23

The Laboratory

23.1. ELECTROACOUSTIC PREREQUISITES

It is good to define the “musical system,” to analyze its contents, and to understand the way it is structured. But we must have the tools to do this; without an electroacoustic laboratory the above analysis would have been impossible. When an artistic investigation becomes analytical, it may need equipment, not just to measure objects but also to display them in a different light, to “play them,” as it were. The laboratory we are, in fact, going to describe, although based on an electroacoustic technology, will be little more than a musical instrument or, more precisely, the instrument of our experimentation on musical perceptions.

So the reader may be reassured: what we are going to say about this is no more than what a user should know about his car or a violinist about his violin, except that now we are not talking about playing in a concert but manipulating sounds.

It may seem surprising that several of the sound examples quoted until now were not taken from the electroacoustic context. In fact, if we have chosen our examples from the traditional domain, it is so that the reader knows what we are talking about and can benefit from the analyses involving such sounds, in particular the correlations in book 3.

Now it is quite obvious that without the electroacoustic system this work would have been impossible and even unthinkable, since we would have remained at the stage of the cultural conditioning described in book 4 without having any opportunity to become aware of it; we would have been going round in circles over aesthetic considerations (as is often the case with contemporary music: a prisoner of the system yet bewildered by its development). Only by overstepping the “boundaries of the blueprint,” by tackling obstinately eccentric objects, can the researcher become disorientated enough to be forced to rethink everything. But we need both unusual objects and new ways of treating them. In this chapter we will discuss these new ways of making and hearing, already implicit in book 3.

In reality we should need a whole work. The first chapters have already introduced the idea of this electronic transformation (Umwandlung, to use an expression dear to Hermann Scherchen)1 into music. We will limit ourselves to giving the framework of a development that we will doubtless need to add as a supplement to this treatise, which will only deal with the fundamentals of the research. Besides, adding more technical details unwisely or prematurely runs the risk of putting off the readers for whom it is written, who would think it is intended only for technicians.

The essential understanding, which everyone should make his own, is not about electronic sound synthesis, which we have criticized: through the deceptive facilities it provides, it encourages musicians to use electronics in the spirit of the old system and to work with frequencies, tempos, and levels as if they were using real musical values or elementary perception criteria. On the contrary, it is all about discovering that the electroacoustic system, where any sound at all is concerned, is an extraordinary piece of investigative equipment, which brings about the desired deconditioning and is also the instrument for the musical and sound analysis we lacked.

23.2. THE ELECTROACOUSTIC SYSTEM

We intend to describe the electroacoustic system briefly without going into detail about the equipment itself—although it is necessary to understand its basic functions—but recognizing the opportunities it gives at its different levels for sound intervention and manipulation (see fig. 26).

FIGURE 26. The electroacoustic system.

We will identify nine stages or links in this chain, going from natural or artificial sound events to sound projection with loudspeakers at the final stage, through a number of electric or sound modifications with extremely wide-ranging possibilities and repertoires.

1. Sound domains

As any sound can be recorded, coming from nature, from human or animal language, from raw or refined, willed or involuntary, elementary or complex sound events, at various stages of artistic or social development, we are virtually in possession of an immense wealth of experimental sound material.

2. Sound factures

These, which are generally implicit in natural sounds, must be made explicit as soon as we are dealing with deliberate manipulations. We could suggest analyzing them as follows: into (1) sound bodies and (2) ways of using them, making them vibrate or maintaining those vibrations.

We can see that traditional musical instruments are a combination of (1) and (2), plus something else: a registration, which we will not permit ourselves to assume in sound in general. Electronic sounds, as well, are involved with sources (1) and ways of using them (2), through a particular system of multidimensional registration.

3. Microphone recording

This may involve one or several microphones, their positioning in relation to the source, and the conductive medium that links them to the sound body (air, water, contact, intermediary device).

4. Electroacoustic modulation

Each microphone delivers an electric current that faithfully reproduces the elastic vibrations it has recorded from the source; moreover, it can be used together with:

• an attenuator or amplifier, D1, D2, D3 . . .

• a set of filters or corrector, F1, F2, F3 . . .

• a reverberation device, R1, R2, R3 . . .

In addition, the various microphones are linked together by a mixing device M, which controls the modulations from each one and provides a resultant modulation.

