Canaries in the coal mine: the pure vowel.
I UNDERSTAND THAT MINERS used to take birds along with them when working underground. Canaries are particularly sensitive to poisonous gasses, which are of ten not detectable by human noses. As long as the birds were singing, everything was fine. If the singing stopped and the bird was not healthy, it was a warning that the miners had better get to the surface immediately!
This is not going to be an article on the singing of birds (although there is a huge literature on this topic). The thesis of this little article is that a distorted vowel is like a "dead canary in the mine" to the observant teacher. It is a warning that the voice is in a compromised condition. Addressing the vowel quality is therefore a critical step in the process of putting the voice right and should not be ignored.
I will avoid here a discussion of the acoustics of the vocal tract; I do not believe that another explanation of formants and partials is necessary in this discussion. Neither is this going to be discourse on the adjustment of vowels that an artist may utilize: vowel modification, aggiustamento, vowel migration, etc. This is a critically important skill that singers must learn to do, although there are the purists who claim that no change in vowel quality is ever needed or allowed. I appreciate how useful this position can be pedagogically, even if the truth of it in absolute terms is not supportable. Most of us realize that some shift of vowel quality is helpful if we are to optimize the resonance characteristics of the vocal tract as we sing throughout the full range of the voice. This idea was codified in the work of Berton Coffin. Dr. Coffin's systematic approach might be a bit bewildering upon first exposure to his famous vowel chart, but the acoustic science behind it is absolutely valid. In summary, slight adjustments to the formants (a formant is a resonance of the vocal tract) are made to maintain an optimal tuning between them and the partials or overtones emitted from the larynx during phonation. Since the partials are constantly changing frequency (because they move every time the pitch changes) the relationship between formants and partials is a constantly changing scenario. The result is a slight modification of the absolute quality of the vowel-call it what you want!
I hear so many singers who do not have a clear concept of vowel quality and who sing distorted vowels even in the range where no systematic adjustment should be necessary. Ifa singer cannot sing an Italianate [a] vowel anywhere in his range, the technique is fundamentally flawed and he will be in a constant state of excessive tension and compromised vocal production. Vowel distortion is a symptom of a poor concept of tonal quality and/or excessive tension in the tongue. This idea is the heart of the concept of the "pure vowel" which was addressed by Tosi in 1723.
Let the scholar be obliged to pronounce the vowels distinctly, that they may be heard for such as they are. Some singers think to pronounce the first, and you hear the second; if the fault is not the master's, it is of those singers who are scarce got out of their first lessons. They study to sing with affectation, as if ashamed to open their mouths. Others, on the contrary stretching theirs too much, confound these two vowels with the fourth, making it impossible to comprehend whether they have said "balla" or "bella," "sesso" or "sasso" "mare" or "more." (1)
What does it mean to pronounce a vowel "distinctly"? Only part of the issue is intelligibility, as Tosi suggested at the end of the quote above. There is much more to it than this. There is an aural definition of vowel quality in the collective consciousness of good teachers and singers everywhere. If it were not so there would be no agreement on what constitutes beautiful singing at all. The existence of a universal defined vowel quality may be "pie in the sky," but is fundamental to what we do as voice teachers, of course this assumes keen discernment in the ear of a voice teacher. We should all be able to agree upon what [a] should sound like in any voice, whether male or female, low or high. If this is so, then we should be able to recognize when it is distorted. This realization should motivate us to help the singer--at any level, correct the distortion. Cornelius Reid said,
The full import of Tosi's advice becomes more forceful when considered in its practical implications. As the direction concerning vowel purity is more specifically given to include the degree of loudness, posture, and exact shading of the vowel, a student has been given an idea, i.e., the creation of a pure, undistorted vowel quality. Fixing his attention upon this idea of quality expressed in terms of vowel, he will, instead of producing a "made" quality artificial to his voice, create a genuine phase of tone color, unaffected and not in any way "put on" This is the initial and most important concept in voice training. (2)
There have been several theories over the years concerning the origin of vowels. You can find sources that say that vowels are formed by the vocal folds. We now know that this is not the case. The vowel is formed primarily by the position of the tongue, although ali the articulators (lips, jaw, soft palate) can influence the vowel color. The position of the tongue in the oral cavity defines the frequencies of the formants of the vocal tract. The frequency of the two lowest formants defines the vowel quality. The la] vowel is produced with the tongue literally pulled back in the throat. We have the idea that [a] is a relaxed vowel with a big open space in the back of the oral cavity. In reality there is a narrow constriction between the back of the tongue and the posterior pharyngeal wall. If there is not, then it no longer sounds like [a]. It requires a certain amount of tension in the tongue to pronounce [a] distinctly. For [i] the tongue is tensed in the front and the primary constriction is between the blade of the tongue and the alveolar ridge--right behind the front teeth. It is classified phonetically as a tense vowel because it takes a lot of effort by the muscles of the tongue to put itself into the right position to produce [i]. Ali the other vowels--the pure ones and the distorted ones are produced by appropriate tension in the tongue. That is what makes this such a difficulty. It is not that tension distorts the vowel, excessive tension distorts the vowel. How do you know if there is excessive tension? The vowel is distorted. It is as simple and as complex as that!
