"Is the ear a natural microphone?"
"Is the ear a natural microphone?" (SON Magazine no. 72, June 1976)
Thirteenth interview of the series Alain Gerber × Alfred Tomatis in SON Magazine. In no. 72, June 1976, Tomatis dismantles the mechanistic conception of hearing born in the 1930s — when collectors were placed on the auditory nerve of dead animals and “microphonic” reactions were observed. He defends the view that the ear is not a passive sensor but an organ of intentional listening endowed with a “predictive” state: in order to speak, one must self-listen to what has not yet been said (the stammerer is deaf to himself because he cannot predict). The left ear listens “as a dilettante” to the global orchestra, the right ear “takes aim” like a marksman. Tomatis identifies two universal switching points (~800 Hz and ~3,000 Hz) and issues the challenge to microphone manufacturers: “it is better to possess a poor ear desirous of listening, than a very good one that will hear nothing”.
“SON” Magazine — no. 72 — June 1976
Is the ear a natural microphone?
Alfred A. TOMATIS
Interview gathered by Alain Gerber
Presentation
In the previous issue of SON Magazine, Professor Tomatis spoke to us about the human voice which, for him, is the most beautiful instrument in the world. This month, he wonders for us about the resemblance between the human ear and the microphone and gives us the fruit of his research.
A seductive but simplistic analogy
Alain Gerber: Professor Tomatis, human industry has developed an instrument of listening which is the microphone. Does there exist a relation between the structure of this apparatus and that of the ear? In other words, can one say, reversing the problem, that the ear is a kind of natural microphone?
Alfred Tomatis: That is a rather complex question you raise, because one can answer it only after having examined a certain number of theories proposed by researchers for more than half a century. The first tendency was, precisely, to represent the ear as a microphone that would integrate the whole message destined for it. It is very tempting to see things in this way. The trouble is that it is also very simplistic! If many scientists have yielded to the temptation, this is due to the fact that the ear, in our civilisation at least, was only late raised to the dignity of an object of study.
Consider that only 400 years ago, it was not even known that there were ossicles in the ear! The discovery of these essential elements was, moreover, only the fruit of chance… In short, in this domain, one has long been content with approximations. It sufficed that a vague resemblance was observed between, on the one hand, the organ furnished with its pavilion and, on the other, a collector, for an ear-microphone analogy to be established.
The 1930s experiment on the dead animal
Add to this that around 1930, more thorough research made it possible to establish that the ear really did respond to stimulation like a microphone. By placing collectors on the auditory nerve, after the labyrinth, it became possible to observe that the ear reacted to speech, for example, exactly in the same way as a microphone. The most curious thing is that this reaction occurred even in an already dead animal. Of course, the natural microphone went out as the cell perished, but one was no less justified in thinking that there existed a structural analogy with the machine, and that, in the living being, the functioning of this microphonic structure was relatively independent of the higher processes. In other words, the ear was a kind of machine, like the microphone itself.
The major oversight: cortex and body
A. G.: And it was a mistake?
A. T.: There again, it is not so simple to answer you. On the one hand, it is exact that the ear acts according to this schema. But on the other, it is false! Let us say, to be more precise, that this way of conceiving things, acceptable in itself, supposes and suggests other theoretical implications which, themselves, are erroneous. What will happen, indeed, if one applies too strictly this mechanistic conception?
By sending a sonic impulse into one ear, one will think to find on the other side a response identical to that which could be measured on a microphone in an acoustics laboratory. Well, one risks being seriously disappointed! For one is forgetting one thing in all this affair, namely that the ear is not only a pavilion and an internal system (of which, it must be said, most researchers have never quite known how it functioned!).
One must extend the notion of ear to the cortex if one wants to understand anything about it. The auditory apparatus comprises a cerebral dimension that it is not possible to ignore. For my part, I go even further. The further I advance in my work, the more I incline to think that it is not only the brain, but the whole body, that comes into play in the process of listening.
A microphone with psychological responses
A. G.: The human ear would therefore be a microphone capable of having… psychological responses?
A. T.: Exactly! It is a microphone that responds not only according to its physical sensitivity to quantitative sonic stimulations, but also and above all according to its affective sensitivity to a qualitative aspect of these stimulations.