5. Recording

Whether this is done by mechanical, magnetic, or optical engraving, it gives a practically faithful “sound image” (except for corrections in engraving and playback) from M (see the reservations made in chapter 3), as we can verify by listening to a control loudspeaker LS0.

6. Manipulating the recording by editing and mixing (transformations)

This is the possibility of intervening either in the succession or the superimposition in time of the sound segments, including the possibility of shaping the dynamic profiles of a given segment or adjusting the levels of two simultaneous or successive segments. In this box we will only put manipulations that do not affect the integrity of the matter of the sounds and only interfere with their form. Only scissors and the copying or mixing potentiometer are used. Finally, we obtain a sound object that combines segments taken from one or several component objects, having perhaps modified their dynamics.

7. Manipulating by modifying the recording (transmutation)

This is what we have just excluded: it involves not only the repetition of the last two manipulations, electroacoustic modulations F′ and R′ (described in 4), but in addition speeding up and slowing down or total transposition, where the time and spectrum of the frequencies are linked by an inverse multiplication or demultiplication coefficient. These manipulations, of course, give rise to new possibilities for mixing M′.

8. Synchronous playback

This is a parallel technique to multiple-microphone sound recording, followed by blending; we can either blend two or three magnetic or optical tracks, making sure they are in synch, or reproduce each of these tracks on a separate loudspeaker channel (still in synch) through multitrack playback, in which each channel may in addition be undergoing some of the electroacoustic modulations mentioned above. Nevertheless, for the sake of simplicity, we will assume that the latter are merely corrective—that is, that they are intended simply to get rid of possible faults in transmission to the loudspeakers of the signals on the recording medium, or to improve their output in high fidelity.

9. Spatial projection

We must be careful to distinguish between this last stage and the previous one. Not that it is not implicit in it: there is no multitrack playback without the beginnings of spatialization. But further levels of freedom come from the positioning of the loudspeakers, whether they are linked together or spread out in the field of sound reproduction, and the possible movements of sounds from one loudspeaker to another (spatial kinematics of sound projection).

The essential aim of spatialization, which is often confused with some strange myth of “spatial music,” is to improve the definition of objects through their distribution in space, since it so happens that the ear distinguishes two simultaneous sounds better if one comes from the right and the other from the left. We are not dealing here with a luxury added on to our hearing but something to facilitate it. Before even mentioning space and sound architecture, we should talk about the identification of objects and their coexistence. Where they are is of little consequence; it is what this enables that is important: an incomparably clearer, richer, more subtle perception of their contents. In the same way, binocular vision gives the third dimension and by putting things in perspective with each other allows us to judge their properties and relationships better.

Finally, we should note the difference between stereophony, which consists in reproducing real sources in space, and spatialization, which consists in dispersing recorded objects (hence tracks) on loudspeakers judiciously situated in the three dimensions of space.

23.3. REPERCUSSIONS OF THE SYSTEM ON FUNDAMENTAL RESEARCH

As we have said, these nine points are so many chapter headings for a possible “treatise on the use of electroacoustic resources in music” and include developments in applied music just as much as in fundamental sound research. At this point we will set aside everything that is not necessary to the latter, that is:

1. We will enter domains of sound only to take samples from them. We will not investigate them any further. We will, however, briefly describe the general workings of the way the sound body is used in music.

2. We will endeavor to discover the general rules that the facture of sound objects should obey—that is, the essential rules for identifying, choosing, and classifying them.

3. We will not go any further into sound recording except to suggest, as we have already done, that there is a difference between the “shot” of the object it is able to give and live listening (a phenomenon analogous to viewing angles, enlargement of images, visual salience, planes, and lighting).

4. Generally speaking, we will not allow ourselves to use electroacoustic manipulations other than regulating the level with the potentiometer, for several reasons, the main one being as follows: we observe that they merely muddle the characteristic features of the objects, without radically transforming or changing them, and that we perceive them in themselves as the trace of the sound, and therefore as transformation; in general, then, they distance us from effective understanding of the objects. But as laboratory manipulations we will use them for the purpose of analysis or when appropriate as correctives.