This is the reason that the identity of our target "pure" vowels is so important. It is like putting just the right amount of salt in a recipe. If there is too much you taste the salt, not enough and it loses something that it needs. Just the right amount of salt is required. If vowels do not have just the right color, just the right timbre then there is probably excessive tension in the tongue that is going to be a problem.
Head posture can also influence the quality of a vowel. If the head is tilted (chin against the sternum) the vowel is going to be distorted because the shape of the air space in the vocal tract has been distorted. This particular position usually darkens the vowel and gives the impression of more color or weight in the tone. I of ten see singers with heads lowered. This is to be avoided, even if we see our favorite professional singers doing it. If the head is habitually turned to one side or the other, the vowels will all be distorted. I call this "choir folder syndrome." The posture of the head relative to the body is turned--usually to the left--because we all have spent so many hours holding a folder in the left hand. The head can be too elevated as well--this tends to bring an open, yell-like quality to the vowel. Trying to correct these postural concerns can be a real difficulty, but it must be done if we are to help the singer to a better place.
It is common to find singers who are trying to impose a quality that they believe should be there but isn't by nature. Not to pick on mezzo sopranos, but it is very common to hear them darkening their vowels to make the voice "rich" or "dark" or "big" enough. (Do I need mention that baritones, sopranos, and tenors do this as well?) This is an artificial means of adding weight or color to the voice and there is always a price to pay for it. The distorted vowel is almost always produced by excessive tension in the tongue and it can absolutely take over the whole mechanism and leave it so tied up that it cannot do anything successfully. Flexibility can be compromised, pitch can be affected, and access to the low voice and the high voice is also inhibited. How many times have I personally encountered a soprano who did not have access to her high voice whose head remained tilted down (chin toward sternum) as she tried to sing above the staff. If I can get her to simply lift her head, sing a bright vowel and drop her jaw the voice easily goes through [F.sub.5]. all this could have been avoided if their ears had been trained to sing pure vowel qualities.
If a vowel is distorted it is the result of unnecessary tension somewhere in the vocal tract, most typically tension in the tongue. We can eliminate the excessive tension through specific exercises designed to get the tongue into motion and by focusing on the quality of the vowel. Both of these are legitimate approaches to purifying the vowel. Simply things like repeating [la] quickly on a scale or arpeggio can help free up the front of the tongue. Likewise [ka] or [ga] addresses the back of the tongue. I say that if a muscle is tense, give it something to do! I also use [mnja]--an exaggerated "hummy" glide performed very slowly, relishing every moment of the tongue in motion to release tension in the tongue, of course, at the same time we must create the correct vowel identity in the imagination of the student. If students cannot imagine it, they can not sing it clearly. They cannot imagine it if they have no concept of what it should sound like.
I have always loved the sound of happy canaries!
(1.) Pier Francesco Tosi, Observations on the Florid Song, trans. Mr. Galliard, ed. Michael Pilkington (London: Stainer & Bell, 1987), 6-7.
(2.) Cornelius Reid, Bel Canto: Principles and Practice (New York: Coleman-Ross Co., 1950), 58.
Stephen F. Austin, MM, PhD is associate professor voice and voice pedagogy at North Texas State University. Dr. Austin is a singer, voice teacher, and a trained voice scientist, having earned the PhD in voice science under the direction Dr. Ingo Titze at the University Iowa. He has published original scientific investigations in Journal Voice and has published articles on pedagogy, voice science, and historical pedagogy in Journal Singing, Australian Voice, The Choral Journal, and Classical Singer. He has been primary author and editor the popular "Provenance" column in the Journal Singing since 2005. He is a popular presenter and lecturer and has given presentations and master classes around the world. Dr. Austin serves on the editorial board the Journal Singing, is chair the Voice Science Advisory Committee NATS, and is a member the Scientific Advisory Board The Voice Foundation. He is a successful studio teacher with former students who sing throughout the United States and Europe and hold university faculty appointments around the country. Dr. Austin conducts The Vocal Pedagogy Workshop every summer on the campus the University North Texas where he teaches practical application important scientific concepts and pedagogic tenets the historical Italian School singing for voice teachers and singers.