal forms of the voice. The voice as stringed instrument partook of the lucidity and rational intervals of the Apollonian lyre. The voice as wind instrument was full of reminders of the respiring and expiring human body. In the one, the voice toned the body as tense as a string; in the other, the body, no more than a balloon or saggy bladder, was reduced to wheezing eructation. The lyric voice is virile, virtuoso, inviolate, untouched by human hand; the bagpipe voice is odorous, exhausted and mortal. The story of the contest between Apollo and Marsyas enacts this distinction (Connor 2003). Eliot’s lines from the final section of The Waste Land – ‘A woman drew her long black hair out tight/And fiddled whisper music from those strings’ seems to show the bathetic diminution of the stretched string into sterile, insect-like susurration.

So, although there is nothing in the voice that is not made of breath, though voice is breath through and through, there is yet a ravine that runs through voice, cleaving the true, transfigured voice from the mere unvoiced breath, and holding voice apart from that in the voice that is yet not voice. It is above all the noise of the breath that has seemed to constitute this shadow song, this whisper music, the voice of the unvoiced in the voiced. It should be observed here that, in what follows, it is the idea or the ideal of an absolute distinction between the voiced and the unvoiced that is at issue. As one might expect, the increase in physiological understanding and means of observing and measuring the processes of speech production has complicated this simple distinction between the voiced and the unvoiced. Phoneticians now distinguish (and argue over) many different hybrid types, such as ‘voiced aspirates’, ‘breathy voice’ and ‘whispery voice’ (Stuart-Smith 2004, 162-6). But my concern is not with phonetics, but with phonophenomenology (which will sometimes be to say phoney or funnyfarm phonetics). For there is no theatre of the mind or body that so teems with magical thinking than that which relates to the forms and powers of the voice. The voice is not easily to be distilled out from the complex delusions, fantasies and fixations to which it gives rise, for these fantasies are formative, performative: they determine our comprehension and experience of the voice.

No aspect of phonetics is so overdetermined in this way than the seemingly straightforward and self-announcing distinction between the prime elements of speech. For centuries, Aristotle’s metaphysical claim that only those sounds that have been informed by soul are really voice has been cast in phonetic terms, the distinction between soul and the soulless corresponding to the distinction between vowels and consonants. Vowels are said to be formed in the larynx by the constriction of the vocal cords, producing a musical tone accompanied by harmonics. Consonants are thought to be formed in the mouth, and are the result, as the venerable Henry Sweet describes them, ‘of audible friction, squeezing or stopping of the breath in some part of the mouth (or occasionally of the throat)’ (Sweet 1877, sect. 99). Consonants (‘co-sounders’ - medeklinkers in Dutch) are so-called because they need to be sounded together with vowels in order to form expressive meanings. Consonants do not seem to constitute viable or expressive sounds on their own. A consonant or consonantal cluster can express a feeling  - tsk, ch, grr, pff – but there are few consonantal clusters that can singlehandedly express a concept, or perform a specific grammatical function, as ‘I’, or ‘a’. As John D. Peters has shown in his admirable history of the vowel (2006), the vowels have often been thought of as the soul of speech, with consonants serving for its body. Indeed, vowels have sometimes been thought of as speech itself – active, living and ephemeral as it is presumed to be -  with consonants allied more naturally with the letters which fix and represent it. This may very well derive from, or perhaps itself determine, the fact that music seems fundamentally vocalic and that we have had to wait until Stockhausen, Berio and the ‘extended vocal technique’ of Trevor Wishart for music that takes seriously the musical possibilities of consonants and the unvoiced.

The distinction between vowels and consonants has also been ethnicised, in terms of a distinction between Hebrew, traditionally regarded by Christian commentators as the primary language, the language that was closest to the divine, and Greek. This is because written Hebrew does not phonetically represent vowels, while Greek was the first written language to derive a notation for them. It is on these grounds that, in his Essay on the Origin of Languages (1772), Johann Gottfried Herder distinguishes between the spirit of Hebrew and the dead letter of Greek. For Herder, as for many others, the distinction between vowels and consonants is a distinction within language which is equivalent to the distinction between two forms of language, the spoken and the written. Vowels are the spirit which giveth life, whereas consonants are the letter which killeth. By refraining from writing the vowels, Hebrew is thought to keep them ineffable and inviolate; whereas, by circumscribing the vowels in script, the new Greek dispensation denatures and defiles them:

With us the vowels are the first and most lively thing and the door hinges of language; with the Hebrews they are not written. Why? Because they could not be written. Their pronunciation was so lively and finely organized, their breath was so spiritual and ethereal, that it evaporated and could not be captured in letters. Only for the first time with the Greeks were these living aspirations unraveled into proper vowels, which, however, still needed the help of breathing [Spiritus], etc. – whereas with the Easterners speech was, so to say, entirely breathing, continuous breath [Hauch] and spirit [Geist] of the mouth, as they also so often name it in their painting poems. It was the life-breath [Othem]of God, wafting air, which the ear snatched up, and the dead letters which they painted down were merely the corpse which in reading had to be ensouled with the spirit of life [Lebensgeist]. (Herder 71-2)

A contemporary version of this kind of phonesthetic word-magic is furnished by Michel Serres, who asks us to hear in the rough aspiration of the Hebrew ruagh, spirit, something like the raw, pure audible-sensible of the breathing body, before it has been consigned to the prisonhouse of language:

The first cry of Genesis, at the dawning of the world, above the hubbub, God says ruagh, a rasping alliteration of the breath, at the back of the palate, in the hollow of the throat before language, behind the root of the tongue, there, where the breath clears the throat and recognises the divine; ruagh, breath, breathing, wind, gust of the spirit, at its last gasp, mastering the fanfare of the heart. (Serres 1998, 421; my translation)

But there is another side to breath, the mechanical side, the side that belongs the sonorous engine of the mouth. It is for this reason that Herder, while believing that language derives from animal ‘sounds of sensation’, insists nevertheless that ‘it is no organization of the mouth which produces language … no breathing machine but a creature taking awareness invented language!’ (Herder 2002, 90)



But the sound represented by the letter h is eccentric to most of the schemes for classifying language. Although h is usually thought of as a kind of consonant, it does not perform the function that we have heard is characteristic of consonants, namely that of stopping, delaying or detouring the efflux of air through the mouth. Aspiration is a feature of all of the consonants, with the  sizzle of sibilant ‘s’ and fricative ‘ff’. or the little detonations of air released in the wake of dental ‘d’ and ‘t’, or plosive ‘p’. But the sound supposed to be signalled by the aspirate letter h is a kind of pure debouchure, orally undetained and minimally modified by the mouth. It is, we might say, the degree zero of consonance, It is a consonant in the sense that it is lacking in voice, but vowel-like in that it is open and unobstructed. Indeed, it has sometimes been described as an ‘aspirated vowel’ (de Brosses 1765, 181). Aristotle points to ‘our inability to speak when we are breathing either out or in - we can only do so by holding our breath; we make the movements with the breath so checked’ (De Anima 2.8; Aristotle 1993, 33). It is the absence or minimal presence of this checking in the case of h, which means that it belongs neither to voice nor to noise, neither larynx nor mouth.

In one sense, the letter h represents a perplexing, even a menacing anomaly; its existence being wholly accessory or parasitic. And yet, because of this, it is everywhere, not just at the beginning of English words, where it holds its place of honour, but also secreted semi-silently within them, as in combinations like ch, and gh, and words like cough and enough. Ben Jonson, who helped make his name by suppressing the h in the middle of it, noted the ambivalence of the letter at the beginning of his English Grammar, saying that ‘H. Is rarely other than an aspiration in Power, though a Letter in Form’ (Jonson 1640, 35). The slight flicker of meaning in the phrase ‘aspiration in power’, which seems to allow us to think of the aspiration to power as well as the power of aspiration, is typical of Jonson’s wry facetiousness at moments in this text. His later expansion of these remarks turns on a play between Biblical letter and spirit:

Whether it be a Letter or not, hath been much examined by the Ancients, and by some of the Greek Party too much condemned, and thrown out of the Alphabet, as an Aspirate meerly, and in request only before Vowels in the beginnings of words. The Welsh retain it still after many Consonants. But, be it a Letter, or Spirit, we have great use of it in our tongue, both before, and after Vowells. And though I dare not say, she is (as I have heard one call her) the Queene mother of Consonants: yet she is the life, and quickening of them (Jonson 1640, 48)

But the ubiquity of h means that it can also be read symbolically as the principle of conjunction, as the necessity of commixture, which makes it appropriate for discussions of sexual complementarity or conjuncture. Johann Buxtorf tells us, for example, that  ‘Isch [Hebrew man] differs from Ischah  [woman] only by the letter H  which is an aspiration, noting that the woman was made of man, and as it were breathed out of his side’ (Buxtorf 1657, 42n). Another, lengthier reflection on the hymeneal conjuncture implied by the h goes further by marrying the graphic and the phonic properties of h:

But of all letters, it is the hardest for the body of man or Woman, alone of it selfe to imitate an H. For it consists of two severall disjunct, parts of letters: that is to say of two I I: both which are signes of the singular and first person; and are of themselves, both good formes of building too, but unles there come some-what, that (after a friendly manner) may joyne them together, they both still remaine singular and alone: and the building can never come into its desired and beautifull forme. Wherefore, if either man or woman, (being alone and built according to the singular and first person I) doe desire to change for a better: There is no better way to establish and make them most firmely grow into this well approved forme, then (by the love of their hearts) to reach each other their hands in direct sinceritie, thus, I---I: And let the even and straight course of marriage, fully and firmely establish them into one letter, H. Which not only by uniting of two bodies, makes them eaven: but by bringing them into the forme of this letter H, makes their eaven, Heaven: if they continue in the love, which first joyned them: which is, indeed Heaven upon earth. (Austin 1637, 82-3)

The aspirating function of h has led to difficulties and disputes for centuries in Europe. Cicero grumbled about the influx of affected h sounds in Latin, which perhaps arose in imitation of the imagined elegance of Athenians, who aspirated more obviously than Romans, while his contemporary Catullus has a poem mocking one ‘Arrius’ who adds hs illegitimately or ridiculously wherever aspiration is possible: ‘Chommoda dicebat, si quando commoda vellet/Dicere, et hinsidias Arrius insidias’: ‘ “Chomfortable”, said Arrius, when “comfortable” was what he wanted to say,/ and “ambush” came out as “hambush” ’ (lxxxiii; my translation). At the end of the fourth century AD, Augustine regretted the hypersensitivity of grammarians of his own time towards incorrect aspiration: ‘si contra disciplinam grammaticam, sine aspiratione primae syllabae ’ominem dixerit, displiceat magis hominibus, quam si contra tua precepta hominem oderit, cum sit homo’ - ‘if someone should, against the strictures of grammarians, pronounce as ’ominem the opening of ‘hominem’, then he will be abominated by more men than for illicit hatred ad hominem’ (Confessions 1.18; my translation gives Augustine’s Latin a perhaps gratuitous helping hand). Jonathan Sheehan has shown the centrality of the letter h, which is used principally to lengthen median vowels in German, as in Bahn, to discussions of orthographic reform in eighteenth-century Germany.  On the one hand, rationalist reformers attempted to ‘tame the overuse of the letter h’ in the interests of ‘a unified pure High German’ (Sheehan, 35). Against them stood writers like the critic Johann Georg Hamann, who followed Jakob Boehme in finding a mystical power in the letter; according to Boehme, the insertion of the character signifying the breathing h into the five vowels of the tetragrammaton ‘shows how the holy name breathes itself out’ (Boehme 1730, 331-2, quoted Sheehan 1998, 37).

But aspiration has probably never had a more fussed and tussled-over history than in British English. Following the Norman invasion of Britain, h sounds, which had often been more guttural than now, and resembled ch or gh, began to become less marked in English, in line with the tendency for aspirates to vanish in Romance languages like French and Italian. Pronunciation and orthography reinforced each other, with ‘ostler’ replacing ‘hostler’ for example. This parallels the development of ch and gh, which have tended either to move forward from the throat into the middle of the mouth, losing much of their expectorant attack, or to be hardened into the letter ‘k’, prior to their being passed over in silence, as witnessed by the mute ks at the beginning of words like ‘knight’, ‘kneecap’ and ‘knowledge’, and the mute gs of ‘gnaw’ and ‘gnome’.

But for some reason English began to resist this erosion and a spirited revival of the aspirate h set in from the eighteenth century onwards. It is as an effect of this revival that the general tendency to drop (or disdain) initial aspirates becomes read as idleness or ignorance, and the careful highlighting of hs a sign of careful and correct attention to the detail of language. The most surprising of the ways in which this has happened is in the pernickety pronunciation of the wh of words like which, when, whisk and, of course, whisper. Speakers who take care to emphasise these sounds in words spelled with wh may well imagine that they are pronouncing the word as it is written, though in fact they are recalling the hw of Old English words like hwy, or, most notably, the untranslatable calling-to-attention of the word Hwaet! (‘Oi! Listen up!’) that opens The Dream of the Rood.

