A Right to Sing


A Right to Sing

Essays on Singing

By Joseph Shore



Table of contents




Vocal mechanics:

  1. What do you mean lower my larynx?
  2. Is All 2800 created equal?
  3. The Illusion of placement





These essays in singing were written while I was a Professor of Voice at The University of British Columbia. They are written in a particularly polemical style because I was trying to engage the minds of the young students who had never studied any of the subjects. It may seem therefore that I debate a great deal with other teachers and voice scientists. This style belies the actuality that I have far more in common with these men than appears. Also, in the years since these lectures I have changed my opinions on several issues herein discussed. I seemed to debate quite a bit with Richard Miller on issues of larynx height and breathing. I now consider that Miller and I are very close on these issues. In the system know as “Appoggio” which we both teach, and to which I am indebted to him for much material, the proper breath is all that is “consciously” attended to in order to secure laryngeal lowering. The sternothyroid muscles, which Sundberg so properly identified as the muscles which lower the larynx, are cued to contract by the tracheal pull of the breath. It is certainly true that conscious attempts to lower the larynx are contrary to Appoggio and usually result in constriction. This being said, the information about laryngeal lowering and the production of the singer’s formant is totally accurate. Miller agrees that the breath for Appoggio results in a larynx that is “relatively low” and stable. His earlier writing which equated the low larynx with the depressed larynx has reference to a German technique in which laryngeal lowering is both excessive and attended to by conscious manipulation. We are both in agreement that such a technique represents a misunderstanding of classic Appoggio utilized by the International School of Singing.


Essentially, these lectures introduce the subject of voice science by taking the beginning student through Johan Sundberg’s book, The Science of the Singing Voice (Northern Illinois University Press).


After some years of reflection I now have to consider the work of Fredric Husler to be bogus in many regards. I had great hopes that Husler’s work might offer insight into proper technique and indeed there is some truth in his writings. But his system as a whole is a misunderstanding of the anatomical and acoustical characteristics of good singing. I have therefore highly edited “Huslerian” comments in this edition of these lectures.


It is my hope that these essays on singing will spur the young singer on to investigate the basic issues of voice science and leave behind the subjective vagaries of simple empiricism. This does not mean, however, that a typical singing lesson is devoted to the discussion of acoustics and muscular nomenclature. While the teacher must know this, the knowledge must be digested and conveyed to the student through a variety of means: demonstration, images, scales, and biofeedback instruments. Vennard’s students have told me that he himself utilized a rather simple, “folksy” method of teaching, something I have also tried to emulate.



These first three long lectures are all about vocal technique, the mechanics of singing. But that does not mean that I view the event of singing as coldly as that sounds; far from it indeed. To me my faith in God, His love for all people and singing are linked together. I love my singers and I love the act of singing. When you get past the scientific lectures of singing you will get to the “heart” of the matter, and it is the heart that matters. When I teach my students I am not just a walking textbook of voice science. I love them. I feel honored and tremendously blessed to be a part of their lives in such an intimate way. How I love my singers. I have had feast and famine as a voice teacher. I have had so many students in my studio that I did nothing but teach six days a week, from early morning until late in the evening, and then I have had the unfortunate circumstance of having no students at all.


My voice studio has always enjoyed having singers in all types of music. I do not consider voice a “classical” thing only. Some of the students who have touched my life the deepest have been non-classical; Gospel, Christian Contemporary, Jazz, Pop, they have all enriched my life and made me a better person and teacher.


