Anarchic Transformations: Trevor Wishart & Vox 5

Hola my lovelies! After the break is the text of an essay comprising my thoughts on Trevor Wishart's seminal work Vox 5, which has been a massive inspiration to me. Wishart's stereo mix of the piece itself is embedded just here, for your listening pleasure:

(PS ...I apologise for the lack of footnotes, but seriously, what do you want for nothing? Rubber biscuit?)


Leigh Landy has noted that a valuable method of achieving a successful appreciation of any given composition can be found in the answers to three very simple questions: ‘what’, ‘how’ and ‘why’. To make this point explicit may seem unnecessary; however, Landy asserts that discussion of the ‘why’ has often been found lacking in, if not altogether absent from, analysis of the work of modern composers who integrate electronic processes into their music. In particular, he stresses the need for an understanding of the ‘why’ - what he describes, in theatrical terms, as “the dramaturgy of a composition”, as a potentially helpful “grip to (first) listeners where relevant”, and suggests that, for this reason, “such an aim should always be encapsulated in our programme and CD booklets”. Fortunately, Trevor Wishart is a composer who has made both his working methods and his creative philosophy very clear to those who wish to research them. However, I have still found certain aspects of the ‘why’ of VOX 5 to be rather underrepresented in many of the articles to discuss the work, a point that I shall return to below. It shall be the this essay’s endeavour, then, to explicitly address all three of Landy’s questions when considering the piece and its context.

The first question that must be answered is, of course, the ‘what’, a clear understanding of which is a necessary grounding for any discussion of the ‘how’ or the ‘why’. VOX 5, then, is an electroacoustic piece of just over six minutes’ duration, composed by Wishart in 1986, using the computer workstations at the Institut de Recherche et Coordination Acoustique / Musique (IRCAM). VOX 5 is based around a “supervoice”, representative of the Hindu god Shiva, built from samples Wishart had made of his own extended vocal techniques. This voice metamorphoses into and through various other sound-events taken, in the main, from the natural world. The piece is intended to be heard in four-channel surround sound (although a stereo version is also available on CD), and it is, as its name suggests, the fifth section of Wishart’s VOX cycle, forming, in the composer’s own words, “a bridging recitative into the finale of this series (VOX 6), and... also [acting as] a “poetic summary” of the preceding material”. The cycle itself is intended as a musical narration of myth of Shiva: the ‘preceding material’, VOXs 1 to 4, chart the various aspects of Shiva’s role as creator-destroyer, while VOX 6 represents a summative dance on the part of the god (the dance of destruction being a key element of the Shiva myth).

This short description of the work is already beginning to beg a number of ‘why’ questions that must be dealt with. Not the least of these concerns Wishart’s choice of the electroacoustic medium itself; an issue to which he devotes a significant proportion of his 1985 book On Sonic Art. Wishart explains that his philosophy of composition can be explained in terms of his wishing to move away from what he calls “lattice-oriented” music (ie that based on traditional notation, the composition of which, it is argued, necessarily becomes subsistent upon the primary need for sound-objects to be ordered on a stave according to their pitch and duration), towards a method of composition whereby any aspect of a sound can be taken as its primary musical raison-d'ĂȘtre. That there is a whole world of sound which remains unnotatable in the western musical tradition is a simple point to prove: by its very nature sound is, after all, a continuum, and “the infinite is not simply notatable”, especially when one is constricted to only 12 fixed pitches and durations easily divisible by the number of beats in any given bar. Furthermore, it is not simply an infinitude of pitches and durations that we are presented with: Wishart notes, too, that a fully-formed concept of timbre is conspicuous by its absence from the two-dimensional ‘lattice’. It is true that timbral concern does have a place in traditional western notation inasmuch as scores will include an indication of the instruments to be used in the realisation of a piece. However, this practice in no way recognises the nature of timbre as a multi-dimensional and infinitely variable aspect of sound. Crucially, too, this multidimensional infinity of timbral possiblity is true for any sound: as Wishart notes, it is wrong to assume that “one only runs into new, non-lattice-based, conceptions of musical ordering when dealing with sounds having complex mass or noise characteristics”. In short, what Wishart hopes to achieve is a “direct musical communion” with his audience, rather than being reduced to making music mediated by the clumsy signifiers of what he sees as a needlessly oppressive system of notation.