5. Ultimately, it is the most faithful recording that will be of most use to us—faithful, of course, to the object and the recording “angle.” Faithful recording, without transformations, will not dispense with the need to go back to sound recording, combined with skilled, inventive manipulation of the sound body; rather the contrary—in the same way the photographer is keen not to use film gimmickry, and wants the photograph to examine faces or visual objects as frankly, as eloquently, as possible, simply using the ways the camera has of reinforcing the real as it is seen by the eye. So we reject electroacoustic sound cosmetics but claim complete liberty to approach and define a “type of sound” (theoretically perceptible to the ear) radically changed in its proportions by microphones.

6 and 7. The same attitude guides our preferences in the matter of using manipulations either by editing or filtering. Editing retains the matter of the object (6), filtering its form (7); we value these two manipulations because through them we can lay the emphasis alternately on each of these two fundamental aspects of the morphology of objects.

8 and 9. These techniques mainly concern applied experimental music. Of course, they also enable us to study other properties of listening. For example, in chapter 3 we saw the impact of the visual (section 3.7) in sound impressions. We should never forget that objects dispersed in space do not have remotely the same effect as the same objects, just as synchronous, coming from a single loudspeaker.

It remains for us to develop the following points:

• how a sound body functions when it is used in music;

• the opportunities for inventing sound objects by means of factures that use the combined resources of sound bodies and sound recording;

• the techniques for preparing the object from the recording of it;

• more advanced and recent techniques after critical examination of the earlier ones, it being clearly understood that, reserving the work of defining research approaches and experimental disciplines for a later analysis, we will limit ourselves here to a description of ways of using the equipment available.

23.4. DESCRIPTION AND USE OF SOUND BODIES

A good technological description of sound bodies should enable us to generalize the concept of musical instrument and lead us to the beginnings of a deconditioning from traditional practices, through analyzing what belongs to sound in general and to the musical in particular in each of them. We will borrow the Baschet brothers’ analysis, the most classical possible, yet it led them to brilliant innovations in instrument making, particularly through the use of vibrating rods and wands suspended in the air by ingenious systems of their own devising. We could comment along with them that there are only two essential elements in a sound body: what vibrates and what causes vibration. A child playing with a rubber band is like the child with the grass. Each of them possesses the minimum—a vibrator: rubber band or grass; a stimulator: a finger using pizzicato or breath that sustains the sound. Here, too, we can see the two basic types of sustainment: one short-lived and “passive,” the other permanent and “active.”

The third element, dispensable but frequently added to a sound body intended for music, is a resonator. Yet we must point out that the term is ambiguous and masks two functions, which must be differentiated from each other and which could lead on to two distinct devices. The function of a resonator, properly speaking, is to characterize or transform timbre through the addition of formants (commonly occurring zones of frequency in the spectrum of a given sound) or acoustic filtering. Thus, the violin’s belly colors the vibration of the string without imposing any particular formant on it, whereas the buccal and nasal cavities, used in various ways, give rise to specific formants that turn the sounds from the vocal cords into different vowels. Quite different from this is the “coupler” or “radiator,” intended to adjust the sound body’s “acoustic impedance” to the surrounding conditions: this, for example, is the role of the horn on trumpets or loudspeakers, which in Baschet instruments is taken on by antennae or metallic diffusion surfaces, which in fact improve acoustic output to an astonishing extent.

Finally, from a viewpoint this time frankly favoring the musical (and no longer only the musicianly), we find a registration, that is, a modulation device that in most cases is nominal pitches (percussive keyboards or holes in a bellows).

The interaction of the five elements—two (vibrator and stimulator) indispensable, two (resonator and coupler) desirable, then the fifth, both valuable and tendentious (register)—gives, as may be expected, innumerable combinations through which, provided we can produce them effectively, we could already to a great extent rethink “live” sound making. When we go on to consider the potential of linkage with surrounding conditions, of recording itself, of amplification and the various electroacoustic modifications, we are dumbfounded at the extent of the opportunities opened up to us; there is nothing to equal them except the ignorance of our contemporaries and their idleness as they wait for an electronic Father Christmas who will bring them a music machine, ready-made.