Among some linguists, the h became a sign of the burly native vitality of English speech, as opposed to the effeteness of the French invader. John Spanton writes that ‘[a]s England became a stronghold of Norman feudalism, the aspirates and gutturals of the Saxon element were often suppressed; thus diluting the vigour of the English speech by an infusion of the Norman element’ (Spanton 1894, 8). One of the doughtiest defenders of the h was Alfred Leach, whose The Letter H Past and Present (1880) promises to show ‘the antique origin, the unbroken line of descent, and the rough, sturdy ancestry of our English H’  (Leach 1880 33). Leach sees a direct line of transmission from the ‘spiritus asper’ or ‘rough breathing’ marked in Ancient Greek - a sign derived from the left-hand half of a bifurcated Greek H  - to the guttural aspirates of Celtic and British words. Accordingly, he applauds the revival of the h sound during the nineteenth century – signalled most conclusively in the word ‘herb’, which was widely pronounced for much of the nineteenth century without an initial aspirate, as it still is in most parts of the US – as a reassertion of the hearty British spirit after the waning of Norman influence: ‘when the language of the vanquished began to overcome that of the conqueror, the Aspirate must have entered upon a new era, and H’s again have prevailed in the land’ (Leach 1880, 33). Oddly, but entirely typically, the dropped h, or, more accurately, the indifference to h, that is characteristic of many dialects, especially Cockney, has also been defended on precisely the same grounds, as an expression of a pungently indigenous nonconformism.

Nevertheless, despite the rallying of the fortunes of h, Leach acknowledges the melancholy and in the end perhaps irresistible course towards oblivion on which the sound and the letter are set. Already, ‘the new H had not the vigour of the old one – the guttural ‘ of the Anglo-Saxon….the powers of H were gradually, surely, and steadily waning, until, at length, its strong guttural sound finally and completely evanesced towards the latter half of last century’ (Leach 1880, 33). His book ends with the acknowledgement that

the tones and modes that constituted nature’s primitive eloquence must fall gradually into disuse. The strong breathing and the guttural breathing, having been the most expressive emotional interpreters of the early savage, are repugnant to the artificial sedateness and studied reserve of the modern speaker. In the speech of the well-bred Englishman, the hale old English H has melted into a soft Aspirate, and even this is likely to be soon altogether lost (Leach 1880, 81)

Indeed, somewhat improbably, Leach sees the aspirate sound, not just as an individual tragedy, but rather as the route to desuetude taken by many sounds

Any letter doomed to die out of a word or a language, generally attempts to depart gracefully by first acquiring the nature of an aspirate-consonant, and then turns into a perfect H; under this form it relies upon h-dropping mortals to give it quiet burial, and unobtrusively confide it to Oblivion. (Leach 1880, 81)

The now-you-hear-it-now-you-don’t condition of h in English made it the target of heightened vigilance and anxiety from the eighteenth century onwards. The already ambivalent condition of the aspirate is redoubled by  its vacillating position in English writing, as a dubious mute. As such, it becomes an authentic shibboleth, so named after the word that the Ephraimite people, attempting to flee from the victorious Gileadites over the river Jordan, could not pronounce, those who betrayed themselves by lisping it in the Ephraimite fashion as sibboleth, having their throats slit by the stickler Gileadites (Judges 12: 5-6). Most authorities interpret the word shibboleth as meaning ‘an ear of corn’, though there is a minority opinion that it refers to a stream or torrent, which obviously has a neat relation to the site of linguistic inquisition. Similarly, there seems to be some subliminal communication here between sound and script in the case of h, for the form of the capital letter H is believed to derive, through Greek, from a Phoenician letter, itself modelled on an Egyptian hieroglyph, having the form of a three-barred fence, and thus picturing the act of regulated passage.

From the middle of eighteenth century onwards, a stream of popular publications defined and disputed the correct pronunciation of the initial letter h, sometimes cast in the form of comic petitions voiced by the letter itself. Poor Letter H (1854) is an address to the vowels by the letter h, punningly appealing for regulation in its use: ‘sometimes I have the most honourable aspirations to be first and foremost; and then at other times I am so humble that I only want to let my next little brothers speak; but they must speak softly, or maybe I shall be offended’ (H. 1854, 4).

