The researcher who wants to “work at his instrument” (creating sounds) and his ear at the same time will quite naturally have to go back to his instrumental and musicianship classes. This now consists in making, not scales, but what has taken their place: sound reels, in the usual language of the studio, and dictation—that is, descriptions (and not notation) of sounds.
This is the experimental approach, very different from the compositional approach, whether concrete or electronic. The “concrete” musician, in too much of a hurry to compose, goes straight to editing. The electronic composer, overequipped, sits down at his keyboard of frequencies or formants, or constructs a blueprint with parameters. We would yet again advise both of them not to jump straight from ignorance to inspiration, from archaic concepts to modern techniques.
What is this sound-homework like? We must be careful to distinguish between the two possible pathways: either translation, a sort of dictation where we are taught to describe the sounds on a reel given to us, or prose composition, where we have to create sounds starting from the schema of their structure. By way of example, we will take a reel and go down the first pathway.
“Translation” consists in listening to a given object in order to describe its specific contents and organization, its internal morphology, as explicitly as possible. Systematic training in translation will be aided by the creation of translation reels, containing a series of sounds with as wide a range as possible of origins and factures. Generally speaking, there is a number of sources to make these sounds:
• natural sources: trombone, flute, trumpet, piano, vibraphone, marimba, suspended cymbal, gong, sheet metal with bow or bass drumstick, for example;
• electronic sources: a sine-wave generator; a filter for accentuating or lessening the “timbre” of the pure sounds delivered by the aforementioned generator; a white-noise generator with filter for “cutting out” slices of sound of various thicknesses.
To create our objects we can stimulate these sound bodies in different ways: use the bow or the bass drumstick for the sheet metal, manipulate the electronic sources in various ways (any frequency for the sine-wave generator, more or less provided with “timbre” by the filter).
We can also position microphones at various distances from the sound sources, in different places and in varying numbers; finally, as we make the recording we can modify the “forms” of the sounds with the potentiometer, filter them, use the echo chamber, and so forth.
Thus we will be making our objects by playing an “instrument” that on each occasion will be the ensemble of sound source and means of recording. So we can make the following sounds, for example:
1. A tonic sound formed with the sinusoidal sound generator, “given timbre” by the filter amplitude, with its form provided by the potentiometer;
2. A tonic sound with vibrato formed with the same generator, with a frequency modulation to create the vibrato, and its form, as above, provided by the potentiometer;
3. A complex sound with a profile pianissimo—forte—pianissimo obtained from the filtered white-noise generator (only 1/3 of an octave remains), its profile also made with the potentiometer;
4. A complex impulse with resonance (an impulse of white noise plus echo chamber);
5. A flute sound with progressive attack, normal duration, then decay;
6. A trumpet sound with a sforzando at the beginning;
7. A trumpet sound with an ascending gruppetto right at the end;
8. A piano sound recorded on one microphone, imperceptibly merging (fade-in-fade-out) into a flute sound of the same pitch from a second microphone (making the output levels from the two microphones vary in reverse order at the time of recording);
9. A trombone sound with the following profile: pp < ff > pp < ff;
10. A gong sound, damped in the middle of the resonance;
11. A cymbal sound, damped just after the attack;
12. A thud from a wooden bass drumstick on the sounding board of a piano, with the resonance;
13. A large W note on a metal sheet recorded on two microphones, the first recording a complex bass resonance, the second locating a high tonic partial (the intensity of the sound is kept constant by raising the level on the potentiometer for about half the duration of the resonance);
14. A very complex channeled X sound (beating a metal sheet, a piano, gong, vibraphone, marimba).
We will now try to ignore how these sounds are made, which we explained to the reader in order to illustrate our point but which must of course be unknown to the listener trying his hand at musical dictation, since he is practicing acousmatic listening to sound for its own sake, reduced listening.
So this listener will notice
• balanced notes in which the three temporal phases appear clearly: attack, continuation, decay: flute sound (no. 5) and electronic sound (no. 1);
• balanced notes with conspicuous attack: sforzando attack: trumpet (no. 6); piano percussion-attack in the piano-flute fade-in-fade-out (no. 8);
• balanced notes with conspicuous decay: trumpet with ascending gruppetto at the end (no. 7), gong damped near the middle (no. 10);
• finally, balanced notes with a conspicuous continuation: the harmonic development of the metal sheet where we witness a slow transformation in the color of the sound (no. 13); and the electronic sound vibrato (sound no. 2).
It is nevertheless quite rare to find balanced notes—that is, notes where the three temporal phases (attack, continuation, decay) are clearly perceptible; most of the time two of these phases or even all three are merged into one. We will call these deponent notes.