23.5. FACTURES: INVENTION OF SOUND OBJECTS AND SOUND RECORDING

Perhaps we have been using the word facture incorrectly, in a broader, more practical sense than the one that has been and will be used in typomorphology. In this latter connotation it will mean finding formal signs in the sound object of the way it was made. Here we will use it to refer to all the ingenuity (rarely revealed in acousmatic listening) that goes into the creation of varied, formed, willed sounds in an infinitely more extended register of matters and forms than either the musician conditioned to registers or the physicist obsessed with filters and frequencies could ever have imagined a priori. If, for example, I put a piece of sheet metal in front of the microphone, I seem to have neither varied sound elements nor parameters of freedom in the way of playing and hearing it. A traditional musician will immediately demand a set of metal sheets arranged in a register, and an electronics expert will rush to dissect this sound into slices of frequency or millimeters of duration. And so the evidence passes them by: all sorts of musicianly initiatives are possible, which have nothing to do with musical registration or acoustic measuring. From the studio point of view a host of sound bodies can be made to vibrate in all sorts of different ways. From the listening booth point of view, as we have seen, the sound recorder can make an original “take” of the sound object by adjusting the position and the setting of the microphones, just as the cinematographer or the “lighting” director in the cinema can choose the angle, the distance, the lighting for the object to be photographed. These apparent restrictions conceal immense potential.

So what can I do with this piece of sheet metal, put like this in front of active “phonography”? Like the young violinist, remember the teacher’s advice: one to be ready, two to play, making sure this is one movement. Playing the violin linked muscle and bow, a kinesthetic reflex and a mechanical sense. I rediscover all this, much more crudely, in this sheet metal, which I scratch, hit, stroke in all sorts of ways, overcoming the human sense of dignity that would make this situation ridiculous. Finally, depending on the position of the microphone (or microphones), the setting of the potentiometer, the filters (sometimes in real time), I obtain surfaces, a detail, enlargements, a new coloration of every variant I draw out of the same gesture, which does not astonish merely the uninitiated; the professional, trained in a technique of fidelity, hardly ever knows what he holds at his fingertips, that often unimaginable extension of a technique that is all unobtrusiveness.

Now we will go back to our sheet metal and count the degrees of freedom I can combine in order to obtain so many and such varied objects from this one body that the acousmatician will not believe his ears. All sorts of percussive or sustainment factures, using many a stimulator, are possible: various drumsticks, used in all sorts of ways, from simple percussion to a continuous roll; scratching, also going from the mellow to a screech, through staccatos where the material of the stick comes into play: wood, metal, rubber, and so forth. Moreover, all the interesting places on the sheet metal can be explored: the surface, the edge, and practically all the points where it is fixed or suspended; we can use various tensions, degrees of flexing, and, finally, perhaps, coupling it to other resonators.

From the point of view of the microphone there are all these different approaches, involving the surroundings right to the point of contact, and the opportunities given by positioning them in responsive places in the acoustic field. But we could also imagine a temporal intervention in the facture on the part of the operator. If we wait until after the attack to turn up the potentiometer, we obtain a sound that will be astonishing in its originality as much as its authenticity: it is both a sound we know well and one we do not recognize. Where fairly long objects are concerned, therefore, a skillfully handled potentiometer can be as effective and subtle as the violinist’s bow.

We can, finally, combine the sound of this sheet of metal with other types of percussion, such as the bass register of the piano, which is so similar to it. Thus we compose a sound object from two combined sources; we can make a hundred attempts before retaining the one that is preferred for the profile of its form, its scintillating matter. A technique such as this rapidly enters the zone of rhythmic articulations and melodic modulations, in short, live composition through juxtaposition and superimposition. We can see here the favored procedures in concrete music, in direct contact with the stuff of sound, like the sculptor with his clay.

23.6. PREPARING THE OBJECT

We have already mentioned the theoretical dissection of the object along the lines of its dimensions in physics. If we situate the physical object in its reference trihedron of ox, y, z (see fig. 27), taking time as the abscissas (x-axis), intensity as ordinates (y-axis) (oz vertically), and oy as end axis, the frequencies, the normal “cutting planes” at ox, y, and z are cutting at a given moment, cutting at a given frequency, and cutting at a given level. Cutting in accordance with time is not only the easiest (all that is needed is a pair of scissors) but also the most reversible: it works just as well for analysis as for synthesis. Cutting according to level is of no particular interest and in addition is very difficult to do electronically. Finally, cutting according to frequencies, by means of a set of bandpass filters, presents the pitfalls we all know about: it rarely isolates a real band of frequencies, due to the demanding technical norms it requires; moreover, the “form” of the object victoriously resists this dismemberment. As for the three-dimensional representation of the physical object, it is little more than a curio. Once we have had the satisfaction of seeing the perfect chord or the beating of a gong in the form of abstract sculpture, we can only regret this costly waste of time. The author, who has spent time on this in the past,2 cannot warn the reader too strongly against this amusing piece of physics, which can be of no use whatsoever to music. We will return to the recording booth: what manipulations have we to hand?

