Everything about singing is organized around the ear; it is the superior regulator. The buck stops there. How does this come about? The ear has a double role, based on two polarities. The vestibule organizes the motor activity involved in singing. This is linked to the act of listening carried out by the cochlea. The process begins with assuming a listening posture to which the singing posture will be added.
But first we need to discuss singing in the context of communication systems. All communication entails a speaker and a listener. When the listener replies, he becomes the speaker and the speaker becomes the listener. As soon as either one of the two decides to enter the dialogue, there is an element of control, allowing that person to monitor himself. Consequently, the speaker is the first listener of his own speech. In fact, upon deciding to speak, his brain activates phonation so that a message can be emitted. Control is activated simultaneously, regulating different parameters such as intensity, timber, articulation, attack and release of sounds, the melody of the phrase, and the choice of words. Every aspect of language is monitored in this way, and with singing there is even greater control. This control resides in the ear, where three pathways insure this function.
In reality, we are describing three circuits, two of which arise in the larynx. The first and best circuit is bone conduction. The second circuit, air conduction, relies on muscles and tendons. It gives poor results, and you must avoid using it. By knowing the various loops, you can understand the mechanics of phonation. The two channels you will be using are the internal bone conduction circuit and the external mouth circuit. Each has a completely separate role and the use of each is entirely distinct. The first controls the voice and the second controls articulation. The second depends on the first. Without bone conduction, there is no emission, even if articulation is good.
The mouth circuit does not permit good control over phonation because of its anatomy. Any sound emitted is always complex. It contains a fundamental tone and an associated gamut of harmonics. Once it is launched into the air, the sound is dispersed and no longer subject to control. The high partials travel in a straight line. This effect is even more pronounced at higher pitches, also called directional. The low frequencies, contained in all sounds that are emitted, expand in a circle, bathing the outside of the ear. So when we hear ourselves we hear a preponderance of lows.
When we listen to our own voice on a recording, we are alway surprised to hear how we sound. While making sounds in a room with good reverberation, the feedback we get allows us to control high and medium frequencies as well as lows. This is an example of cybernetic control. Singing obeys the same laws of regulation and hearing becomes listening.
From The Ear and the Voice by Alfred A. Tomatis (Scarecrow Press, 2005)