• Either we find an attack followed by a decay (the usual case in percussion-resonances): in the very complex X (no. 14), the thud on the piano sounding board (no. 12) or the white-noise impulse with artificial resonance from the echo chamber (no. 4);
• Or the temporal phases are not distinct from each other and we go imperceptibly from the attack to the continuation, then to the decay, and it becomes difficult to demarcate these different phases: electronic sound (no. 3) and trombone sound (no. 9);
• Or else at the limit we find only an attack, so short is the sound: cymbal damped immediately after the attack (no. 11), and the electronic impulse no. 4 without echo chamber.
Reading our summary we notice that in truth, even if typology and morphology give some means of describing and notating and a certain sense of preparedness for experimentation, the description of sound, as far as dictation is concerned, is still short of terminology and a logical method of analysis. To carry out accurate dictation, we will have to return to this mode of listening once we are in possession of the ideas in book 6.
Many sounds nevertheless seem to require an analysis of their external morphology first. What, in effect, stands out most of all when we listen to them is that they are made up of distinct elements, with forms that are separate from each other. This occurs very frequently. Moreover, some sounds have “impurities,” elements that are not heard as an integral part of the object. We will make a distinction between these two typical examples of observable external morphology, which could be compared respectively to compounds (pure entities) and mixtures (impure entities) in chemistry:
(a) In the first example acousmatic listening can entirely contradict the specification sheet. The skilled reel-maker can take in his public by creating a coherent object in which different sound sources are superimposed or linked together; he can also play a single source yet give the sound several phases, to the point that it could be broken down into as many distinct objects. The sounds in question are compound objects made of several simultaneous elements or composite objects formed of several elements in succession. There are also dubious cases, unstable compounds, where the analysis depends to some extent on the listener’s conditioning.
A high-register piano sound is, objectively speaking, a double sound. It can be considered as compound if through habit it is perceived as a single sound to which the striking of the hammer simply adds a particular color, or it can be considered as composite if through a different type of conditioning the ear is accustomed to hearing the striking of the hammer and the resonance of the string one after the other.
(b) Some examples do not fit into this situation where combined objects coexist. What happens, for example, if at the end of a long vibration (string, cymbal, etc.) another element is added, embroidering its own particular anecdote on to the sound? We are confused because the unity of the first sound wins out. Rather than refer to compound or composite sounds we will say that a sound such as this is accidental. We will call the added sound the accident of the first.
An incident can just as well happen during the course of a sound. This is what we will call interference caused by some technical fault that is added to the sound and is neither wanted nor heard as a property of the sound. This type of impurity may be excessive background noise, a poor adhesive, a copying error, distortion, and so forth.
By linking a metal sheet and the bass piano through superimposition and fade-in-fade-out we can, as may easily be imagined, obtain an extremely varied collection of compound or composite objects. Some will be noticeably coherent and unique; others will display two distinct phases. Generally, the box they are classified in will ultimately depend on the listening context and intention.
Even when they are well linked, the two sounds (metal sheet and bass piano) will present marked differences of harmonic mass: so we really must make a distinction between two musical objects (according to the criteria that will be explained in book 6); but if the dynamic profiles match and complement each other to the point that they simulate true unity of facture, we will only hear one sound object. We can see from this example to what extent the application of the major distinctions in our typomorphology will depend in practice on particular applications and the listening criteria at the time. Besides, the role of our classification system is not, as we have seen, to put permanent labels on sounds but to open up the ear to the richness of the contents of sound; it is not a question of giving the composer of the future a stupidly inflexible code but of making him attentive to the way the object can fulfill a variety of sound functions.
So, generally speaking, we could say:
1. From a morphological point of view we can never be quite sure that the object is definitively split up into its “isotopes.” A so-called thick sound may perhaps, after several months or years of training, appear to be reducible into constituent sounds. Other, composite, sounds will acquire unity when used in a particular way. Morphology, unsurprisingly, thus depends just as much on other surrounding sounds as on the listener’s conditioning. What is important is to agree on a quick way of naming objects, describing them and still leaving a way open for breaking them down or building them up again.
2. From a typological point of view we are tempted to say this sound is a sample, a cell, a homogeneous weft. Nothing is as simple as the paradigm in an academic paper. Indeed, a disorganized scraping of the bow, a short sound bursting with values, a sustained bass note are by and large amenable to this type of classification, but here also context plays a part.
(a) First we will try to communicate the interdependence of a group-object and its constituent parts, whether they are simultaneous (compound, like a piano chord) or in succession (composite, like a drumroll). In our notation we will indicate the difference between these two levels of analysis (group and elements) by using upper- or lowercase letters.