Everyone has a right to sing! An acquaintance told me this story. He went to elementary school in the Canadian prairies. It was a one room school house with one teacher who taught everything. She taught some things better than others, of course, and tried to bluff her way through some areas. When it came time for the music section, she devised a plan to help her get through the time. She soon noticed that some children could make a rather pleasing sound and others had a lot of trouble. Since she didn’t really know how to work with children’s voices, she hit upon the idea of dividing the class into two sections for music: the “singers,” and “the listeners.” All of the children needing some help with their singing were classed as “listeners.” It was the “listeners” job to sit there in class and simply “listen” to the “singers.” “Unfortunately” sometimes a “listener” would get carried away with the pretty song the others were singing and a wisp of sound would escape from his lips, almost unnoticed; “almost unnoticed,” because the teacher noticed. When this would happen the teacher would stop the singing abruptly, shake her hand and her head and say, “Now now, some of the listeners are singing.”  Some of the “listeners” were so scarred from the experience that they could not bring themselves to allow another sound to come out of their mouths for their entire lives, not in church, not at a birthday party, not even in the shower. Let me explain this one thing. Everybody has the right to sing! I am happy to teach beginners as well as seasoned professionals. I am happy to teach people with only a wisp of talent and I am happy to teach the next great opera singer. I love singers and I have the honor of sharing their lives in this most wonderful and intimate way.


Joseph Shore,

Indianapolis, 2002






Lectures on Larynx posture and Vowel Formation

By Joseph Shore

Some student is always asking me this question. It seems to be a highly charged personal issue with many singers. Especially since information about the larynx and the vocal tract is relatively new, anecdotal, empirical knowledge has been handed down from teacher to student, often in a confusing manner. Dr. Robert Sataloff wrote in SCIENTIFIC AMERICAN recently, “Twenty years ago the human voice was a mystery. Little was known about how it works or how to care for it” (“The Human Voice,” SCIENTIFIC AMERICAN, Dec. 1992). No mechanism can be studied fully without a working model and no model existed for the human vocal instrument until recently. In 1960 Gunnar Fant discovered that the vocal tract worked like an acoustical resonating tube. Ten years later, Sundberg expanded on that discovery and clarified larynx posture according to the formants of the vocal tract.


There are fewer issues in vocal pedagogy more debated than larynx posture. The acoustical  research of Gunnar Fant and Johan Sundberg have finally given us some objective information to use in these ancient disputes. It is that information that I wish to use here as a sounding-board for further commentary. But first let us look at sound itself.

Sundberg says this about sound:

Regarded from a purely physical point of view, sound consists of microscopic and quick variations of the air pressure. In order to be perceived as a sound by human hearing, these variations must be more rapid than 2U per second, or, in other words, the frequency must be above 20 Hz and less
rapid than 20,000 per second; that is, the frequency must be lower than 20,000 Hz. When we grow older and lose some hearing ability, the upper limit, in particular, may drop considerably. Furthermore, these air pressure
variations must neither be too small (because then we do not hear anything at all) nor too large (because then we do not hear as much as we feel pain in our ears).

Provided that the air pressure is forced to vary in an adequate way, we can hear such variations as sound. Consequently, the voice organ generates air pressure variations. It is not too difficult to imagine why this muse be so. Simplifying a bit, the process can be described in the following way. Every time the vocal folds separate and allow a small amount of air to enter into the larynx tube, the air pressure in that cavity rises somewhat. This increase in the air pressure then rapidly propagates upward in the vocal tract, so that the air pressure in the larynx tube drops again. When the vocal folds separate the next time and pass the next air pulse into the larynx tube, it again rises. The result then is an air pressure that varies in synchrony with the vocal fold vibrations. This is equivalent to the generation of sound, as was just mentioned. This primary sound generated by the vocal fold vibrations is called .the voice source. By and large, the voice source is similar for all voiced sounds produced at the same frequency and intensity of phonation. As there may be huge differences between different voiced sounds, which are equal in pitch and loudness, we conclude that what we hear is not the sound coming directly from the vocal fold oscillator. Something very essential happens to the sound on its way from the glottis to the lip opening.


It was mentioned that the vocal tract is a resonator. This implies that its ability to transfer sound is strongly dependent on the frequency of the sound to be transferred. Tones with frequencies equal to the formant frequencies arc most favored; they arc carried without any problems, or are even helped. There are four or five formants of significance in the human vocal tract. If the glottal oscillator emits a tone having a frequency equal to a formant frequency, this tone will be radiated with a much greater amplitude than other tones whose frequencies do not match a formant frequency.

