Table of Contents

ABSTRACT-SHAPING— With this term I wish to designate operations in the studio that alter the natural and referential* character of a sound, extracting from it 'interior'-sounding criteria (see interiority*), presenting a musical potential more abstract than its original reference*. Transformations of any order, like filtering*, enveloping* or pitch-shifting* alter sounds by partially or totally removing traces of their original causes.

ACOUSMATIC— expression derived from 'acousmata', a term employed to designate Pythagoras's disciples who used to listen to his speeches through a curtain hiding his visual presence. It is now employed to designate the situation in which listeners to electroacoustic music can no longer identify the source of a sound. The performance of electroacoustic music is optimized when pieces are played in an 'acousmatic' situation: when nothing relevant is exposed to the eyes, leaving the ears the complete freedom of personal imagination. Listeners in an acousmatic concert are usually also remunerated through a sounding environment composed of a comprehensive range of loudspeakers. (See Musique Acousmatique*).

ACTUALIZATION— Designates the mental attitude of composers towards the attribution of places and functions of sounds and events in a composition. When a composer decides on what a sound or event 'is' in the piece, this sound or event passes from a potential to an actual state. See potentialization* and abstract-shaping*.

CDP— Composer's Desktop Project. Name given to the workstation designed by the group of the same name. Although somewhat demanding in time, CDP's applications for electroacoustic music composition are various and very powerful. (Cf. Appendix: Technical Considerations).

COMB-FILTER(ING)— To pass a sound through a group of filters, generally distributed equidistantly. This mode of sound processing changes the spectral content of sounds around the selected frequencies, possibly boosting them to the extent of generating resonance*, when the width of the filtered areas become very thin. Sinusoidal frequencies then start to oscillate in response to the feedback created in those frequencies.

CROSSFADE— To fade in one or more channels of sound while fading out others in order to cross between the two.

DSP— stands for Digital Signal Processing, referring to all modes of digital processes that can be applied to soundfiles*, samples*, or recorded sounds. The aim is usually to transform the sound in some way.

DYNAMIC PROCESSES— See dynamo*.

DYNAMO— Expression derived from the idea of 'dynamic process', a term used by composers of electroacoustic music to designate aspects of a piece, or part of it, which follow a dynamic structuring process. This can either result from a sound's reference acquiring a musical function, or from the effect of the settings of DSP* devices which impose some sort of energetic evolution on sounds. (Nowadays MIDI devices too are becoming involved in the fabrication of similar processes — 'acoustic models' in GRM's terminology in the manual for 'MacSoutiLs'.) I prefer the expression 'dynamo' to 'structure' insofar as the latter is not always 'dynamic', and is compromised by a complementary notion of 'object'. In the case of VOLTA REDONDA the dynamo is a spiral (cf. chapter 4). (A spiral could be described also as a structure, even as a graphic event, but it is the distribution of the energy orientating the development of the piece in this shape that really matters).

I would avoid the expression 'dynamic process' because a dynamo can be less complex than a process: the idea of process may suggest a nexus of causes and effects, whereas a dynamo can imply a single and simple energetic function.

This notion, especially when the dynamo is 'natural' (cf. chapter 2), may invite composers to believe in a 'deeper-naturalistic' music based on more objective instances than, for example, more 'cultural' 'écriture'-based compositions. In my opinion if such a link is established — which would imply in finding true sense and full meanings for musics — it deserves to be inhibited by music's ambiguity.

(Regarding the mutual confrontation between 'natural' references and the 'cultural' or learned side of music, I favour a more phenomenological approach, avoiding the intricacies of this metaphysical labyrinth. The question as to whether the experience of natural events shapes musical perception or whether musical disposition shapes the perception of nature leads to a whole branch of philosophic questioning that, if not dated, cannot be resolved in this commentary).