To articulate meaning without relying on outmoded signification, Wishart talks of the creation of “sound landscapes” within which “musical gesture” can function. By ‘sound landscape’, Wishart is referring to those aspects of the sound experience of an acousmatic piece, the source of which can be ‘recognised’ by the audience. ‘Recognition’, here, is not taken to mean a full and true understanding of the source and history of each individual sound-object; rather, we are talking of “the nature of the perceived acoustic space [the sense of width and depth provided by, for example, reverberation]; the disposition of sound-objects within the space [their position in and movement through the space]; [and] the recognition of individual sound-objects”. Wishart notes, too, that one can play with the audience’s perception of sonic context by using various combinations of real and unreal sound-objects placed in real and unreal spaces. In this way, the sonic ambiguity inherent in acousmatic music can be used as a compositional tool rather than a barrier between the composer and audience.

Landy has talked of the ‘something to hold on to’ factor in this context: the creation of sound landscapes that are not altogether alien to the listener who “is more interested in the appreciation of music than in being confused by a given work”. It is, perhaps, partly for this reason that Wishart has been an enthusiastic champion of the use of extended vocal technique in electroacoustic composition. Stephen Connor has asserted that Trevor Wishart’s work “represents one of the most sustained attempts to enlarge the conventional repertoire of the voice with noise, not just of inhuman sound, but also the noise components of the voice itself”. Wishart himself explains that it is the cultural resonance inherent in all vocalised sound that allows him to experiment in this way without losing his audience’s trust: he notes that “where human beings are heard to produce sounds, then we will tend to impute intention to the sonic event. We will hear it at some level as an utterance”. The vocalisations of VOX 5, therefore, imbue the piece with a certain air of deliberate intent, even before we take into account their musical arrangement: put simply, if we are presented with a human voice, we naturally assume that what we are hearing must mean something (even if the vocalisation is alien enough that we struggle to ascertain exactly what this meaning is).

Indeed, there are a multitude of clues in any vocal sound that give us insight into the intent behind it: any word (and, by extension, any vocal articulation at all, whether recognised as language or not) can by articulated with anger, surprise, disappointment, love or fear - Wishart’s own Red Bird, in which “the various phonemic objects contained in the phrase ‘listen to reason’ were spoken by a male and a female speaker in a number of different ways indicated by adverbs”, contains exemplary proof of this. Furthermore, Brattica and Sassanelli have noted that certain phonemes have a kinaesthetic quality to them, rendering them linkable to objects and activities in the world outside one’s own mouth: they explain that “we feel the formations inside the mouth and thereby associate them with the activities or the shapes of objects in the external world”. Wishart is also well aware of this phenomenon, and provides a number of examples in On Sonic Art: for example, the phoneme Spr- can be seen to represent “a flow (or store) of energy (s), passed through a constriction (p) and continuing but being broken up into an iteration (r-), as in spray, sprinkle, sprout, spring”. This experience is universal: similar phonemic analogs can be found in a great number of widely disparate languages.

Of course, one need not necessarily be speaking any language for these aspects of vocal meaning to be relevant, and this is a point worthy of consideration when one is discussing a piece based around the voice of a god - which, by its very nature, we might expect to encompass the totality of human language whilst, at the same time, being some degree removed from it. Shiva’s voice must, essentially, be recognisable and alien at the same time, and it is thanks to the sheer range of the human voice, in terms of sonic potential, that this is an aim that is achievable in a piece of music such as VOX 5. Indeed, this is part of that which first attracted Wishart to the possibilities of the voice as a musical instrument. He sees the voice as being “intrinsically multi-media”, able to encompass various concepts of sound, song and utterance at once. In Wishart’s own words, he simply found that, in the days when he was forced to work with very limited studio equipment, “the most manipulable sound-source of all was my own voice”. As such, any composer using voice-objects has at his disposal a massive repertoire of ‘recognisable’ sounds, and by concentrating on this area of sonic possibility Wishart has allowed himself the freedom to distance himself from ‘lattice-based’ events and employ a compositional method in which timbre is of primary musical importance: he can play with phonemes and vocal noises, transforming and manipulating them so that their meaning is obscured or even radically altered, whilst at the same time being confident that he is using a source-material that is intrinsically understandable by virtue of its human origin. It is in this context of timbral exploration by way of vocalisation, therefore, that VOX 5 should be analysed.