Thus in a complex X bell sound we can discern several sounds that are themselves complex x1, x2, x3 . . . or even partials n1, n2, n3 . . .
We will notate the whole sound: X (x1, x2, x3, n1, n2, n3 . . .).
But it may be that I also wish to examine one of these elements separately x1 or x2, n1 or n2, but without ignoring the fact that it is part of X and is influenced by it. This time I will write: x1 (X) or n2 (X).
In these different formulae, then, the first term unambiguously indicates the object that is the focus of my interest: notated in uppercase letters, it contains the objects notated in lowercase letters in parentheses after it, or, notated in lowercase letters, it is itself contained in the group notated in uppercase letters in parentheses.
(b) But these symbols only mention the link between the group-object and its components, and vice versa; they do not take into account the relationships between the component objects. We have, in fact, separated the letters representing these component objects with commas, without prejudging their relationships of simultaneity or succession as a consequence. The double-high-register piano sound, for example (or any other percussion instrument where the noise of the percussion is heard at the same time as the resonance), is an extreme example of a compound sound: the two elements coexist in duration and are closely linked by their causality. To denote this type of link inside a tonic note in conventional listening, we will use the two juxtaposed symbols N (x.n). But we could also highlight the typologically double nature of a sound such as this by ignoring conventional listening and then simply write x.n, a formula that represents a listening mode where equal attention is given to both aspects of the sound.
On the contrary, if the elements succeed each other in duration, we will use the addition sign to indicate that we are dealing with a composite object. This is the case with the drumroll notated X (x+x+x+x . . .). The same sign will be used for a juxtaposition of two parts inside an object, such as can be obtained through editing. A bowed staccato continuing with a held note gives us another example of composite sound: it has unity, but it also follows on from a note N′ and a note N. It could be considered their sum, and notated N′ + N. The staccato itself, analyzed separately, would have the following formula:
N″ (n′ + n′ + n′. . .).
The above indications are not enough if we happen to have a fade-in-fade-out type of development: the squeaking of a metal sheet followed by its resonance gives an example of two very different, though not independent, types of sustainment. The resonance is already there during the rubbing, but it is masked; when it comes into the foreground, the squeaking noise has disappeared but remains present indirectly because of the partials it has released. Thus, the two elements are too closely linked for the addition sign to be used but not simultaneous enough for the multiplication sign; so in this case we will use a forward slash, representing the fade-in-fade-out: X′/X. Which introduces a time element and distinguishes this composite sound from the compound sound X′.X.
(c) We can generalize this system of notation by combining these various possibilities; for example, a trill in the high register of the piano could be notated this way:
Σ (x.n1 + x.n2).
It may be useful to explain that in compound sounds, X.N for example, the order in which the component elements are juxtaposed does not necessarily determine their arrangement; this notation is simply an attempt to take into account the simultaneity of a number of different elements, any of which may be called on to dominate depending on the context.
It would be good, first, to place the neophyte in a setting for acousmatic listening, the most favorable to the deconditioning necessary for a full understanding of the analytical side of musical research, by keeping him away from any reference points and also from preoccupations about how things are made.
But it would not be good teaching to keep him for too long simply practicing hearing: soon we must give him the opportunity to close the loop, to experiment all by himself on the concept of the facture of a sound. In short, we must give him sounds to make.
The obvious route is to discover the phenomenon of sustainment and its corollaries: allure and grain. Through sustainment, morphology becomes clear and typology justified. So by bringing together what conditioning had so completely separated—a violin, piano, or cymbal pizzicato, on the one hand, and their “held” notes, on the other—we get to the idea of the object and are freed from concentrating on specific instruments. The various methods—rubbing, breath, or resonance—that produce the same grain effects, the vibratos—with the fingers, a reed, or the glottis—that produce similar allures very quickly lead to a sense of facture and facilitate the discovery of reduced listening: too many different causes produce the same effects for us to go on trying to discover the explanation of sound objects in these causes alone.
The beginner will then endeavor to apply these notions to make sounds for himself by working on sustainments that give him different morphologies and also different types of sounds. The important thing is to take care over the diversity of sources, the guarantee of a proper exercise in both making and hearing, enabling him to learn to ignore the criteria usually thought to be the only important ones but that obscure most listening. Thus, whatever the sound body, the tessitura, the means of sustainment, the nature of the sound itself (tonic, complex, or variable), the experimenter will find himself forced to draw out only the elements that present themselves to the diligence of his gesture as of his listening: factures or sustainment criteria.