ÉCOUTE RÉDUITE— Expression invented by Pierre Schaeffer to designate the more abstract mode of listening (Schaeffer, 1966), where the source is disregarded as attention is directed towards an appreciation of the inner qualities of sounds. According to Michel Chion:

ÉCRITURE— French term widely used in the French contemporary music scene that not only refers to composition as an act of writing. It has also come to work as an alibi for the composer's correct decisions and determinations. The term is sometimes used to imply criticism of electroacoustic music's supposed 'indeterminacy'. Pierre Boulez, or perhaps his translator into English, forewarns us that:

Pages later, composer Antoine Bonnet clarifies:

One needs to try only once to compose in the studio without employing 'symbolic reasoning' to understand that this is not really a problem of the poverty of the English language or of the electroacoustic studio.

ENVELOPE— Generally speaking, an envelope is a graphic or mental representation of the changes in the dynamic (or pitch) evolution of a sound. Although belonging to technical jargon, it is almost identical to Pierre Schaeffer's 'profil dynamique' and 'profil mélodique' (1966), two of the perceptual 'critères de forme' (as distinct from 'critères de matière') of sound objects. Envelope differs from a 'profile' in that while the former is a concept of technique, the latter relates to perception and can only be identified as a profile as long as a necessary duration allows its perception as such. A too fast or too slowly-evolving envelope may not present a profile, but rather a brief pulse or an endless sustain.

The term 'envelope' can be found in synthesisers and other devices, being also applied for modulations of volume, filter, pitch and other parameters.

FFT (or Fast Fourier Transform)— is a tool used to represent graphically the spectral constituents of a sound. The sound is mathematically analyzed and is reduced to its various overtone components in their dynamic evolution over time (see envelope*). The graphic representation of an FFT usually plots the amplitudes of the frequency bands over time, offering a highly readable and visual representation of the sound being examined.

FILTER— see comb—filtering.

FORMANT-CHORD— Expression I use to designate a functional abstract-shaping* employed in VOLTA REDONDA. It is something between a chord and a formant*, acting like the latter in the sense that it creates a spectral coherence throughout the pitch range. Its presence throughout the piece relates functionally to a chord. In VOLTA REDONDA the formant-chord consists of a set of boosted frequencies in different sounds enabling a sort of canvas to which other functional and dynamic items relate. (Cf. chapter 4).

FORMANT— 'A peak of frequencies in the spectrum of a sound which may be associated with a resonance in the mechanism of the production of the sound. Vowel sounds may possess more than one formant in different parts of the spectrum. The term may also be used in relation to musical instruments.' (Borwick, 'Sound Recording Practice', 1987, p.537).

GRAIN— One of Schaeffer's three criteria of matter, the other two being 'masse' and 'timbre harmonique', grain describes the 'grating' perceptual qualities of sounds that are not smooth. It designates a granular quality. Schaeffer's 'grain' should not be confused with Barry Truax's technical concept of 'granular synthesis' (cf. chapter 3). The latter refers to a small portion of sound that can be combined with others to create larger sounds or events in a compostition.

INTERIORITY— The contrary of reference*, 'interiority' designates the qualities of a sound that do not refer to external causes/sources. It is close to what one hears during what Pierre Schaeffer called 'écoute réduite'*. When attention is drawn away from sound's exteriority, or reference, the 'reduced' experience can be at least as full and complicated in its own way, including the perception of mass, grain*, etc. — qualities of the 'interior' of sounds.

ITEM— Whenever the word 'item' is related to a musical context, it means either one single sound or a single group (of sounds) behaving in a collective and functional way. (One exception exists in CANONS/CHAOS where some 'items' are too small to be perceived as sounds.) For instance, an impact is a 'sound' and an avalanche is an event. My fingers drumming on the table can make either an iterative* event or a set of five brief sounds. (It depends only on my attention and on the speed of my fingers).

For the composition of these pieces, most of the time an item is, technically speaking, an isolated unit recorded either as a sample* or as a soundfile*.

ITERATION— ITERATIVE— While classifying the modes of 'execution' ('facture') as criteria of sounds, Pierre Schaeffer designates 'iterative' execution:

LOOP— A loop consists of continuously repeated playback of sounds. In the analogue domain of tape recorders a loop consists of pieces of tape whose end is joined to its beginning. Looping using a digital sampler* offers different possibilities, especially if stereo looping is possible: if each channel has a slightly different size, phasing effects affecting the spatial or the temporal dimensions can be explored. Samplers offering more than one loop per sound enable the contruction of increasing or decreasing loops (cf. chapter 3). Computer programs like SoundDesigner™ allow playlist looping, which enables the progressive increase or decrease in the durations of each of the repetitions (cf. chapter 4).

