Read an Excerpt
THE MYSTERY OF THE SEVEN VOWELS
In Theory and Practice
By Joscelyn Godwin
Phanes PressCopyright © 1991 Joscelyn Godwin
All rights reserved.
Mapping the Vowels
Few things in our daily lives are more mysterious than the vowels; few things more essential than these forms through which we shape our speech. As close to us as our own breath, almost as intimate as our thoughts, the vowels are hidden by their very proximity, like the object one searches for while clutching it in one's hand. Yet once they are noticed, paths open up on every side, leading to unsuspected revelations of sound, sense, and symbolism. This small book, which I believe is the first one on the subject in English, is intended to point out some of those paths.
Without the vowels, we would be in a sorry state, having to talk to one another in hums and clicks, hisses and groans, as the beasts do: a possibility, but not an alluring one. A ... E ... I ... O ... U ... As we listen to the familiar sounds, which the inner ear of our imagination is even now forming as we read them, we may wonder what it is that actually distinguishes one vowel from another. It is not in the air coming from the lungs, which is the prime matter of our speech. Nor is it to be found in the vocal chords that flank the air-passage and vibrate with the current of breath to sound high notes or low. It is in the mouth that tone is transmuted into word, music into meaning. This is the vestibule of the body, open on one side to the world, and on the other to those dark chambers which so few of us understand or are even aware of. The mouth has a dual role, and its two functions mirror each other. First, it is the place where food is worked on by teeth and tongue and converted into fit nourishment for the body. Second, it is the place where the raw sounds of our vocal chords are worked into language, fit nourishment for the world of ideas that surrounds us.
Though the English alphabet names only five or six vowels, dictionaries give phonetic symbols for about twenty, and experts distinguish many dozens. Theoretically their number is limitless, since vowels merge imperceptibly into one another just as musical pitches do. But although pitch is an infinite field, musical convention requires it to be divided into discrete points—in Western music, to twelve different notes per octave—in order to play its particular games with them. The game of language likewise selects only a few vowel-shades as the currency of its conventions: a different currency in each country, or even in each class and district. Then it represents these by an even more limited number of symbols.
Consequently, the written vowels of any language are only a poor approximation of the spoken ones; no symbolic scheme is equal to the subtle differences so obvious in speech. For example, the initial A of the word "actually" becomes a U in Northern English (uctually); a short E in the "Oxford" accent (ectually); a short I to the stage Frenchman (ictually) and an O to the Swede (octually); and in the Southern drawl of the United States, a froglike noise in the back of the throat. When the pioneer of speech studies, Sir Richard Paget, compared the fourteen vowels of his own "Southern English" or "Public School" pronunciation to a set of thirteen vowels pronounced by a French speaker, he found only one (tout or who) that was identical in both lists, though both languages are content to use the same five or six symbols.
Confronted with the elusive and flexible nature of this phenomenon, various researchers have tried to set out the vowels in some logical order, or to map them in more than one dimension. Certain extreme positions of the mouth then serve as the coordinates, or the boundaries. One such limit is the widest possible cavity, with the widest possible aperture: the AH as performed for doctors and dentists. Another is the widest inside cavity coupled with the smallest aperture: a pinhole OO so exaggerated that no one would use it in ordinary speech. The tightly-squeezed EE is a third limit, with the tongue at the roof of the mouth and the lips forming a thin smile.
Having established these three limits, one can explore the wide range of intermediate vowel-colors for which our alphabet furnishes only an approximate notation. Beginning with the AH of calm, one can either move the tongue forward and the lips into a grin until the point of eat is reached; or else one can constrict the lips while keeping the tongue down, eventually reaching the third coordinate of hoot. To avoid ambiguity, I follow the principle of showing every vowel in the context of a word.
These are a fairly comprehensive set of fourteen English vowels, arranged as they often are in pyramid shape. But in order to complete the base, one has to call on foreign languages, such as French:
si fille tu deux oeuf mot mou
or German, where the characteristic sounds of über and möchte are also present.
