Even before we are blinded by the brilliance of science, we are tempted by the logic that facilitates it. Faced with the phenomenon of music, our first reaction, in the spirit of Descartes, will be to analyze it in our minds, to “break it down into as many small parts as will be needed . . . the better to resolve” the problem it presents to us. And the logical analysis that seems to impose itself unbidden is what makes it appear as a chain of successive events.
(a) At the top of the list we will put the visible origin of sounds, the instrumental gesture: first comes the performer, his physiology, his technique, his art.
(b) Then comes instrumental vibration itself, strings and membranes, which spreads at the speed of sound right to our ears: this is the purely acoustic result of the player’s activity.
(c) This traditional trajectory, these ancestral acoustics are becoming more complicated nowadays: the electroacoustic music system, microphones and tape recorders, amplifiers and loudspeakers, stereophonic or otherwise, broadcast on the wireless or not, are now, in a great number of cases, coming between player and listener.
(d) At the entrance to the ear the physiology and the acoustics of sensations await us. The union between watts and the organ of Corti, the labyrinth and frequencies, is fragile, it is true, but no one will dispute that we must resign ourselves to it.
(e) Then musical impressions occur in the listener, by means of what Henri Piéron goes so far as to call aesthesio-neurones or, more simply, aesthesiones.1 This is psychophysiology.
(f) And the listener recognizes the work the composer intended him to hear, which is truly a stroke of good fortune. This is pure psychology, aesthetics even.
(g) We could, in fact, have started with the composer and his intentions, which are recorded on the score by the musical signs that, according to the good Danhauser,2 enable us “to read and write music as easily as a book,” with the result that all the performer has to do is follow them. We are at the level of pure music, Art.
This is the canonical breakdown, coach by coach, of the train that runs through that fertile land from aesthetics to symbolism, from symbol to the mechanism of the muscles, from muscles to frequencies, from frequencies to the auditory nerves, and from the auditory nerves to you know what.
We could, however, look at this chain from two different points of view: it represents either a series of artistic activities transmitted from a more or less gifted composer to a more or less sensitive and informed listener, by means of the virtuoso and the sound recordist, or a succession of results (score, physical phenomenon, physiological stimulus, musical perception), in which the problem of music is raised only at the two extremities: the imagined and the perceived work. The first point of view immediately brings with it observations that cast doubt on the order we have just neatly established.
First, none of our characters keeps to the limited role that the event assigns to him within the chain. Take, for example, the performer. If the performer really followed after the composer, and came before the listener, simply as an intermediary, instrumental education would be like training a horse, associating a reflex gesture with the sight of the sign written on the score. In reality, we know it is nothing of the sort and that a good teacher bases his instrumental teaching on an analysis of the circuit of communication. The formation of a violin note is not worked on like a reflex but as an intention: the intention of forming a clean sound that “carries”—giving it, of course, the qualities required by the sign (accuracy, duration, nuance) and the intention also of affecting the listener emotionally, through a particular quality (vibrato, holding the sound). Not only does the instrumentalist hear what he is doing, but he hears it as the listener in the concert hall would hear it, like a painter who can paint in close-up what has to be seen from a distance.
We could do this sort of analysis for each of our characters, whose activity, including the sound engineer’s, linked to or anticipating what the others are doing, does indeed go round the entire circuit of communication, from intention to reception. We need only repeat what is common knowledge: from the listener’s point of view an instrumental experience, even by an amateur, can help to guide his listening; as for the composer, he prehears as he composes, anticipating what will finally be perceived, and sometimes he even writes for a particular virtuoso.
This sort of complicity, which allows each activity to be carried out in relation to the others, apparently depends on shared experience. On the one hand, this can operate within any given culture, language, or musical field. On the other hand, in relation to this shared conditioning, it highlights the freedom for personal expression left to each individual talent.
The second point of view, in contrast, which gives us a succession of “things in themselves,” which can be studied for themselves, seems likely to lead us to “objective” knowledge and scientific, universally valid, types of truths.
This movement into science can take place in two different ways, depending on whether more or less emphasis is placed on the abstract aspect of music as a language or on its realization in sound. We will consider the first point now and deal with the second in the next section.