FIGURE 27. The reference trihedron.

We have already mentioned the first: intervening in duration, with scissors and editing.

The second is with the potentiometer: it consists in raising or lowering the overall levels of the object, or intervening in its profile.

The third, rather than “filtering,” consists in total transposition by speeding up or slowing down, transposing both the spectrum and the rhythm of the object at the same time. In this way the form is dilated or condensed, depending respectively on whether the sound matter is transferred into the bass or the high register.

The above interventions are techniques belonging to physics. Our only originality is in making them solely for the purposes of musical evaluation while drawing a clear distinction between these two aspects of our experimental activity.

Suppose, in fact, that our experimenter is endeavoring to differentiate form and matter in objects, finding it difficult to evaluate both at once: he might decide, in his experiments, to emphasize form at the expense of matter or vice versa. Suppose, then, that the form, too complex and in a rich harmonic resonance, makes it difficult to listen to one moment of the sound. How can this instant be isolated to allow the ear to hear it comfortably?

Phonogènes are intended to provide an answer to this question, much more than to perform harmonic noise marches.3 When we only had the total transposition they made possible, we discovered this additional use, which resolved the problem after a fashion: we slowed the sound right down and located the zone of interest. We cut a piece of sound out of this zone, appropriately dilated, and “looped” it. Once it was speeded up again, it found its tessitura and played itself over and over again. Of course, the samples retained a certain dynamic, and it was difficult to “homogenize” a loop like this properly (i.e., cancel out its form). Nevertheless, we must say that despite these imperfections, it enabled the experimenter to take hold of a slice of listening and a temporal element of its form.

Here is an example of the importance of such crude manipulations, taken from the beginning of our research. It would never have occurred to anyone to compare a bell and a wind instrument. Now since 1948 our first and crudest experiments with the “cut bell” have led us on to this new way of looking at sounds and establishing connections between them, which neither acoustics nor traditional instruments would ever have suggested. The “bell loop,” in truth, was at that time no more than a closed groove from a bell taken at a carefully chosen moment of its resonance, and it made a sound like a flute. Hence our initial idea that the concept of instrumental timbre, hardly at all linked, as was claimed, to the presence of a characteristic spectrum, had to be completely rethought.

Therefore, to pursue these studies on timbres, we had to isolate the component matter and the component form, which now appeared as two criteria of musical listening, independent of elementary acoustic parameters. The systematic variations of these two components in relation to each other clearly indicated the most interesting direction for experimentation.

23.7. THE TRANSPOSITIONS OF THE OBJECT

The above techniques, in their crudeness, are like irreversible biological incisions. The combination of scissors and glue, the potentiometer and the phonogène, did allow us some syntheses, as well, but at the cost of what damage! But this is how our research progressed: successive copies, fragments dilated then condensed, levels raised then lowered, nothing that had much chance of appearing pleasing to a musical user.

That heroic period was set to be superseded by a happier, better equipped, phase, benefiting directly from its trials and errors. The important thing, as we have suggested, was to bring into prominence notions such as sound form and matter, without which there is no point in thinking about the device we will describe now and which, precisely, allows us to grasp these independently.