Below, by way of example, is the plan for a “reel of sustained sounds” as an exercise in making or hearing in the sense of “prose composition.”
Reminder of the definition: the sustainment of a sound object is what maintains it in duration. Sustainment is therefore different from causality (especially the initial causality, on which the attack depends); it determines the continuation of the object, an essential element in duration.
(A) Categories of sustainment
Sustainment conforms to different “laws” or categories of its own causality:
1. No cause of duration: nonexistent or short-lived sustainment (e.g., whip, woodblock).
2. The environment may prolong or color the sound after the attack: sustainment through resonance (e.g., piano, guitar).
3. Regular prolongation of the sound by a renewed input of energy caused by a single law: sustained reed, more or less regular rubbing of a bow, electronic oscillation, and so forth. In this category we will distinguish among:
(a) fixed sustainment: a strictly constant supply of energy;
(b) modulated sustainment (predetermined dynamic): electronic sound for example; and
(c) active sustainment directly from the performer (wind or bowed instruments; ondes martenot).
4. The sustainment is not regular, although there is a single causal law; in this case it will be:
(a) irregular fluctuating (rubbing cymbals, maracas . . .);
(b) disordered (clumsy bowing, drumroll . . .).
5. The energetic input no longer has a single law, but:
(a) a series of willed or chance inputs (cascade of objects, rapid incessant manipulation of a potentiometer);
(b) the repetition of one identical fragment (staccato, tremolo, beating, etc.).
6. Finally, two or several categories of sustainment may coexist in one sound; then we will use the terms:
(a) compound sustainment if they are juxtaposed;
(b) composite sustainment if they follow each other.
The part of the reel with this exercise on it will have to have between six and twelve examples of each type of sustainment, if possible from each of the three domains: traditional, concrete, and electronic music—that is, in total eleven series (1, 2, 3a, b, and c, 4a and b, 5a and b, 6a and b).
(B) Transition between categories of sustained sounds
The point here is to show that there are no fixed barriers between these various categories:
1. Continuity between short-lived sounds, which always have a little resonance, and resonant sounds.
2. Continuity between resonant and sustained sounds, particularly with electric forms of sustainment.
3. Sustained sounds: subtle transition from fixed to active sustainment.
4. Register, clearly continuous, of all more or less fluctuating active fixed sounds (thus a good singer or violinist is distinguished from a bad singer or violinist).
5. Allure of prolonged sounds. The characteristic allure of a fluctuating sound (vibrato) will lead to a prolonged cyclic sound, while an irregular fluctuating sound will rapidly lead to a disordered sample.
6. Finally, prolonged types of sustainment, cyclic or irregular, rapidly destroy the coherence of a single object by producing either cells, samples, motifs, or sequences.
In fact, the idea and the perception of sustainment presupposes a certain morphological coherence that establishes the object; otherwise, we come back to a typological problem.
The firmness of this instruction will have been noted. In some respects it seems to go against classical instrumental teaching. The latter, in effect, concentrates on a specific instrument, from which an increasingly skilled technique draws objects that conform more and more to a given musical code and a certain aesthetics of sonority. But the training we advocate uses many instruments, some of them with age-old pedigrees, while others are unnamable or have to be thrown on the scrap heap after a single trial. If we want to be free of systems and generalize the use of sound bodies, we must bring all possible types of sonority into play regardless of the hierarchy. Moreover, we give our beginner no external model for making sound; it is enough for us if he learns to handle the bow and the bass drumstick, the microphone and the potentiometer judiciously and not without skill. To make what? Certainly not sounds that are valid according to particular musical criteria (which, in any case, it would be very difficult to define at this stage of research) but simply sounds as “decontextualized” as possible from the context of the traditional musical system and, at the same time, as successful as possible in terms of interest, originality, and subtlety, in other words, in their form and content, evaluated by reduced listening.
Such work, we must admit, involves a certain aesthetic sense. We should note, however, that an aesthetics like this is still instinctive, irrational, sensual almost. Its demands are not, for all that, any less urgently felt than elsewhere. Thus, the experimental musician will say quickly and decisively that a certain sound is “very good,” another “suitable,” and yet another “of no interest.” What is certain is that the ill-defined freedom given to the performer of “any sound at all” cannot be fully exploited unless that performer submits to two disciplines: one involves the learning of new instruments, leading to practical virtuosities in making and recording sound; the other consists in rediscovering, through an imagination freed from known sonorities (which, in any case, does not stop him from using them), a way of reinventing sound. Is it necessary to add that without diligent practice, together with original gifts, we will never attain a standard high enough to inspire a generally accepted aesthetics?