A. G.: The formula is a little abstract…
A. T.: Plainly, it means that our ear does not content itself with registering the stimulations of the surrounding sound world. It is also endowed with the power to accept or to refuse them. And if you still find that too abstract, I shall tell you that it attributes to them a value, and that sounds, according to this value, are more or less well received, integrated by the psyche and the personality as a whole.
When these sounds happen to be the words pronounced by the parents at the moment of language acquisition, refusal or acceptance determines in very large measure the degree and rapidity of integration of this language by the subject. You see that there is nothing more concrete and that it is a question of a serious problem. Our “ear”, thus understood, can render us dyslexic, stammering, mute; it can take charge of a neurosis and contribute to its fixation, etc. — all things that are out of all proportion with the possibilities of a microphone.
The microphone and repression
A. G.: The microphone, for its part, is not capable of being deaf because it does not want to hear?
A. T.: That is so. It can record, or it cannot. The ear, for its part, may very well fail to hear messages that are however within reach of the auditory apparatus. In fact, it perceives them, but rejects them before the vigilant consciousness can take hold of them. It is indeed a matter of a mechanism of psychological repression.
This being said, one should not believe that the microphone has no problems of its own to settle. Of course, it does not suffer from disorders of a psychoanalytic order! But all the same. All is not as clear in its functioning as what is inscribed on the notice that accompanies it. On paper, a good microphone is always linear and possesses all the characteristics necessary to satisfy the user. But when one sets it going and looks closely at what happens, one realises that the response is not as uniform as one could have hoped. It seems indeed that there are elective passbands. Thus a microphone, supposedly linear from 0 to 10,000 hertz, presents at certain levels phenomena that one might almost call of resonance.
At these levels, the responses are therefore as good as possible, while at others, it works much less well. In sum, the response of a microphone is more or less good according to physical phenomena more or less known, more or less complex, and also according to the montages that can be made and the feedbacks that may occur.
Elective passbands of the ear
A. G.: And here, one finds again the analogy with the human ear?
A. T.: Indeed! The human ear should be linear, or roughly so, from 16 to 16,000 or 20,000 periods, for a stimulation situated between 40 and 60 decibels. Yet, according to individual aptitudes, learning, the impedance of the place and a thousand other parameters, there are passbands that will be immediately maximal as regards response, while others will be mediocre.
A. G.: Must one then return to the “microphonic” theory of the ear?
A. T.: In fact, no. It is indeed some time since this theory had to be abandoned, because it did not explain the distribution of frequencies on the ear. Let us say that on the plane of physiology, it was much too lacunar.
“Predictive” listening and the stammerer deaf to himself
A. G.: Where do we now stand?
A. T.: Efforts are being made to specify the differences that exist and remain irreducible between the ear and an ordinary microphone. The first of these differences is that the ear is highly selective. Listening to an orchestra, it can, at will, focalise on the flute or the second violin — which the microphone does not know how to do very well (when one uses it, it is very hard to play on the feedback). This focalisation goes so far that one may even speak in certain cases of a “predictive state”.
Let me explain. Take the example of a subject who is about to speak. As I have already had the opportunity to tell you, this subject will be the first to listen to himself. To listen to himself, he is going to have to use the auditory apparatus as a microphonic apparatus of self-listening. Now, to master this process properly, one must in a way self-listen to what has not yet been said: one must self-listen to what the ear is only on the point of hearing. One is in the presence of a veritable phenomenon of premonition, which must absolutely come into play for us to dominate our speech perfectly. So true is this that the stammerer is typically an individual incapable of this premonition. Because he cannot “predict” the system, he is as it were deaf to himself.
The ear that predicts the B flat
A. G.: In other words, this “predictive” state modifies listening…
A. T.: Exactly. Experiments have moreover shown that the ear had a response curve that modified itself according to the sound it “predicted”. Some thirty years ago, phonologists had put forward the hypothesis that, according to a fairly similar schema, the brain “prepared” the larynx. You think of a B flat, the larynx is already preparing to emit it. Perhaps this goes a bit far. But in the case of the ear, it seems beyond doubt that it indeed predicts the B flat in question.