A skilled musician can analyze a work, not as communication from one mind to another but for its own structure, its intrinsic proportions. At a pinch, that is, at the level of abstraction of a perfect score, this analysis has absolutely nothing to do with performance. Appalling instrumentalists or disastrous transmission can indeed “massacre” a classical work; but “massacred,” it nonetheless remains what it is, just as a mutilated body is still a body.
Insofar as it is language, music has, in fact, all the right properties—those, as Husserl would say, of a “spiritual objectivity,” distinct from the ways it can be reproduced or realized:
In the same way we can distinguish the engraving itself from the thousands of reproductions of it . . . present, in each reproduction, like an identical ideal entity. . . . The same is true when we speak of the Kreutzer Sonata, as opposed to particular performances of it. It may very well itself be made of sounds, but it is nevertheless an ideal unit, and its sounds are just as much ideal units. Its sounds are not at all the sounds of auditory perception, sounds as a perceptible thing, which, precisely, exist in reality only in an actual reproduction and in the perception of this. . . . Like the whole, the part is an ideal entity that becomes real hic et nunc only in the shape of individualization in the real world.3
We are not talking about disembodied music but about some highly developed forms of music, based on objects that are so perfectly understood, or at least so exclusively used as signs, that their realization in sound is, as it were, immaterial or at least secondary. Such is Bach’s The Art of the Fugue, where the composer’s genius actually allows any instrumental arrangement of the voices.
It often seems premature for contemporary music to aspire immediately to these lofty heights; besides, for our part, without precluding any reference to a possible musical language at the level of objects, we will endeavor to examine which objects, or better, which type of objects, could be made into the most universal music possible: we will limit our excursions into the abstract side of music to this; in addition, in this quest we will not lose sight of the postulate, for us fundamental, that all music is made to be heard. So we will link any possible musical language to values established at the perceptual level.
Here we should note that the analysis of music into abstract structures—that is, into terms that are meaningful to the intellect and not to perception—has tempted many a mind. Contemporary experiments show that it is possible to go a long way down that road, to the point of looking to mathematical functions or chance theories for the organizational rules of musical language. These attempts are scientific only a posteriori: insofar as they are “experiments just to hear.” It is clear, however, that they are not of the first importance for us, since we wish to hear before understanding and in order to understand. We take it as read that, even if the Art of the Fugue can be completely reduced to a numbers game, the meaning of this game is in its manifestation in sound, because ultimately it is entirely based on criteria of musical perception, which arithmetic may represent but certainly does not determine.
The tendency toward the work-in-itself, which must not only have a totally internal organization and rigorous figuring, but in which the sound components themselves, thoroughly familiar, can all be expressed in abstract terms, “parameters,” is emerging today as the most powerful myth in contemporary music. It seems to tie in with a scientific approach to music based on the elements that allow it to be played.
If this were really how it is, these musical elements would be given simultaneously as signs, and our whole approach through perceived objects would become otiose. We could doubtless experiment on the relationships between these signs, which now correlate to a physical signal, and our musical sensibility, but that would be secondary research and would avoid the arduous task that for us, as will be seen throughout this work, consists in selecting the object insofar as it is meaningful material for a possible music.
So it is worth examining this to see if it is a necessary evil or if there is a more direct way, which would make the construction of a piece of music flow from the immediate use of sound materials that come from a physical synthesis of simple elements.
This seductive hypothesis has not only been advanced but enthusiastically and persistently applied by the electronic school of Cologne, whose theoretician was the distinguished Werner Meyer-Eppler, a physicist from Bonn University, taken from us too soon to pursue its validity and measure its consequences. He would doubtless have moved on from a position that is already out of date. We, however, with the aim of highlighting this point of view, will take the liberty of quoting some of this physicist’s words from a lecture given in 1951:
Because we can now produce sounds electronically, the modern composer is no longer tied to sounds made in advance, which can only be modified within very narrow limits, depending on the directions for performance: he can create his sound material himself. So the initial product he uses no longer has to be identified, as tradition dictated, by its instrumental timbre (for example, oboe or harpsichord sound, etc.).