MIDI VOLUME— Parameter that controls the overall volume of a midi channel. It enables several operations, the most common being the mixing of sequenced* channels.

MIDI— Musical Instrument Digital Interface designates the standard protocol for communication between electronic instruments and equipment with MIDI implementation. 'At its simplest level, it means that playing a note on one MIDI synthesizer will send out three eight-bit bytes from the 'MIDIout' socket on the back of the machine, which may be routed to the 'MIDIin' socket on another machine and so tell it to play the same note at the same velocity, that is, hit with the same force.' (Jonathan Gibbs in Borwick, 'Sound Recording Practice', 1987, p.348).

MIKE-SHAPING — I agree with Michel Chion that the action of recording new ITEMS* for use in a composition of electroacoustic music is not as sterile as the expression 'sound-recording' would imply. He proposes:

This recording action depends on and determines what will happen in any future application of its results in the composition. It may demand a very complicated configuration of microphones, or a simple one perhaps with movement of microphones during the recording; it may vary from picking up embrionic sounds that will fit in a piece, to the recording of fully completed parts of it, which establishes an action of authorial intention. Through lack of a proper term, I would like to split this notion in two, proposing to the first an inelegant neologism — 'mike-shaping' (for Chion's 'tournage sonore' as far as only the use of microphones is concerned). The second idea proposes that behind the authorial action with a microphone there is an attitude common to actions performed with other media, which should be acknowledged as well. (See potentialization*). With the expression 'mike-shaping' I am trying to call attention to the specific role of the microphone as an active tool for the shaping of sounds during their first potentialization.

I believe that the employment of the word 'sonore' in 'tournage sonore' implies a dividing line between what is still sound and what would be music. As previously described (cf. chapter 1), such a line is unclear at all stages of composition; it presupposes that during the process of sound collection one cannot extract what might already be a whole and actual* section of a piece. Although 'electroacousticians' may prefer to perform abstract-shaping* on the sounds, it cannot be denied that sounds in which there are no transforming interferences might be used.

MIXING — To combine sounds from different sources (instruments, sequences, sections, textures, etc..), recorded on separate tracks into a single (and usually stereo) recording.

MUSIQUE ACOUSMATIQUE — 'La Musique Acousmatique ou l'Art des Sons Projetés', note in the Encyclopaedia Universalis, 1984, signed by François Bayle. It is the name given by Bayle to electroacoustic musics which are not conditioned by visual cues (see acousmatic*), and to electroacoustic music diffused in concert over loudspeakers thereby hiding the origins and causes of sounds. In some francophone countries it has become a widely accepted label for designating electroacoustic music.

MUSIQUE ANECDOTIQUE — Name given to the genre inside electroacoustic music that employs recognizable sounds more for their 'anecdotic' or narrative aspect than for their abstract potential. The first piece to be named thus ("Hétérozygote" by Luc Ferrari) was not the first anecdotic work. It was preceded by many of Schaeffer's compositions and the music of other composers.

MUSIQUE CONCRÈTE —

OBJET SONORE —Concept created by Pierre Schaeffer in the fifties to designate any sound unit perceived in 'écoute réduite'*. It is described by means of perceptual criteria which are independent of the sound's meaning and cause. 'Objet musical' is a sound unit detected and considered inside and according to its function in a musical context.

OBJETS INCONVENABLES — In Schaeffer's 'Typologie' (1966, 1967), troublesome objects are those which either present too little ('objets redondants') or too much information for the ears ('objets trop eccentriques'). As a definition of too-eccentric objects:

OVERTONE — A component of a complex wave which may or may not be an integral multiple of the fundamental frequency. When overtones are all integral multiples, the sound produced is clearly pitched and easier to blend harmonically with others. In that case, overtones are called harmonics.