As such, a short word on the techniques employed in the composition of the piece - its ‘how’ - is in order. In order to impose both narrative meaning and musical order on his collected vocal fragments, Wishart transforms his material, creating timbral metamorphoses that link two (or more) sound-objects into one single sound-event (or, as Landy has it, a single “sonorous gesture”) which unfolds over time. Timbral transformation is not a concept unique to electroacoustic music: Landy notes, for example, Haydn’s interest in “instrumental exchanges”, typified by his common practice of having a pitch “played by one instrument before a cadence and then have[ing] it seemingly magically taken over by another one immediately afterwards”, thereby creating an exchange of timbre. However, a modern composer using electronic processes has an infinitely larger set of transformative tools at his disposal.

In the composition of VOX 5, Wishart favoured a method of ‘resynthesis’ in which the Phase Vocoder was used to analyse the spectral characteristics of inputted sound data. Once this had been completed, the composer generated entirely new timbral morphologies which enabled him to link his original sound-objects. The main processes used to achieve these transformations were those of spectral manipulation (what Wishart also calls spectral ‘stretching’) and interpolation. Spectral manipulation was used most prominently to create the bell-like sounds that predominate in the piece between 2’39” and 3’03”. The concept of these ‘vocal bells’ was a particularly problematic one for the composer to realise: the spectrum of a bell sound is, by its nature, inharmonic, and as such it represents to Wishart “one of the few sounds the voice cannot imitate”. The ‘specsh’ program that Wishart had already written could only perform linear stretches on input spectra, preserving the harmonic relationships inherent in the original sound-object. Therefore, a new program had to be created - ‘spece’ - which “allowed me to vary the type of interpolation used in the stretching process” and hence create inharmonic outputs that were then time stretched, after their original vocal attack, in order to create the sound-events heard in VOX 5, where clearly vocal attacks birth bell-like timbres.

For the most part, however, Wishart was able to use the mimicking ability of the voice, discussed above, and its onomatopoeic nature, in order to create the sounds he wanted: morphologies that could more easily be electronically integrated with other concrete sounds through spectral interpolation. In his Audible Design, Wishart is careful to point out that a successful interpolation should always leave us with the feeling that “we have moved away from one type of sound and arrived at a different type”. In other words, we should clearly recognise both source and goal. However, these two states are, crucially, all we should recognise. To take an example from VOX 5, in the transformation between ‘voice’ and ‘bees’ which begins at 2’03”. Wishart aimed to create a seamless transformation “without any intervening artefacts which might suggest some other physicality / causality, or even betray the technical process involved”. After all, the goal is not simply to create mathematically coherent transformations, but acoustically coherent ones. As Wishart points out, simply mixing two sounds together does not work: our brain is very good at unscrambling the multitude of sounds that our ears will perceive at any one time. The main problems that Wishart identifies are those of “onset synchronisity... and... the parallelism of microfluctuations of the components in any one source, at the same time being different from that in other sources”.