In fact, one of the many fallouts from this was a minor perturbation in the pages of Notes and Queries regarding the pronunciation of the word ‘humble’, which is so distinctively and rendered as ’umble’ by Dickens’s hypocritical Uriah Heep, whose equivocal name contains both a voiced and an unvoiced h. This was protested against by J.S. Warden, who wrote that ‘I was always taught in my childhood to sink the h in this word, and was confirmed in this habit by the usage of all the well-educated people that I met in those days… but this eminent writer [Dickens] has thought fit of late to proscribe this practice by making it the Shibboleth of two of the meanest and vilest characters in his works’  (‘The Letter ‘H’ in Humble’ 1853, 54). A number wrote seconding the correctness of the pronunciation ’umble’, though one anonymous correspondent responded with verses in favour of retaining the distinctive English h, in lines that consciously or unconsciously recall Augustine’s quibble on the idea of a phonetic crime against ‘humanity’:

Habituated to the sound of h
In history and histrionic art,
We deem the man a homicide of speech
Maiming humanity in a vital part,
Whose humorous hilarity would treat us,
In lieu of h, with a supposed hiatus. (‘The Letter ‘H in Humble’ 1853, 298)

This is in partial imitation of a pedagogic riddle by the poet Catherine Fanshawe (though sometimes credited to Byron) which was frequently reprinted in the second half of the nineteenth century:

’Twas in heaven pronounced, and twas muttered in hell,
And echo caught faintly the sound as it fell;
On the confines of earth twas permitted to rest,
And the depths of the ocean its presence confest.
Twill be found in the sphere, when tis riven asunder,
Be seen in the lightning, and heard in the thunder.
Twas allotted to man with his earliest breath,
Attends at his birth, and awaits him in death,
Presides oer his happiness, honour, and health,
Is the prop of his house, and the end of his wealth.
In the heaps of the miser tis hoarded with care,
But is sure to be lost on his prodigal heir.
It begins every hope, every wish it must bound,
With the husbandman toils, and with monarchs is crown’d.
Without it the soldier, the seaman may roam,
But woe to the wretch who expels it from home! 
In the whispers of conscience its voice will be found,
Nor een in the whirlpool of passion be drown’d.
Twill not soften the heart; but though deaf be the ear,
It will make it acutely and instantly hear.
Yet in shade let it rest like a delicate flower,
Ah, breathe on it softly – it dies in an hour.  (Fanshawe 1865, 41-2)

The huffing and puffing and hectoring of the uneducated about so-called ‘dropped aitches’, and the apprehension at the impending extinction of the h  produced a nervous desire to multiply hs where there was no warrant for them, a  ‘a pervert method of aspirating’ (Leach 1880,15) that drew even more scorn upon the hapless culprit than the innocent ablation of the h, with all the ferocity reserved for the failed social aspirant: ‘It is not as a rule the very poor who introduce h’s, but the small shopkeeper and the villager who reads at home in the evening instead of going to the public-house’ sniffs one commentator (Hill 1902, 43). Most remarkably, the name of the letter h was subject to this supererogation. The word aitch derives from Teutonic ache and, by the seventeenth century, was routinely pronounced ake, in conformity with the tendency already noted for ch and gh words to be baked into k. Oddly, French, which has almost entirely extirpated the sound of the h, preserves much more of the aspiration in its name for the letter – asch. But the recent revival of the aspirate sound in English makes the name for the letter seem like an anomaly – for, unlike any other letter of the alphabet, apart from w, double-you and y, wy, aitch contains no announcement of its own sound. There is an ache in the word aitch, we may say, for the exemplary aspiration of which it has been deprived, a fact that encouraged John Heywood in 1550 to make not-very-hilarious hay from h’s aches and pains:

H, is worst among letters in the crosse row,
For if thou finde him other in thine elbow,
In thine arme, or leg, in any degree,
In thy head, or teeth, in thy toe or knee,
Into what place so ever H, maie pyke him·
Where ever thou find ache, thou shalt not like hym. (Heywood 1550, sig c1v)