During phonation, our glottal sound generator, that is, the vibrating vocal folds, does not give rise to one single cone. Instead, an entire family or spectrum of tones is generated. The lowest (one in a spectrum is called the fundamental and the other tones arc called overtones.. The fundamental plus these overtones arc called partials.. All of them have different frequencies, otherwise they would not be different tones. Their frequencies form a harmonic series. This means that partial number N has a frequency N times the frequency of the lowest partial, which is the fundamental. The second partial has a frequency twice that of the fundamental, the third partial has a frequency three times that of the fundamental, and so on. The frequencies of the partials arc simply integer multiples of the frequency of the fundamental.                           


Thus, the glottal oscillator delivers an entire bouquet of harmonic partials to the vocal tract to be forwarded to its open end, which is the lip opening. The vocal tract treats these partials in various ways because they have different frequencies. The partials lying closest to a formant frequency are helped on their way our, so that they arc stronger in the sound radiated from the lip opening than other partials lying further away from a formant frequency.





The VOICE ORGAN is composed functionally of a power supply, an oscillator and a resonator. The airstream from the lungs is periodically interrupted by the vibrating vocal folds. The resulting sound, the voice source, has a spectrum,  containing a large number of harmonic partials, the amplitude of which decreases uniformly with frequency. The air column within the vocal tract has characteristic modes of vibration, or resonances, called formants (.A B. C)
As the voice source moves through the vocal tract each partial is attenuated in proportion to its distance from formant nearest it in frequency. The formant frequencies thus appear as peaks in the spectrum of the sound radiated from the lips; the peaks establish particular vowel sound.


Here is a graph of the vocal organ: the lungs as the power supply, the airstream, the vocal folds as the oscillator producing the sound stream, resonating in the vocal tract, emerging from the mouth as the modified sound or “radiated spectrum.” This graph is for the speaking voice. The singing voice is modified differently as you would expect.


So, the vocal tract is the resonating tube for the voice, the vocal tract being the tube created by the combined chambers of the pharynx  and the oral cavity. Before we get into this, I hear some old voices saying: “Wait. That’s not all! What about the nose and the sinus cavities.  They are resonators too and contribute to the overall resonance by sympathetic resonation. And what about the chest!! After all we have chest tones!”  Vennard’s studies and his recounting of other scientific studies in this area are now standard. Vennard says about the chest as a resonator:

“One often hears the expression ‘chest resonance’, and as a relic of a former day of pedagogy it may be tolerated, but we must not allow ourselves to include the chest seriously in our list (of resonators).”


“Let us not forget that me resonators are cavities of air, and while there is air in the lungs it certainly is not free to vibrate as an integral mass. The chest is not a resonator because it is not a cavity. The thorax is filled with a soft, spongy material similar to that used artificially for insulation. It would tend to absorb, not augment.                                                       


“If the chest were a resonator it would be constantly changing its pitch, because even though the ribs may remain fixed the volume is decreasing because of the ascent of the diaphragm. The capacity of the thorax cannot remain constant during exhalation, and there cannot be phonation except with steadily controlled exhalation.  To refer to the chest as a ”sounding-board” is even worse. No instrument maker would use a cage for a sounding-board, and even supposing he did, he certainly would not fill the spaces and cover the framework with a soft material like flesh. If the chest were a sounding-board, singers would have more volume if they exposed their bodies. No instrument maker would cover his sounding-board with an insulator, and yet we all know that Siegfried sings just as loudly wrapped in a bear-skin as he does in any other circumstances.


“The nasal passages are adapted to the filtering and warming (or in rare cases, for cooling; incoming air) so that it will be cleansed of dust particles, etc., and will be of such a temperature that it will not injure the lungs. The surface is irregular, having fins, or turbinates in the sides like the fins of any mechanical radiator, and it is also fleshy and full of blood vessels, for the purpose of exposing the air to as great a body-temperature area as possible. All this results in two things: While the passage is ideally suited to its function, it is a poor means of taking breath quickly, and it is a poor resonator for either improving or building the tone. For these reasons, most singers both inhale and sing through the mouth.