To create a single synchronised start-time for a sonic gesture is not too demanding a problem when one is able to use a computer workstation. The first of the ‘vocal bell’ sounds in VOX 5, for example, is a fusion of three vocal syllables, the attacks of which have been synchronised, Wishart tells us, exactly to the sample. However, to interpolate between two sources - the spectra of which are changing in time - such that their microfluctuations are perceived to be identical, is a far harder challenge. As Wishart puts it, the goal is not simply to interpolate between two sound-objects. Rather, “we are effectively interpolating the microfluctuations themselves”: in other words, instead of having our spectrum move gradually away from the start value of sound 1, and towards the end value of sound 2, many small interpolations have to be calculated, such that at any given time we are moving from the current value of sound 1 towards the current value of sound 2. This concept can perhaps be better explained diagrammatically: in the image overleaf, two sounds are split up into a number of what Wishart calls ‘frequency domain analysis windows’. Because the waveforms we are dealing with are so complex, we must take each window individually, “interpolating over the differences between the values in successive windows” and “moving from the “wobbling” of one spectrum to the “wobbling” of the other”.

Furthermore, a certain amount of preprocessing has taken place on a number of the source sounds of VOX 5. For example, a technique known as ‘blurring’, ‘flocking’ or ‘wedging’ was used as a type of “spectral thickening” in order to reduce the spectral detail of the complex vocal sounds and, hence, make them more easily transformable. Basically, several slightly modified versions of the same sound were layered on top of each other in such a way that the layers could move independently of each other. This technique not only aided the mathematical process of interpolation (by creating ‘simpler’ sounds with which to work), but the moveable nature of the layers meant that the listener’s attention could be intentionally distracted during certain moments in the transformative process. Another problem that Wishart has noted in sonic transformation is that of the ‘perceptual switch’: the fact that “when perceiving a continuously changing object that we originally recognize, we continue to interpret it as the original object until a certain threshold is reached”, at which time we suddenly perceive it as something other. As the intent was the creation of seamless sound-objects, Wishart devised a method by which one of the layers of a ‘wedged’ sound could be used to capture the attention of the listener “at the moment of maximum ambiguity” by the simple expedient of having it change independently of the transformation currently occurring. Landy calls the aforementioned ‘voice to bees’ transformation the “classical example” of this pioneering method: here, the independent shifting of the vocal buzz’s upper spectrum is what distracts us from the ongoing interpolation between the two sound-sources.

The streaming fluctuations and sonic metamorphoses created through these techniques are what gives VOX 5 its particular character - which is not to say that Wishart’s methods of spectral transformation make for results that are in any way repetitive, or that suffer from diminishing returns over the course of the piece. The composer’s mastery of extended vocal technique makes for a diverse range of sound-objects which, nevertheless, all clearly belong in the same sonic landscape: The object ‘voice to bees’, which gradually unfolds over a number of seconds (a slight crescendo giving life to the transformative act), is very different in effect from the abrupt ‘vocal whinny’ that occurs at 3’43”, its suddenness appearing as an uncontrollable creative explosion. However, both are recognisable, in the narrative context of VOX 5, as the creative outpourings of a godly ‘supervoice’, and for this reason the piece can be said to have a landscape that is incredibly well-realised, being both structurally and narratively sound.

After all, in Hindu mythology Shiva is seen as both creator and destroyer: on the one hand, his dance of death destroys the world and brings all things to an end. However, at the same time this destruction can be seen as an act of renewal, preparing the ground such that a new world can be called into existence. Shiva, then, has a split personality, both sides of which are transformative in nature, and Wishart clearly reflects this in his choice of transformative technique: in this respect, the ‘how’ and ‘why’ of VOX 5 are clearly mutually informative. This insight, although elementary, provides us with a fuller context for the transformative processes Wishart adopted for VOX 5, and this reinforces Landy’s assertion that the ‘why’ of a piece is every bit as important as the ‘what’ and the ‘how’. However, discussion of VOX 5’s narrative context is, I believe, the one area that remains severely underrepresented in a vast majority of that which has been written about the work. There has been a great deal of talk, some of it evidenced above, concerning the ‘why’ of VOX 5 inasmuch as the piece represents Wishart’s desire to create a digital piece building on his analogue work experimenting with timbre, moving away from the proscriptions of the ‘lattice’, and utilising pioneering transformative techniques. However, little has been made of the ‘narrative why’ of the work, a grasp of which is surely every bit as important to a complete understanding of VOX 5 and, of course, the VOX cycle as whole. As such, in the paragraphs that follow I have attempted my own narrative reading of VOX 5. This response to the piece is, of course, entirely subjective; however, I make no apologies for this, and feel that an exploration of this kind is a necessary one.