Springer’s Zeitregler initially very modestly addressed a need felt in broadcasting studios, which are obliged to observe strict time schedules.4 How could a symphony that lasted twenty minutes be made to fit into a slot of nineteen minutes? This was the mundane question addressed by the device invented by Springer, which for our own part we consider worthy of nobler application. In effect, in some way it resolves the problem of infinitesimal cutting, which haunted a number of composers such as Stockhausen when he visited our studio in 1952. Despite our rebukes, he insisted on cutting a tape into millimeter segments in order to be able to stick them together again in a different way. The “thousand-piece loop” was famous for the obstinacy and the trials and tribulations it represented. In practice, cutting taken to this extreme goes beyond the capabilities of the experimenter, as of the tape recorder, for reasons of output and sound quality. Now, Springer’s device resolves the general question, which is very close to the one raised by Stockhausen’s endeavor, of condensing or dilating a sound object that remains the same with regard to form or matter. Suppose, in fact, that cutting and gluing together infinitesimal pieces of magnetic tape is possible. For example, suppose we cut the tape of a sound one second in duration (therefore 380 mm long)5 into segments 1 mm long. We then number these millimeters 1 to 380. We try out the two following edits: in one we stick 1, 3, 5, 7 . . . together (i.e., all the odd numbers) or else 2, 4, 6, 8 . . . (i.e., all the even numbers); in the other, assuming we have a second tape identical to the first and cut in the same way, we stick 1 followed by 1 from the other tape, then 2 followed by 2, and so on. In the first we obtain a sound half the length, in the second twice the length. What will happen if we play these two edited fragments at the initial speed of 38 cm per second? As they are being played at the right speed, the components of the matter of these sounds (pitch, harmonic content) will probably be the same as in the initial sound, although one tiny fraction out of two is absent, or every fraction is repeated twice. But one of these sounds will be half, the other twice, as long—that is, its form will have been condensed or dilated: thus this first procedure acts on the form without affecting the matter of the sound. If, however, we play the short tape at half speed and the long tape at double speed, we transpose the entire sound spectrum an octave lower in the first case and an octave higher in the second. So here we return to the initial duration and therefore the form of the sound, but we have transposed its matter. In this way the Zeitregler enables us to dissociate sound form and matter to a greater extent than could be done through the use of the phonogène.

23.8. TRANSMUTATIONS OF THE OBJECT

While the above technique resolves major problems in both fundamental research (morphological examination of the different moments of the object) and applied music (independence of the rhythm—or durations—and tessitura of the objects), it does not yet give the means for that sound alchemy that is every experimenter’s dream: a radical separation between the form and matter of sounds. We indeed imagined a bow stroke that could be applied to matter other than the violin, or a piano note with the sonority of the violin as its matter. If this were possible, would it not be the source of astonishing transmutations, here and in both fundamental and applied research? Even the best known musical instruments would be revolutionized.

Very recently, Francis Coupigny from the Research Service of the ORTF built an initial prototype that granted this wish: the form modulator, in fact, enables us to mold the matter of a sound homogenized earlier to the form of another sound (using the procedures mentioned above).⁶

Over and above the first experiments “just to see,” we can think of other applications for this: the composer-performer would use it to ensure the formal shaping of objects, the sound matter of which would be judiciously chosen. Once more, as can be seen, this is a “concrete” technique, in direct contact with sound.

23.9. ELECTRONIC GENERATORS

Experimental music cannot ignore electronic sources. The sounds they provide, although not audible until played through a loudspeaker, are part of the domain of sound. They are previously unheard by definition, since they are absent from nature. And this is why their appearance in a group of sounds is always accompanied by certain difficulties about their use: our conditioning to such sounds is still fragile. Even if they do not bring about the easy synthesis of all other sounds, as was initially believed, they nonetheless provide matters and forms we have certainly not yet exploited to the full. Another reason for taking an interest in them is the extreme ease with which they can be manipulated, which gives great flexibility to sound and musical experimentation.

We will give a rapid overview of the resources generally used in the electronic production of sounds.7 On the one hand, there are generators that produce various types of electric signals; on the other hand, there are means for modifying or combining electric signals; finally, we will indicate briefly how these devices are used.

1. Generators

Generally speaking, an electronic music studio will have the following sources:

• Sine wave generators, either registered (e.g., in the tempered scale) or of continuously variable pitch. There are often relatively large numbers of the latter, to allow several frequencies to be added (e.g., the components of a given spectrum).

• Short impulse generators (clacking noise) of variable length and frequency of repetition.

• Square or triangle wave generators.

• Noise generators (sounds in which the spectrum contains all frequencies) with adjustable bandwidth.

2. Devices for modifying or combining sound. Their function is to superimpose the signals from the generators on or make them interact with each other. In general they are as follows:

• Addition: one signal is added to another.

• Attenuation, amplification, frequency filtering: these procedures are self-explanatory.

• Multiplication: a signal continuously produced by two given signals is obtained. The applications of this procedure are as follows:

(a) Frequency modulation: a frequency variation, regular or otherwise, is superimposed on a sound of fixed frequency, initially giving a tremolo effect.

(b) Form modulation: the (sine wave, noise) signal is given an envelope chosen in advance: square or triangular, for example. This procedure in particular gives profiles with predetermined attack, development, and decay.