It is in order to make good this imagined mutilation that the word aitch has in many quarters been given artificial resuscitation to become haitch, a usage which it seems to me is currently gaining ground, or perhaps simply coming out of hiding. The anxious aspirator is like the fetishist, who, according to Freud, goes about patching up imagined amputations of the sexual organ with substitutes, succeeding only in drawing attention to the fact that nothing was missing in the first place. There is perhaps an analogy here between the peephole logic of the h and Roland Barthes’s reading of the castration complex embodied in the difference between z and s, the sibilant half-sisters of h, in his brilliantly overblown S/Z. a study of Balzac’s story of the love between the sculptor Sarrasine and the eunuch singer La Zambinella. There are two curiosities to be observed in Barthes’s text. The first is that he remains tight-lipped about the fact that s and z are conventionally distinguished as unvoiced and voiced respectively, an intriguing omission in a text that is so absorbed with the relations between voice and castration. The second is that Barthes’s gaze seems to have been snagged by the scissoring zipper of the Z on the page, so that he reads the Z in the name of La Zambinella, not as voiced virility, but  as ‘the letter of mutilation’ (Barthes 1974, 106), and carries this across to its sound: ‘phonetically, Z stings like a chastising lash, an avenging insect’ (Barthes 1974, 106).


Whisper Who Dares

Until the beginning of the twentieth century, unvoiced consonants were known to phoneticians as ‘whisper-letters’. The whisper is distinguished from the lisp, hiss, sigh and whistle. For these are local or accidental occasions of the unvoiced puncturing the flow of discourse. But the whisper is distinguished by being an entire mode of discourse in itself. Whispering is not something that happens in speech, it is something that happens to it. The whisper is a spectre-speaking, a mirror or mode of minority for speech. The consonants were sometimes regarded as a kind of necessary abeyance, an active gap in nature; so Richard Baxter advised his readers that ‘[t]he night is part of the useful order of the creation, as well as the day. The vacant interspace in your writing, is needful as well as the words: Every letter should not be a vowel, nor every character a Capital’ (Baxter 1664, 83). But the whisper is not the chiaroscuro provided by the interplay of death and life in the consonants and vowels; rather, it is entirely nocturnal, the whole of speech transposed into the key of H. The whisper is not the complement of speech, as the consonant is the complement of the vowel; it is its fetch, phantom or facsimile, which doubles without touching on its original.

The whisper is not only voiceless, it is also, and for that reason, mindless. It is perhaps because of this that whispering seems so often to arise as an accidental or imagined locution produced by the passage of air through some merely material obstruction. The wind whispers, especially in leaves and grass. In the story of Midas, the king’s counsellor cannot keep the news about his master’s ass’s ears  to himself, and so confides it to a brook, only for the whisper to be taken up by the reeds and bruited across the countryside.

The whisper signifies intimacy and secrecy. It is the mode in which I most naturally speak to or overhear myself. As such, it has religious or supernatural overtones, the whisper being the favoured mode of communication both of angels and of demons. The intimacy of the whisper gives it strong erotic overtones, too as in the many popular songs in which whispering features. The practice known as ‘horse-whispering’ suggests that the whisper allows one to cross the linguistic barrier separating the human from the animal.  The successful novel (Evans 1995) and film of this name has given rise to a series of books with titles like The Cat Whisperer (Bessant 2001), The Dog Whisperer (Richardson and Cole 2001, Owens and Eckroate 2007), The Rabbit Whisperer (Tarrant 2005) and, most intriguing of all, The Tarantula Whisperer (Pasten 1999), all promising insights into how to communicate confidentially with one’s pets. More remarkable still is the phrase a ‘pig’s whisper’, which was current until the beginning of the twentieth century, and was used to mean a jiffy, or brief instant, or as an emphatic form of the ordinary variety of whisper.

But the whisper signifies not just the keeping but also, as in the Midas story, the incontinent spilling of secrets. The whisper is devious and dangerous and, as such, is close to rumour. For, if the whisper appears always to be so close as for it to be uncertain whether it belongs to the inside or the outside, the whisper is always also scattered, with no fixed abode. Its very lack of amplitude is what seems to allow the whisper to be so easily amplified, as suggested by Swift’s claim to know ‘a Lie that now disturbs half the Kingdom with its Noise, which I can remember in its Whisper-hood’ (The Examiner, 15.5, 1710). The observation, in Shakespeare’s Henry V, that ‘From camp to camp through the foul womb of night/The hum of either army stilly sounds,/That the fixed sentinels almost receive/The secret whispers of each other's watch’ (Henry V, iv. prologue), draws together proximity and propagation in a telling way.