Wishart has stated that he found the “anarchist element” of the Shiva myth particularly attractive. After the destructive climax of VOX 4, we experience the world through Shiva, and sense its being in a state of flux. This sense of a collapsing, or even collapsed, world can be felt in the wide variety of images, both natural and man-made, which pass through Wishart’s sonic landscape. The first sound we hear, for example, is the gradual approach of a flock of crows, their ‘cawing’ placed inside an atmosphere of empty wind noise and bubbling undercurrents which conspire to create an ominously post-apocalyptic image. This multitude of birds, however, is subsumed into the first vocal entrance at 1’05”: it is as if Shiva is taking a great breath in, swallowing these final signifiers of a devastated world before spitting out the sounds and images that constitute the rest of the piece. In the final minute of VOX 4 we are left with a society that has disintegrated (indeed, to my ears the manic laughter that begins at 9’49” in that piece can be seen to mirror the crows here, particularly if they are both taken to signify the aftermath of society’s breakdown): now, it seems, the world itself is no more, its memory existing only as aspects of Shiva’s character which manifest themselves in the god’s superhuman vocal utterances. VOX 5 is very much a piece which embodies Wishart’s interest in transformation as “change, modulating and continually active”: what better image is there to illustrate this concept than that of a god who begins a piece by subsuming the last chaotic remnants of creation into his being?

If, indeed, Shiva has completed his demolition of the world by swallowing it - an apt metaphor for the process of ultimate godly destruction - then the passage from 1’05” to 2’03” could almost be taken as a journey down the god’s oesophagus and into his stomach, where all manner of recently-swallowed creatures, both human and inhuman, float and flounder in his gastric juices. Between 2’00” and 2’03” these objects appear to be corked, as if we are shooting back out again, through some other pipe, leaving the body through Shiva’s trachea just as the sound of an immense swarm of bees is being vocalised. The bees, once clear of the god, seem to fly away, and by 2’25” we are left once again with the vocal sound that created them. It is almost as if Shiva, having swallowed the world in one go, is struggling to keep it down, resulting in many tiny, impromptu, acts of creation. The god’s stomach is simply churning with far too many things whose primary function is to exist outside of him, and as such he is becoming somewhat bilious. This would be in keeping, of course, with the Hindu belief that the world will be created anew after its destruction: even Shiva will not be able to indefinitely contain an entire world of being that wants so badly to be brought back into existence.

For example, the vocal ‘stabs’ that begin at 2’53” (‘Vuh!’, ‘Too!’, ‘Zee!’ etcetera) could be construed as an aspect of Shiva’s own speaking voice. However, they become overtaken for a time by another multitude of goblin-like squabbling voices, which try to escape from the original vocalisations before being caught and confined by the ‘t-t-t-t-t-t-t-t-t-t-t-t’ sound which causes this section to crumple in on itself by 3’08”. A wrathful section immediately follows, with ‘electric shock’ sounds (from 3’11”) and a whirling ‘szszszsz’ noise chasing a high-pitched ululation that grows in in pitch and amplitude before involuntarily releasing a slow, watery and low-pitched yawn of creation at 3’26”, from which will ultimately spring the aforementioned ‘vocal whinny’ at 3’43”, as well as mock baby’s cry that morphs into a vocalised engine noise at 3’45” and following. Even more evocative of this idea of Shiva desperately trying to contain that which he has created is the section between 3’53” and 3’58”, where two spinning voices break away from a multitudinous crowd, whirl around the speakers and are shot down explosively, both ending in a stifled ‘oof’. The image here has become Shiva as riot-policeman, resorting to ever-more violent methods in an attempt to keep the peace.