“The quality of tone as resonated by the nose is well known even to the layman. It is a “honky,” muffled sound, which should not be confused with what is sometimes called “nasal twang.” A good illustration of it is heard in the speech of those unfortunates whose palates are defective and cannot shut off the nose from the other resonators. Anyone can make a cruel imitation of such speech simply by allowing his velum to drop so radically that the mouth is largely shut off and the tone must exit through the nostrils. No matter how well the pharynx is distended, the tone cannot emerge unimpaired except by way of the mouth.


“How much of this quality is admissible to singing ? I lean toward the opinion that it should be eliminated entirely; that the closure of the naso-pharynx should be complete. As a demonstration that nasal resonance is of negligible value, I sometimes sing a sustained tone for my students while alternately closing and releasing my nostrils with my fingers. There is practically no change in quality.



“In 1954, Wooldridge attempted to isolate the contribution of the nasal passages to the singing voice by comparing the vowels produced by six professional singers under two conditions: – normal, and with the nasal passages filled with cotton gauze. He was unable to find significant differences between the spectra of the vowels produced under the two conditions, and a jury of expert listeners was unable to distinguish the two conditions by hearing tape recordings. Wooldridge concluded, “The term ‘nasal resonance’ is without validity in describing voice quality in the singing voice,” (p. 39). A repetition of this experiment by five male singers, including myself, confirmed the original findings, (Vennard8).

“In the experiment on nasality to which I have referred (Par. 343) we not only filled our nasal passages with gauze but also had our maxillary sinuses more than half filled with water, (Fig.41). These are the largest of the sinuses, and putting water in them would change their resonant properties noticeably if they were of any importance. Our singing under these conditions was compared with our normal singing by 86 vocal authorities from the United States and 25 from Holland. The Dutch judges were included because Europeans frequently judge American speech to be nasal but they were no more able to detect the difference between the normal and the abnormal singing than were the American listeners. Our conclusion was that neither “nasal resonance” nor “sinus resonance” has validity. To dispose finally of the idea that these tiny air spaces, with their minute openings into the other resonators, could be of any value other than as indicators to the singer himself, let me quote a typical scientific authority, Schaeffer.

It is very unlikely that the paranasal sinuses exert any influence upon vocalization. The ostia of the sinuses are so small and not infrequently encroached upon by neighboring parts that one naturally wonders how the chambers can have any modifying influence on the sound waves. Moreover, the great variations in the size and arrangement of the sinuses would preclude any constancy of influence. The theory that the paranasal sinuses impart resonance to the voice must doubtless be abandoned.”



So, the only resonator of the voice source in the model we are using is the vocal tract itself, which forms a resonating tube with areas of resonance called formants.  This is the case if the velum is raised against the pharyngeal wall to shut off entrance to the nasal passages. (The trachea may indeed have some effect on the sound. This is still being explored, notably by Dr. Ingo Titze. But the trachea is not an adjustable cavity. So we will leave its discussion until later.) The presence of areas of resonance or FORMANTS in the vocal tract means that certain frequencies of the voice stream are amplified and sped on their way to the mouth.  Sundberg shows that the vocal tract, with the larynx in the high, speaking position, has four formants which effectively modify the sound as shown in this slide of the voice organ. You can see the slope is much more gradual in decreasing amplitude of the partials.


So, the four formants in the vocal tract, when the larynx is high, or in the speaking posture, amplify the upper partials only slightly so that the decrease in amplitude or slope of the voice is less. The peaks in the radiated spectrum where the formants have disrupted the even slope are heard by the human ear as vowels.


This slide shows the graph of the spectral energy of the word “WHO” when spoken.  You see that the highest energy is in the lower frequencies and the upper partials have a steep slope downward in energy.  The other slide is the same word, “WHO,” sung by an opera singer with a lowered larynx with correspondingly broadened base of the pharynx.  You will notice the energy peak between 2500 and 3000 Herz.







VOWELS SOUND DIFFERENT in speech and in singing and the difference is visible in their recorded spectra. Here the spectra of the vowel In “who” as spoken and as sung by a male opera singer are compared. What is significantly different about the sung spectrum is the spectral-energy peak that appears in it between about 2,500 and 3,000 hertz. The new peak is called the singing formant.