PAN(ORAMICS) — Control that enables the placement or movement of sounds in the stereo field. MIDI* control offers 128 positions to be chosen between extreme left and right.

PHASE-SHIFTING — The effect produced by creating (and sometimes increasing), during playback, a gap of time between two copies of one recorded item. When the difference is small the effect is close to that of a sweeping filter*. Augmenting the gap will lead to panoramic* motions and further to 'echoes'.

PITCH-BEND — A portamento-like glide in pitch. In MIDI* pitch-bends are created by means of a dedicated wheel or by means of stored information (e.g.: in a sequencer) being sent to the pitch-bend control.

PITCH-SHIFTING — The application of a function to alter the pitches of sounds, most commonly changing their duration as well. Pitch-shifting without this side-effect is related to TIME-STRETCHING*. In some machines this function can be controlled by oscillators, by hand-drawing or even by dynamic envelopes extracted from the sound itself or from other sounds. The expression designates only the effect on pitched sounds, but such a process can also be applied to non-periodic, non-pitched sounds or whole events.

PLAYLIST — Facility offered by some hard-disk recording systems that enable the playback of selected parts of a soundfile in a pre-defined order (or list).

POTENTIALIZATION — (of sound objects, characters or events). I employ this quasi-Aristotelian term to describe a composer's relationship towards items selected to be incorporated into a composition. No matter whether by sampling*, synthesis, recording, mike-shaping*, etc. 'potentialization' concerns the attribution of musical potential and power to sounding items. This definition does not rule out utilizing sources from any media — eg. copying from tapes or CDs, resampling from existing sampling libraries, etc. This concept has affinities with Michel Chion's 'sons fixés' (cf. footnote 15 at the Introduction), and 'tournage sonore' (cf. MIKE-SHAPING*) but stresses not the medium but the compositional attitude.

4X— Real-time digital signal processor developed at IRCAM by Giuseppe di Giugno to enable musics demanding sound generation and treatment in real-time, i.e.: during performances.

REFERENCE — Property of recorded sounds which expose, suggest or at least do not hide the source to which they belong. A vector pointing in the opposite direction to interiority* (See also'écoute réduite'*), it refers to the 'non-abstractness' of sounds, i.e.: to their context away from intrinsic criteria of perception. (Somewhat arbitrarily, and in order to help my descriptions, I do not include synthetic sounds in the category of referential sounds even though they can refer to their synthetic source, nor recorded instrumental sounds that refer to the instrument, nor sounds that remind us of their electroacoustic manipulations.) By referential sounds I should like to designate only those which point towards a more macroscopic setting, referring to natural scenes and phenomena, human or mechanical activity. It is not their real origin that matters, but their power of evoking extrinsic settings. A referential sound can even be of synthetic origin: e.g.: a filter sweep in a band of white noise could refer to 'wind'.

RESONANCE — Resonance happens when one or more frequencies of a sound are increased in amplitude (volume) by means of a boosting filter* so as to trigger a feedback of the selected frequency(ies). The resulting effect will be perceived as a sort of sustained pitch, or pitches shadowing the original sound. See also comb-filtering*.

SAMPLE — (SAMPLER)— (SAMPLING)— A sample is a portion of sound locked in digital memory, which can be played back, transposed and undergo several other tranformations which are possible in a dedicated device called a sampler. Most of these functions can be triggered either by the sampler's controls, by a MIDI instrument or through other digital interfaces: computers, MIDI data boxes. (See soundfile*).

SEQUENCING — The recording of MIDI information in a dedicated machine (software or hardware) that can play it back as a computer file. This operation can be either performed in real-time, by playing notes on a MIDI equipped instrument, or entered step by step through the computer's interfaces. Sequences thus recorded are called tracks and can be mixed-down, reversed, time-scaled*, merged etc.

SEUILS TEMPORELS DE L'OREILLE, LES— I include this paragraph as a further development of two entries in chapters 1 and 2 which comment on the lack of correspondence between acoustics and perception, a theme extensively explored by Pierre Schaeffer. This is his third theme of reflection in the 'Solfège de l'Objet Sonore'*:

SOUNDFILE — A sound as a computer file, available for playback and all other DSP* treatments. Technically speaking it is the same as a sample*, but it is called a soundfile in programs that will not play it back in the same way as a sampler (with pitch transpositions and MIDI controls).