By the time VOX 5 ends, this peaceful state appears to have been reached. Shiva brings into being a heavy storm, beginning with a great thunderclap at 4’07”, and it is this that appears to finally settle his stomach long enough for him to enact the final celebratory dance that VOX 6 represents. In an image which, for me, recalls the biblical metaphor of Christ’s death, the final voice disappears, whimpering, under a thunderclap at 5’08”, and for the last full minute of the work we are left with the thunderstorm and rain playing us out, gradually fading away into nothingness.

It is unlikely that the above few paragraphs have fully managed to describe the ‘true’ narrative of VOX 5 - i.e., that running through Wishart’s own mind during the composition of the piece. Indeed, the vast majority of my response to the work may well be entirely flawed in this respect. The important point, however, is that VOX 5 is a work that invites such interpretation. Simon Emmerson has suggested that “perhaps all artists are in the end practitioners of trompe l’oeille”, since “art is about suggestion (even, on occasion, deception)”. It is, perhaps, one of the defining characteristics of great art that it invites opinion, discussion and interpretation in and of itself, quite apart from the techniques used to create it. If we neglect to discuss the artistic ‘why’ then, as Landy points out, we are doing a great disservice to the sonic artistry involved in the creation of a piece such as VOX 5, as well as providing a potential barrier to listeners who do not mind how a piece of music was made; indeed, whose path to meaning might, at first, be confused by the method of presentation inherent in the electroacoustic medium itself.

Hopefully, then, this essay will have gone some way to addressing all three of Landy’s questions: indeed, it is testament to the artistry inherent in VOX 5 that both its ‘how’ and its ‘why’ are equally strong. It is a groundbreaking work in a technical sense - Miller Puckette has written that VOX 5 was the piece that finally convinced him that “the phase vocoder can give rise to musical results which would be hard or impossible to come by otherwise”. However, it is, perhaps, equally as impressive when viewed on no other terms than its being an artistic retelling of the Shiva myth. Its metaphors are both striking and, despite clearly belonging in their own particular landscape, overseen by a godly ‘supervoice’, open to all manner of interpretation. Any one person’s reading of the narrative of the piece will clearly be their own, and not anybody else’s. However, assuming a cursory understanding of the anarchic and double-sided character of Shiva, all such readings will, in the end, reflect an aspect of the reality of VOX 5, and Wishart’s accomplishment in the creation of this reality is to be as equally valued as his creation of the techniques that aided him in this work.


T Wishart, Audible Design: A Plain and Easy Introduction to Practical Sound Composition (1994)

T Wishart, On Sonic Art (rev ed 1996)

The Concise Encyclopedia of Living Faiths (ed RC Zaehner 1977)

E Brattico & F Sassanelli, ‘Perception and Musical Preferences in Wishart’s Work’, in Journal of New Music Research Vol 29 No 2 (2000), pp107-119

S Connor, ‘The Decomposing Voice of Postmodern Music’, in New Literary History Vol 32 No 3 (2001), pp467-483

S Emmerson, ‘‘Live’ Versus ‘Real Time’’, in Contemporary Music Review Vol 10 Issue 2 (1994), pp95-101

L Landy, ‘Sound Transformations in Electroacoustic Music’, part of the Composers’ Desktop Project (1993)

L Landy, ‘The “Something to Hold On To Factor” in Timbral Composition’, in Contemporary Music Review Vol 10 Part 2 (1994), pp49-60

M Puckette, ‘Something Digital’, in Computer Music Journal Vol 15 No 4 (1991), pp65-69

T Wishart, ‘The Composition of VOX 5’, in Computer Music Journal Vol 12 No 4 (1988), pp21-27

T Wishart, ‘From Architecture to Chemistry’, in Journal of New Musical Research Vol 22 No 4 (1993), pp301-316

D Witts, ‘Trevor Wishart and ‘Vox’’, in The Musical Times Vol 129 No 1747 (1988), pp452-454


  1. Good grief this lad can string words together!

  2. im going to listen Wishart's stereo mix, i think it is really good, when i read this blog it catch my attention

  3. Wishart is the master. No doubt. Although I hear that online viagra is also pretty swish?