A. G.: Yet it must still have a musical culture!
A. T.: Of course. You know that the ear presents responses at the level of the musculature. Well, these responses have value only if the ear is already educated. It is a phenomenon I have many times had the opportunity to observe in practice. For the muscles of the malleus and of the stapes to be in a position to respond, the association of three sounds is needed. This addition of stimulations must come first so that afterwards the muscles will be capable of responding in the presence of a single sound.
I take the opportunity to say that according to certain recent works, the muscle of the stapes innervated by the facial nerve benefits from a complementary innervation by the same nerve as the tympanum. This nerve is the vagus nerve. And the vagus nerve, as if by chance, is the nerve of affectivity.
The two ears and the marksman
A. G.: But we have not one, but indeed two ears!
A. T.: Yes, and there again, one cannot understand anything if one places the participation of the cerebral cortex in parentheses. In themselves, our two ears are receivers of information built on the same model. But from the moment the brain comes into play, they differentiate themselves according to the tasks confided to them by this higher level. It is said that auditory “bilaterality” serves to localise sounds in space. That is true, but as Cyrano said, “it is a little short…”. There are many other things to say!
A. G.: You spoke of a differentiation…
A. T.: Indeed. Let us take up the example of the orchestra again. The left ear is the one that will hear the global sound. The right ear is the one that will wander among the music stands, snatch a piccolo trait here, a trombone phrase there, etc. The first listens almost “as a dilettante”. The second is much more active; it “takes aim” at particular sounds.
Besides, the comparison with the marksman suggests itself. Do you know that the greatest sharpshooters keep both eyes open to take aim? One sees the landscape, the other focalises on the centre of the target. Just as there is double vision in the marksman, there is double listening in the one who masters his auditory apparatus well. To remain in the domain of comparisons, one might also say that the two ears are like the two hands of a pianist: the right plays the melody, the left takes charge of the accompaniment.
Towards a selective microphone?
A. G.: The essential trait of human listening appears once again as its power of selection. Could one not imagine a microphone endowed too with selectivity?
A. T.: We are not capable of it for the moment. However, I do not despair that one day one will manage to fabricate an electro-acoustic apparatus whose characteristics come close to those of the ear.
What happens, indeed, when the ear “decides” to hear? It opens what one might call gates. Now, gates exist in the domain of electronics too. The Electronic Ear, for example, supposes a whole system of electronic “gates”, thanks to which we can constrain a human ear to listen to something it does not want to hear. This apparatus enables us to have a more precise idea of what a selective microphone might be.
I do not want to enter into too much detail, but we have already been able to establish a certain number of non-negligible data. I have observed, among other things, that the point of rotation between welcoming, accepting listening and rejecting listening, between relaxed listening and focused listening, lies more or less always at the same level.
A. G.: That is to say?
A. T.: Around 800 Hz. There is there a kind of switching point, in all likelihood the same for all men in all regions of the globe, as well as for animals (this is what I am in the process of verifying). A second hindrance is observable around 3,000 Hz. Why? I do not know. I simply observe. What is certain is that it is not a question of a cut in audition with the upper parts of the spectrum, since a subject who hears sounds prepared at 6,000 or 8,000 Hz integrates very poorly too the sounds prepared at 3,000 Hz.
The challenge to microphone manufacturers
The fact remains that if the microphone manufacturers wish to advance, and to achieve a true technological revolution in their domain, they must begin by studying more closely the way in which the human ear functions. There is here a passionate field to clear. If one seriously tackles the problem — and we are here to communicate the results of our work to all those whom the question interests — one will certainly arrive one day at developing a machine that “knows” how to take aim electively at certain sounds.
You will have understood that it is not a question of improving existing equipment, of going further in the same direction, but indeed of changing point of view, of making the system function in another perspective. It is better to possess a poor ear desirous of listening, than a very good one that will hear nothing.
Place of this interview in the series
This interview is the thirteenth of a series of fifteen. For the complete contents, see the mother-article of the series.
Source: Alain Gerber, “Is the ear a natural microphone? — Alfred A. Tomatis”, SON Magazine no. 72, Paris, June 1976. Digitisation: Christophe Besson, June 2010.