(c) Frequency translation: a given signal is transposed by a given value. For example, the sum of two sine waves of 100 and 200 Hz (one octave apart), shifted 30 Hz, will give sounds of 130 and 230 Hz, which are no longer one octave apart. Special effects can be obtained with this procedure, even the inversion of spectra.

3. The electroacoustic procedures already described, in particular:

• Reverberation: an artificial echo (obtained mechanically or electronically) is added to the sound.

• Magnetic tape interventions: either the playback speed can be modified, thus obtaining total transpositions, or pieces of tape can be cut out, stuck together, and edited.

4. Performance devices to facilitate the exploitation of these various techniques:

• Keyboards, for instruments with registered or registrable sounds (from the Trautonium and similar instruments for manipulating new sounds to the ondes martenot for using new timbres in a classical manner).

• Manual adjustments with continuous variation: knobs, faders, triggers, ribbons. We should point out that keyboards have been constructed with keys that can be pressed down to various depths for adjusting the intensity, or with lateral movement to start and regulate frequency modulation (tremolo).

• Desks for switching on/off, selecting and mixing.

5. Various devices for encoding and recording procedures that enable a particular sequence of sounds to be prepared in advance, for example in the form of a series of perforations in a paper tape, which then start up the necessary equipment and combinations in the order and at the times required. Under this last heading we could also include automatic composition procedures using electronic calculators (ILLIAC experiments in the United States, P. Barbaud’s recent experiments in Paris). We should point out, nevertheless, that these latter experiments are peripheral to our study, inasmuch as they postulate a fixed structure for musical language (e.g., the rules of traditional harmony or serial music): for us, obviously, the discovery of the properties of sound objects and musical structures must come first in the logical order of priorities.

23.10. THE BARE ESSENTIALS

Since the simultaneous launching, just a few years apart, of musique concrète in Paris in 1948 and electronic music in Cologne in 1950, several studios have been fitted out worldwide.8 What historian, what sociologist who is both philosopher and technician, will learn the lesson from these bold yet speculative experiments?

We will limit ourselves to two observations, not to defend our position come what may but to enable other researchers to work with greater peace of mind. We would suggest that musical research is not necessarily bound up with the implementation of some costly piece of technical equipment requiring specialized skills. Music that is called electronic today often has more acoustic than electronic sounds and in relation to the electroacoustic system uses the manipulations we have described, as well as electronic synthesis. A second observation is about the principles themselves, the intellectual rather than the material investment. An electronic studio is, in fact, usually set up by groups who see it as a technical project—installing a certain number of pieces of equipment. Now, we would point out that the essential does not lie here; it lies in the ear and its largely unexplored musical qualities. A few specialized studios may pursue technological research in this way, and others will benefit from it one day. But the common run of researchers, the innumerable musicians, especially the youngest, who are wondering about this new approach to the domain of music, should know that the essential is within their reach and almost their financial means. To carry out appropriate experiments, they doubtless need a good professional recording studio, not really specialized; but for their training and their “studies” all they need is an amateur tape recorder and microphone, which initially cost no more than a traditional musical instrument.

It is not enough, however, to possess this instrument; they also have to learn to use it. First, they must have a minimum of technique, which, whatever anyone may say, does not require any specific notions from mathematics or physics but, on the contrary, an operator’s abilities: the sound engineer’s. Then there is learning to make and hear at the level of the object, which is what this book is all about. This fundamental point is nearly always passed over in silence: the silence of ignorance or human self-assurance, a flight into the technical, a faith in machines that will take on the responsibility for a music for man without man’s having to learn it? Doubtless all of this. Whatever others may say about a time when technical marvels are a must, we for our part claim that it is possible to have a musical situation where technique is reduced to a rightly subordinate role. We would ask the researcher, however, to submit to an essential personal and group discipline, to follow a simple training process, to take ownership of his technique in order to work at both his instrument and his ear.

Appendix A

(Cf. section 23.7)

THE TIME REGULATOR

The pitch and duration of a sound recorded on magnetic tape are proportional to the speed and duration of playback, respectively, which are, of course, in inverse proportion to each other when an ordinary tape recorder is used. If we change the speed at which the tape passes across the playback head, we will, in effect, have what we have called a “total transposition” of the sound under experiment, which becomes lower the slower (and therefore the longer) the playback and, inversely, higher the faster (i.e., the shorter) the playback.