The terminology of acoustics must be revised, and sounds and noises must be named, not according to their origin but their physical composition. However, in the process, the capacity of the human ear must be taken into account. Since Helmholtz, its ability to analyze acoustic phenomena “from their spectrum” has been acknowledged; and, consequently, given the current development in our understanding of the way the sense of hearing works, it would be appropriate to represent the structure of the causes of our auditory sensations on a time and frequency diagram. The usual notation can also be considered as an approximation of a diagram such as this.
These assertions merit closer inspection.
First, Meyer-Eppler takes the possibility of synthesizing sounds for granted. This has yet to be tested both in practice and in theory.
In practice we will have to show that every sound used in music can be reproduced synthetically with all its musical characteristics and qualities. Thus, electronic technology would be able to explain systematically what traditional technology, empirical and traditional, has been unable to deliver. Since 1951, this has not been the case, rather the opposite (see section 2.9).
Moreover, Meyer-Eppler’s theory is not based on any tests, which he did not have time to carry out, but on a proposition: the “time and frequency” sound diagram would account for the whole phenomenon of sound. For the acoustician, this is quite true. Is it the same for the musician?
Meyer-Eppler does mention the need to “take the capacity of the human ear into account” before identifying spectral and musical sound. But this vague statement scarcely tells us whether he has in mind a study of sensations—thresholds, sensitivity curves—or the perceptions that surface in the musical consciousness. It seems, however, that once the physical makeup of sounds and the workings of that more or less imperfect device, the ear, are understood, there is nothing else really important for him to learn. He does not even claim to give an account of our auditory sensations (as for the concept of perception, this is totally absent from his thinking) but explains them in terms of their material causes, having taken into account the transformations they undergo on the way. More precisely, he considers this explanation as a foregone conclusion and the chain of causes as obvious enough for it to be already possible to preplan music on the basis of physics.
Doubtless not every specialist will go with him all the way. But doubtless none of them would reject his attempt at explanation itself, considered as the height of the scientific mode of discovery. And so we come back to that project, vigorously laid claim to by Fritz Winckel, to whom, incidentally, we owe so many subtle observations and useful cautions:
If we wish to go more deeply into the phenomena we have hinted at, we must above all study the natural laws that govern the production of sounds, and examine their physical or physiological action on the sense of hearing and the brain. Perhaps these factors will enable us to shed light on the way music influences man. Some may bridle at too scientific an approach to these problems and will content themselves with groundless impressions and metaphysical explanations rather than embarking on a systematic study of the secrets of a natural phenomenon. Of course, sounds can be neither touched nor seen; they nonetheless have a physical reality, since they reveal themselves by a variation in air pressure, mechanical vibrations in the middle ear, oscillations in the fluid of the inner ear, and finally electrical impulses conducted by the nerve fibers to the brain. Sound phenomena are, after all, produced by the vibration of material bodies.4
The findings of Winckel’s research could indeed lead to responses of a very different hue. But we must keep to the text. We find statements of varying significance:
• The production and transmission of sounds, from the vibration of material bodies right to the brain, take place by means of mechanisms arising from natural laws. We have absolutely no reason to doubt this, any more than the importance a study of this may have.
• This research, physical or physiological, must be undertaken before anything else if we wish to “go right into” the phenomena of perception and aesthetic evaluation mentioned earlier by Winckel.
• Its ultimate aim would be to explain how music influences man. That is to say, music, itself arising from physical and physiological processes, is presented as an objective reality and, in relation to the listener, as a cause producing certain effects.
Whereas the first statement is simply an observation and working hypothesis, the other two constitute a stance that must have a motive. We are therefore, in return, entitled to ask two kinds of questions.
1. What conditions must be obeyed for the system of explanation outlined here to be valid within its premises?
2. Since it claims an almost exclusive validity, we would take the liberty of wondering on what he bases this claim and, if it cannot be justified, whether there is another, more justified, approach.
In the Cartesian breakdown in section 5.2. we are, as has been seen, obliged to embrace a whole raft of different disciplines.