None of the mentions of whispers or whispering in the King James bible are reassuring or approving. Whispering is the form of conspiracy (which, as ‘breathing together’ means whispering); and whispering is another term for the kind of soothsaying or sorcery that is routinely forbidden through the Old Testament. As James Mason’s Anatomie of Sorcerie (1612) explains, the word ‘charmer’, against which Psalm 58.5 warns,

doth naturally betoken one that whispereth, muttereth, or mumbleth, speaking softly as it were betwixt the teeth. And because the charmers, and inchaunters do so, as it is manifest by experience, and likewise by the 8. cap. of Esa. vers. 19. in these words. And when they shall say vnto you, inquire at them that haue a spirit of diuination, and at the south-sayers which whisper and murmur, &c. and Esa. the 29. cap. vers. 4. Thy voice shall be also out of the ground, like him that hath a spirit of diuination, and thy talking shall whisper out of the dust.’ (Mason 1612, 63-4)

Whispering is associated with the forms of speech also specified as ‘peeping’ and ‘chirping’ at various point with the Old Testament, terms which are associated with the practice of ventriloquism. Ventriloquism was understood at this period, not as a throwing of the voice into the body or person of another, but as a speaking through the body from some place, or by some means, other than the mouth, an improper or displaced form of speaking which might then appear to be coming from elsewhere, and to be magical or devilish. The whisper is this voice, embodied, but without abode. Indeed, ‘whisperer’ was occasionally used as a term for a ventriloquist, for example of one Mr Fanning, a ventriloquist at the court of Charles I who was known as ‘the King’s Whisperer’ (Edmund Dickinson identifies him as ‘Henry King’, though nothing else seems to be known about him than that he resided in Oxford). ’On Mr Fanning the Engastrometh’, a short poem in celebration of his powers, records:

To speak within, and to ones selfe, and yet
Bee heard, is much, yet Fanning doth it:
So tall and stout a man, ‘tis strange to see’t
So like a coward should his words down eat:
The belly hath no ears they say; yet his
Hath ears to hear, and tongue to talk, I wis. (Heath 1650, 37)

The whisper is a speech that appears to be internal, a closet speech or ‘speaking within’, that has insufficient projective force to get untangled from the thicket of tongue and teeth which gives rise to it. And yet, if it holds back from utterance (the word essentially meaning ‘outing’, putting out or bringing forth), a whisper also seems to have no interior core or kernel. For, as the shell, shadow or outward semblance of speech, it is a kind of feigning or counterfeiting out of which all colour, body and melody have been drained. The whisper is kept inside, held back from speaking out loud, and yet it has itself no inside. The whisper is like a sketchy blueprint of a voice, an attenuated ‘fadograph’, to use Joyce’s delicate minting (Joyce 1975, 7). Perhaps the fact that the whisper has neither interiority nor exteriority explains why the two conjoined modes of the whisper are the secret and the rumour.

But if the whisper is so attenuated and spectral, why does it appear so powerfully seductive or so urgently demanding? Why does the whisper have such designs on us? Perhaps it is precisely because it gives us too little, that the almost-but-not-quite nothing of the voice-that-is-not-one, thereby craves from us the making of a voice-body of compensatory intensity. Perhaps all whispers are relatives of the shades of the underworld summoned in Book XI of the Odyssey, who must lap from a trench of milk, honey and blood in order to plump their twittering voices out into audibility.

If the sound of raw or unprocessed air suggests contamination or collapse, then the word ‘aspiration’ should remind us that there is another mode of the unvoiced air. The audibly unvoiced breath is also implicated in expressions of yearning. It is unlikely that any real phonetic difference is signalled in the spelling of ‘Oh’ rather than ‘O’, but the visible presence of the ‘h’ serves to signal the passage of the air, unheard, but tangible, through the aperture of the lips. ‘Oh’ can signify longing, pain, excitement, rapture, intensity of bliss, precisely through the temporary ebb or overcoming of the articulate voice.

So the aspirate air is both carnal flatus and divine afflatus. As such it participates in the duality of air as such, which is never less than double, never unaffected by the ambivalence of the pure and the impure. The unvoiced air moves between the significant and the senseless. In one sense, it is significance broken in upon by the merely phenomenal, unshaped air. But these occasions of incursion are in fact laced intricately through the fabric of speech. In this sense, voice is suffused by the voiceless. Perhaps the secret of h is that it is the ultimate mixed body, the Hermes of the crossroads where vowel and consonant, voice and void, sound and sense, soul and machine, meet and have their fluctuant commerce. The errant sign, sounded on the breath, of the unsteady breath itself, that is ever neither ever absent or present, is never quite held, or ever quite gives out, that is, in Hopkins’s words ‘needful, never spent’ (Hopkins 1970, 93), intimates the cryptically indigent affluence of the air in the dreamscene of language. Here, in the twilight epochs of speech’s intermission, language gives up the ghost, and voice catches its breath.