This vowel-triangle, adequate as it may be for elementary discussion, takes no account of nasality, the factor that distinguishes sofa and not from the French un and on, and which marks much American diction. The degree to which the nasal cavity is allowed to resonate, by opening the soft palate, might be added as an extra dimension to make the triangle into a tetrahedron. Nor does throat-resonance, so audible in the Scandinavian tongues, appear on it. Finally, the whole concept of a map of sustained vowel-sounds would make nonsense to speakers of a tonal language such as Chinese, where the vowels are always in motion. This much is given as preface, lest the reader take too seriously the artificial limits imposed by five- or seven-voweled alphabets, or by the speech habits of Western cultures. Nevertheless, when we turn from how the vowels are formed to how they are received by the ear, we make an acoustical, even a musical discovery of some significance.CHAPTER 2
The physicists tell us that what strikes the ear of the listener to speech or song is a series of pressure-changes in the air, arranged in complicated wave-patterns. These patterns record the irregular "noises" of the consonants and the regular "tones" of the intervening vowels. Even in speech, every vowel possesses a definite musical pitch (or a glissando from one pitch to another), as one can prove by gradually slowing down one's speaking until every vowel is prolonged. But it cannot be these fundamental pitches that govern or are governed by the vowels. If it were, a singer could not articulate the whole range of vowels on practically any pitch, singing a scale to calm, to hoot, to me, etc. The differences must reside somewhere else.
The only possible place for the acoustical definition of the vowels, if the fundamentals are excluded, is the harmonics. One of the first things learnt in acoustics, or in speculative music, is that a tone carries with it a spectrum of more or less audible harmonics, which occupy certain invariant intervals above it. The sequence of intervals is dictated by the whole-number series, either applied as multipliers to the fundamental vibration-count, or as fractions to the fundamental string-length. For example, the first sixteen harmonics of low bass C, produced by a four-foot string vibrating 64 times a second, are as follows:
It is the harmonics that characterize the tone qualities of instruments, and account for the fact that this low C would sound perceptibly different on a cello, a bass clarinet, a French horn, or an organ diapason. The peculiar tone-quality of the clarinet, for example, is due to the fact that it sounds only the odd-numbered harmonics. The flute's pure tone comes from having nearly all of its energy in the fundamental. The soft notes of the French horn have a weak fundamental and a strong second harmonic. One hears the tone-quality of an instrument without being consciously aware of its acoustical cause. Likewise, in hearing vowels, one's ear is actually registering harmonics, of whose existence most people are altogether unconscious.
It used to be thought that the parallel between instrumental timbre and vowels could be taken further, namely that similar to an instrument, each vowel had a characteristic arrangement of harmonics, no matter what pitch it was sung on. Thus perhaps hoot is always fluty, heat always stringlike in tone. The debate lasted until the early part of this century, when the acoustician Dayton Clarence Miller settled the matter in favor of the alternative theory. Miller confirmed, with the help of vowel-photography, what Sir Charles Wheatstone had proposed in 1837 and Hermann Helmholtz had also deduced on the basis of his experiments with resonators: that a vowel does not have a constant harmonic spectrum, but rather certain "formants" or resonant pitch-areas that favor any harmonic that happens to fall within them. Paget had already discovered this for himself in the 1880s, as an undergraduate at Magdalen College, Oxford. The college chaplain's voice, he writes, "was melodious and resonant, and when he intoned the services in Chapel the harmonics of the note on which he was intoning varied with each change of vowel and performed bugle-calls of an almost militant character."
Vowels depend on resonance. If one sings up the scale into a large bottle or jar, or in a tiled bathroom, there will come a particular note at which the vessel resonates and amplifies the sound. The Roman architectural writer Vitruvius tells (De Architectura V) that in the ancient theaters, bronze jars tuned to a wide range of pitches were placed around the stage and auditorium in order to amplify the voices by just such sympathetic resonance. In an analogous way, the cavities in the head and throat form a series of resonators. Those of the throat and nose can be opened or shut at will, while the mouth is subject to the finest adjustment, often being divided by the tongue into two regions with different resonances. Any given position of the cavities will favor a particular set of pitches. With his usual gift for memorable imagery, Paget wrote that "Considered as a musical instrument, the human voice is really a little orchestra of wind instruments—a reed (of the oboe type) and two, three, or four muted whistles (of the ocarina type). In whispered speech the reed does not play; in voiced speech the reed plays through the ocarinas. In the case of nasal vowels all four ocarinas are used—in other cases two only are essential."