So, as far as music is concerned, each of these disciplines should contribute a huge amount of information, and we must, moreover, take particular care to check the “continuity” across disciplines, so that our logical progress is not weakened.
The perilous nature of the undertaking can be seen at once, as can its probably utopian goals, masquerading as logic. We can also see what preconditions it implies, since it claims to trace out a chronological investigatory pathway. How and when will we tackle the phenomenon of music if we must first of all understand the secrets of the workings of the inner ear and also establish a flawless connection between the elementary levels of sensation and the higher levels of perception? What experimental psychologist, what brain surgeon, will come forward to give reliable answers to these questions?
In response to this scientific dream we must bring in other realities that have been the subject of psychological research rather too often neglected by our physicists. Indeed, we can see that they merely refer to “musical sensations,” as if sensations were the prime factor of musical consciousness. Now, sensation is not instantly there in our consciousness; it comes, in general, only from a selection process on the part of perception. There is therefore a “break” in the apparently logical circuit championed by our scientists. We should send them off to read the whole library of research, spanning almost fifty years, by other specialists, just as worthy of respect, on this point.
Even before referring to the work of gestaltists and phenomenologists, we should point out to our physicists that there is a serious risk of error in their seemingly rigorous approach: it is the discontinuity of specialist knowledge that most often rears its ugly head at each new intersection between one discipline and another. We will simply point out, in musical acoustics alone, the ambiguity of the terms used, according to circumstances, to designate both a physical phenomenon and its musical effect.
In fact it must be either one thing or the other: either the science enthusiast that the contemporary musician is must accept the correlations between frequency and pitch, level and intensity, time and duration, spectrum and timbre, and so forth, as established fact, or else, better informed, he must be aware of the care that physicists themselves take to suggest two calibrations, one physical, the other sensory. This could serve as a warning. But doubtless he sees in it further reasons to trust the experts, who, for their part, maintaining out of principle that they are incompetent in music, trust him. So what happens next?
What happens is that extremely rigorous research on the physical level leads to results that cannot be used for music, unless they are inappropriately extended into a field to which they do not apply. This is why most of the response curves of the ear, set up for elementary stimuli, do not apply to complex signals or simultaneous sounds in the context of strictly musical listening, which has nothing in common with the quasi-surgical conditioning of a well-conducted sensory experiment.
We will see below an attempt to explain the more absurd misunderstandings. But since it does not seem possible to fulfill the conditions for an objective scientific approach, we will have to find another way.
In rejecting the above approach to the phenomenon of music, which claims to be scientific because it is based on the physics of sounds, are we rejecting a scientific approach to music?
On the contrary. We are saying that a scientific approach is defined by a method that is adequate for its subject. Suppose we reconsider the “chain” of the phenomena that form a circuit of communication from man to things, and vice versa, in music. At either end we find, on the one hand, the physical universe and, on the other hand, the world of consciousness. There is nothing to say we should not pursue our investigation along parallel lines, using a “pincer movement” and preferably attacking the weakest fronts, which afford the most effective way in. The most modern cybernetic practice teaches us to put all the links in the chain, which form, as it were, a “black box,” “in parentheses,” and to concentrate only on what is going on at the two ends.
If we put stimuli at only one of these ends, there is little chance of our experimenting on music at the other—just as, if we present symphonies only to the ear, there is little chance of our finding the truly experimental level of a psychology of music, at least at the initial stage.
These are, however, the two most common mistakes—the first the prerogative of physicists, the second of some psychologists. If we have clearly understood the error of the former, who impoverish their experimental material too much, we can also begin to see the error of the latter, who take their experimental material from too high a level (melodies, modes, etc.) and thus, despite encouraging statistics, obtain only vague results about aesthetic emotion.
Between these two types of mistake, excess and insufficiency, there should be room for a reasonable experimental program, concerning the object itself. What elementary factors are we listening to in all music? How do we listen? Between sensations, which are only an “unstable,” artificial state of consciousness, and aesthetic emotions, already inaccessible and too complicated, surely there is an experimental field for specifically musical perception, where the stimulus of an external signal and the awareness of a musical signification could be adequately explored?