The device thought up by Springer for his Zeitregler, and that forms the basis of the “universal phonogène” constructed and used by the Groupe de recherches musicales, enables the speed of playback to be separated from the time of playback—that is, the pitch of the sound from its duration. This is how it is done: four playback heads are placed around a small cylinder which turns one way or the other at an adjustable speed on a tape recorder, where the speed at which the tape passes through is also adjustable. The tape adheres to the “playback head” cylinder at an angle of 90 degrees; there is therefore always one head out of four, and one head only, in contact with the tape. The device works in three principal ways:

(a) As an ordinary tape recorder: the heads are motionless, one of them is working, and the tape passes through at standard speed;

(b) To transpose the pitch of a sound only: the speed at which the tape passes through remains standard; thus, the duration of playback is unchanged. But the playback head cylinder is rotated. If this is in the opposite direction from the direction the tape is passing through, the playback is, in effect, at a greater than normal speed (thus, the sound is transposed up); however, the four heads coming one after the other into contact with the tape has the effect that each one replays a part of what the previous one has already played: through these partial repetitions of the sound it is possible to make a sound played back at greater than normal speed last for a normal length of time.

(c) To vary the duration only: the speed at which the tape passes through, and therefore the duration of playback, is changed. Then we rotate the playback head cylinder in such a way that the relative speed at which the tape passes through in relation to the playback heads equals the standard speed—hence the pitch is unchanged. The compensatory mechanisms that here enable us to lengthen or shorten a sound without the pitch being transposed are the same as those described above.

In practice, with a device like this we can obtain the following results: as pitch shifter, several octaves lower, and about a fifth higher; as duration shifter, we can change the normal duration of a sound by (plus or minus) 25 percent. Of course, these results depend on the sound being experimented on (held sound, word, music, etc.).

It is clearly possible to carry out the full range of adjustments between pitch shifter and duration variator and thus obtain modifications of pitch and duration at the same time, which leaves the experimental musician a great deal of freedom of innovation . . . but is occasionally likely to bring some confusion into his manipulations!

Bibliography for Springer’s Zeitregler

Springer, Gravesaner Blätter, no. 1 (1955): 32–37.

Springer, Gravesaner Blätter, no. 11/12 (1958): 3–9.

Springer, Gravesaner Blätter, no. 13 (1959): 80–82.

Appendix B

(Cf. section 23.8)

THE FORM MODULATOR

We have seen that we could act on the dynamic of a sound by simply adjusting the potentiometer, but the precision and speed of this procedure are limited. This is why we turned to electronics to improve it. In a way the “form modulator” is an electronically controlled potentiometer. More precisely, it is a variable-gain amplifier: the relationship between the input/output signal is not constant, as in an ordinary amplifier, but depends on a special voltage brought in at the additional input point, called the “form” point. Thus, the signal entering the machine is “modulated” at its output by the “form” signal. In particular, if the input level is constant, the dynamic of the output signal is the same as the “form” signal.

The “form” signal is an electrical voltage that varies in time, according to a law that the experimenter should be able to choose or determine in the course of hearing. To obtain a form of some sort, Francis Coupigny has created the following devices:

(a) The optical reader: a 70 mm wide band of paper on which a continuous line in the desired form has been drawn passes in front of an optical analyzer system; the electrical voltage delivered by the system is always proportional to the length of the line on the band of paper.

(b) The demodulator: this receives the electrical signal given by an acoustic vibration (coming from a microphone or the playing back of a sound recorded on tape) and at its output delivers a “form” voltage, which may be either directly or inversely proportional to the envelope of the entry signal—that is, the dynamic of the sound it represents.

(c) All other devices that give an appropriately varying electrical voltage can be used as a source of a “form” signal (e.g., a generator of square or triangular signals, etc.).

The general law governing the workings of the device is as follows:

Output dynamic = (original dynamic) multiplied by the

(dynamic of the “form” signal)

If we use the same signal at the main and “form” input points, we obtain this:

output dynamic = (original dynamic)

We can also, with enough connections, obtain the following laws:

output dynamic = exponential function of the original dynamic

(which is, in fact, a violent expansion of the latter); or again:

output dynamic = constant

In the latter case the original sound has become a homogeneous sound.

We should finally point out the possibility of only allowing the “form” voltage to intervene in the modulation process at a particular level of the input signal.