Since these cavities are rather small, their resonances are on the whole high-pitched, even higher than the range of a soprano voice. So, unlike the experiment in which one sings into a large jar, it is not the fundamental tones of the vocal chords that are enhanced, but certain of their harmonics. Whatever harmonics are closest to a given resonance-area will be strengthened to the extent of sounding several times louder than the fundamental itself. For instance, Dayton Miller found that the chief resonances of the vowels ma, maw, mow, and moo amplify a single harmonic so that it contains as much as 90 percent of the total energy of the sound.
So far, there is general agreement that it is the resonant areas that give the vowels their individuality. But consensus breaks down as soon as the attempt is made to quantify the formants of each vowel. Miller found only one for the vowels mentioned above; Paget found two different regions for all the vowels; the Encyclopedia Britannica currently gives three. There is further disagreement on the exact placement of the resonance areas. Early reports diverged partly because of the different languages analyzed (English, German, and Dutch), and partly because of the various methods used: hollow resonators, tuning forks, or simply listening intently to one's own or another's voice. Some were based on whispering alone. Even after 1950, with the application of phonographic methods and spectograms, there was still only a broad statistical correlation between results as published in scientific reference works. Only now is it becoming possible, with the aid of computers, to analyze the acoustics of cavities as complex as those of the speaking apparatus, and to define their resonances. Quantitative analysis of the vowels will always founder on the fact that every voice is unique. One need only consider how readily one can recognize familiar voices from a single word, a cough, or even from the intake of breath, to realize how individual is every person's resonance system or "voice print," and how finely tuned one's ear must be in order to distinguish between them.
One can experience the connection of vowels with harmonics through a simple experiment. The best vowels for this purpose are those with low resonances, who and no. Beginning at the bottom of one's voice range, sing slowly up a scale, listening for the emergence of the upper harmonics. If the partials are not readily audible, try singing a note, then without changing the position of the mouth, send the air through it without the vocal chords, as if whistling. Listen carefully to the pitch of the whistle-tone: it is approximately that of the harmonic to be listened for. Slight adjustments of the tongue and lips enable one to focus the resonance so as to amplify the favored partial. Here are the results of my own experiment, first conducted not in the helpful acoustics of a chapel, but in the course of a long solitary car journey. The notes in the bass clef are the fundamental tones, those in the treble clef the most prominent and audible harmonics. One can see that each vowel has a resonant range of about a major third.
Some people have gone so far as to execute two-part inventions with fundamentals and harmonics—a vocal polyphony exploited to marvelous effect in recent years by David Hykes and his Harmonic Choir. But this is more practicable for male than for female voices. For the higher one sings, the choice of harmonics to occur in the resonant area becomes more and more limited, until it is only the fundamental itself that can be amplified by the primary vowel-resonance. This probably accounts for the "hooty" quality of some voices, especially those of boy sopranos, in the notes toward the top of the treble staff; also for the fact that a coloratura soprano's words are virtually incomprehensible. Women and boys typically speak well below the range in which Western composers require them to sing, thereby availing themselves of a richer spectrum of harmonics.
The findings of acoustics compel one to admit the surprising fact that every time one hears language, whether sung or spoken, one is unconsciously perceiving an intricate melody of high harmonics, and that this is the very thing that carries meaning and enables us to understand one another. Whether we are aware of it or not, this proves that all humans are innately and actively musical, constantly picking out the softest and most transient of tones and discriminating between their nicest shades. This makes nonsense, incidentally, of the concept of "tone-deafness," in the commonly accepted sense of being unable to distinguish one pitch from another. If anyone really lacked this capacity, they could never understand a word anyone said. The so-called tone-deaf are usually people who have been less quick at matching pitches than the majority of children, and have been silenced by school-teachers.
Another conclusion to be drawn from the unconscious hearing of vowel-resonances is that we must have an inborn familiarity with the harmonic series. Somewhere in our perceptive mechanism, the harmonics of vocal tones are being analyzed and presented to our consciousness as vowels. The intervals of the harmonic series must be deeply imbued in our unconscious. This is surely the explanation of why people all round the world respond intuitively to the basic harmonic intervals, and why they are at the basis of every musical system. Some people, on first learning about the harmonic series and in hearing it played, feel an instant kinship with this hierarchy of pure intervals. It is, after all, our most direct perception of the numbers which underlie not just the acoustical world, but the whole physical universe.
Excerpted from THE MYSTERY OF THE SEVEN VOWELS by Joscelyn Godwin. Copyright © 1991 Joscelyn Godwin. Excerpted by permission of Phanes Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.