Aristotle (1993). De Anima Books II and III. Trans. David Hamlyn. Oxford: Clarendon Press.

Austin, William (1637). Haec Homo: Wherein the Excellency of the Creation of Woman is Described, By Way of an Essay. 2nd edn. London: for Ralph Mabbe.

Barthes, Roland (1974). S/Z. Trans. Richard Miller. New York: Hill and Wang.
Baxter, Richard (1664). The Divine Life. London: for Francis Tyton.

Bessant, Claire (2001). The Cat Whisperer: The Secret of How to Talk to Your Cat.
London: John Blake.

Boehme, Jakob (1730). Mysterium Magnum, oder Erklärung über das erste Buch Mosis. N.p.

Brosses, Charles de (1765). Traité de la formation méchanique des langues et des principes physiques de l’étymologie. Paris: Saillant.

Buxtorf, Johann  (1657). The Jewish Synagogue, or, An Historical Narration of the State of the Jewes at This Day Dispersed Over the Face of the Whole Earth. Trans. A.B. London: for H.R. and Thomas Young.

Connor, Steven (2003). ‘Windbags and Skinsongs.’ Online at <http://www.bbk.ac.uk/english/skc/windbags/>
------------------- (2000). Dumbstruck: A Cultural History of Ventriloquism. Oxford: Oxford University Press.

Evans, Nicholas (1975). The Horse Whisperer. London: Bantam.

Fanshawe, Catherine Maria (1865). Memorials. Westminster: Vacher and Sons.

H., Hon. Henry (1854). Poor Letter H: Its Use and Abuse. 2nd edn. London: John Henry Jackson.

Heath, Robert (1650). Clarastella Together With Poems Occasional, Elegies, Epigrams, Satyrs: Epigrams. London: for Humph. Moseley.

Herder. Johann Gottfried (2002). ‘Treatise on the Origin of Language.’ (1772). In  Philosophical Writings, trans. and ed. Michael N. Forster (Cambridge: Cambridge University Press, 65-164

Heywood, John (1550). An Hundred Epigrammes. London: Thomas Berthelet.

Hill, Geoffrey (1902). The Aspirate: Or the Use of the Letter “H” in English, Latin, Greek and Gaelic. London: T. Fisher Unwin.

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

Jonson, Ben (1640). The English Grammar. In The Workes of Benjamin Jonson. London: s.n. 1631-41.

Joyce, James (1975). Finnegans Wake. London: Faber.

Leach, Alfred (1880). The Letter H Past Present and Future, A Treatise: With Rules for the Silent H, Based on Modern Usage; and Notes on WH. London Griffith and Farrar.
‘The Letter “H” in Humble’ (1853). Notes and Queries, 8.194, July 16th, 54; 8.201, September 3, 229-30; 8.204, Sept 24, 298

Mason, James (1612). The Anatomie of Sorcerie Wherein the Wicked Impietie of Charmers, Inchanters, and Such Like, Is Discovered and Confuted. London: John Legatte for Simon Waterson.

Owens, Paul, and Eckroate, Norma (2007). The Dog Whisperer: A Compassionate, Nonviolent Approach to Dog Training. Avon, Mass.: Adams Media; Newton Abbot: David & Charles.

Pasten, Laura (1999). The Tarantula Whisperer: A Celebrity Vet Shares Her Secrets To Communicating With Animals. Berkeley, Cal.: Conari.

Peters, John D. (2006).  ‘A Short History of Vowels.’ Lecture at Sounding Out 3, University of Sunderland, 6 September.

Richardson, John and Cole, Leslye Sharon (2001). The Dog Whisperer: The Essential Guide to Understanding and Raising a Happy Dog. Sydney and London: New Holland.

Serres, Michel (198). Les cinq sens. Paris: Hachette.

Sheehan, Jonathan (1998). ‘Enlightenment Details: Theology, Natural History, and the Letter h.’ Representations, 61, 29-56.

Spanton, John (1894).  The Letter H: Its Old and Modern Uses. Ramsay: S.W. Lynn.

Stuart-Smith, Jane (2004). Sound-Change in Italic. Oxford: Oxford University Press.)

Sweet, Henry (1877). A Handbook of Phonetics.

Tarrant, Ingrid (2005). The Rabbit Whisperer. London: John Blake.