SOLFÈGE DE L'OBJET SONORE — Follow-up to Pierre Schaeffer's most important book (Pierre Schaeffer, 'Traité de l'Objet Sonore', Ed. du Seuil, Paris, 1966), in which, illustrated with sound examples, he studied the relations between acoustics and perception and proposed a new solfège of all sounds by means of his 'Typo-Morphology'. Presented in a trilingual edition (French, English and German), it is the best introduction to his thought.

STRONG ACTION — The strong action of devices for sound transformation and abstract-shaping occurs when these devices impose too much of their own characteristic effects, sometimes fooling composers into believing that the result obtained is due to characteristics belonging to the input sound or to his imagination (cf. chapters 2 and 4). The same can be said when an exaggerated use of devices marks sounds or events with a presence that overwhelms the sounds (and events) themselves.

Of course it is difficult to trace a dividing line separating musical production in two parts, one made up of 'strong actions' and the other not, because the use of devices is in itself an inseparable part of the electroacoustic work. But, more embarrassing than the corpses of machines acknowledged by Boulez (cf. chapter 1), is the amount of electroacoustic music made with them, particularly considering the novelty of their aesthetic impact when they first appeared compared with their rapid obsolescence and descent into cliché.

Even though acknowledging the difficulty of the terrain, I believe that there is something political in the strong action of devices. Strong action is not only responsible for imposing on musics the language of the device, but also, by supporting a musicality based on the device's power, it promotes the music made by the device's first user (or buyer, or commissioner), thereby transferring the values to a more political scale.

As far as electroacoustic music is concerned, aesthetics and industrial power are closely related. (Perhaps it may not be from the richest power that my favourite music comes, but certainly electroacoustic music could not be invented and intensely cultivated outside the nexus concentrated in the more developed continents). Yet it is unreasonable to nurture antipathy toward electroacoustic music only because the world's balance of power is not even. However, as a composer and listener one should adopt a certain awareness of misuses of power in the musical domain, one example of which I discern in 'strong action'. Certainly I would not pretend to dispose of the problem so rapidly by clearly isolating 'use' from 'abuse', so much so because there are many exceptions which almost deny the problem's existence. Even more so if we consider how difficult it is to conceive of a 'neutral' device. The problem is that of knowing if there really is a problem, especially when a so-called strong action happens to be responsible for musically interesting results. Why should I consider these 'bad' or 'abusive' if they are genuinely musical? My difficulty worsens when, shielded by the novelty of the device, the strong action is not yet classifiable as such. Only after collecting examples of the musical repertoire will one be able to detect a pattern of strong actions. Then the existence of the problem is confirmed.

It is curious to notice that 'strong actions' are picked up at a later date by composers outside the developed nexus. Due to the peculiar mechanics merging precocious aesthetic ageing with technological obsolescence, it is easy for us to spot this delayed occurrence of 'strong action'.

TIME-SCALING — In terms of MIDI controls, it means the alterations of the duration of any group of MIDI data in the time domain. A whole sequence can be accelerated or delayed by means of tempo adjustments on a 'conductor track' that imposes its law on all data uniformly. But when certain channels must be independently time-adjusted, or when information concerning only one type of data (like pitch-bend, panoramics or MIDI-volume) must be altered, then a device called 'timescale' can be called to perform this without affecting information otherwise not selected.

TIME-STRETCHING — DSP* operation initially designed to stretch soundfiles* in their duration, ideally without affecting their pitch (or pitches). This is the same device that allows pitch transpositions without changes in the soundfile's duration (see pitch-shifting*). When applied exaggeratedly to complex events, time-stretching abandons its initial goal of changing duration to that of creating less predictable events recalling Pierre Schaeffer's experience in 'Les Seuils Temporels de l'Oreille'*. This relates to what I call 'non-linear response' of elecroacoustic equipment (cf. chapter 1).

Next