SINGING FORMANTS ORIGIN {left} and its utility in singing are demonstrated. An extra formant was inserted between the usual third and fourth formants in an experiment with an electronic resonator that behaves like the vocal tract. The new formant increased the amplitude of the partials near it by more than 20 decibels;


Similarly, an extra formant (achieved by lowering the larynx) supplies the high-frequency peak in the spectrum of a sung vowel. The three



Jussi Bjoerlings Voice is shown………….Here


curves (right) show the averaged distribution of energy in the sound of orchestral music {black), of ordinary speech (gray) and of the late tenor Jussi Bjorling singing with an orchestra.

The distribution is very similar for speech and the orchestra at all frequencies; it is the singer’s voice that produces the peak in the colored curve between 2,000 and 3,000 hertz. In that frequency region a singer’s voice is loud     enough, compared with an orchestra’s sound, to be discerned.







The next slide shows a comparison of the spectral energy of ordinary speech (or singing with the larynx in the speaking positions-deep slope, orchestral music s1ightly less steep slope, and the voice of tenor Jussi Bjoerling.  The peak between 2500 and 3000 Hz in Bjoerling’s voice  is most evident.

Sundberg shows that when the larynx is lowered in such a way as to expand the base of the pharynx, the vocal tract may be said to have an “extra” formant between the normal third and fourth of male singers, which he has called “the singer’s formant.”  The slide above shows the artificial creation of an extra formant with an electronic resonator which behaves like the vocal tract.  The new formant increased the amplitude of the partials near it by more than twenty decibels. Sundberg concludes that the lowered larynx, with the expanded base of the pharynx, is a unique feature of serious singing, since without it the resonating tract has only four formants and the spectral energy of the voice is essentially like speech. This is the essential difference between the sound of the popular singer and the opera singer.

Actually the singer’s formant is also present in speech, since it is impossible to create new formants, but it is tuned to such a high frequency as to have little if any effect on the sound. For the sake of convenience we will talk about the singer’s formant as an “extra formant.”

Sundberg explains:

“It can be calculated that if the area of the outlet of the larynx into the pharynx is less than one sixth of the area of the cross-section of the pharynx, then the larynx is acoustically mismatched with the rest of the vocal tract; IT HAS A RESONANCE FREQUENCY OF ITS OWN, LARGELY INDEPENDENT OF THE REMAINDER OF THE TRACT. THE ONE-SIXTH CONDITION IS LIKELY TO BE MET WHEN THE LARYNX IS LOWERED, BECAUSE THE LOWERING TENDS TO EXPAND THE BOTTOM PART OF THE PHARYNX… THE LOWERING OF THE LARYNX, IN OTHER WORDS, SEEMS TO EXPLAIN THE SINGING-FORMANT PEAK.”

Bartholomew reported investigations at Peabody Conservatory that the “ring” of the voice is around 2800 Hz for men and around 3200 Hz for women.  This could be the resonance of the larynx itself for certain women who generate the singers’ formant.

Bartholomew shows that there probably is no 2800 in the falsetto, unless the  falsetto is very strong. This would also fit with Sundberg, since a weak  falsetto is usually sung with a higher larynx and it is doubtful the “one-sixth” condition would be met.  A “stronger” falsetto would have to be sung in a lowered laryngeal posture and the “one-sixth” condition might be met.


“Depressing” the larynx by pushing down on the tongue, can be shown in X-Rays to narrow the pharynx at the base and therefore we would imagine that the larynx would not have its own resonance.   We will talk about the crucial, but sometimes over-looked differences between  the lowered and the depressed larynx in a moment, but right now lets talk about the obvious relationship of resonating tubes to frequency.  Think of organ pipes.  The lower the frequency of the pitch produced, the longer the pipe, and vice-versa. Long resonating tubes have low frequencies and short tubes have higher frequencies.  Females have shorter vocal tracts than males, as one would expect.  If the morphology of the vocal folds is not compatible with that of the vocal tract, one is not likely to become a singer. PERFECTLY MATCHED MORPHOLOGIES PRESENT THE BEST POSSIBILITIES FOR EXTREMELY EFFICIENT MANIPULATION OF THE TRACT!

The average male length of the vocal tract is about seven inches or 17.5 centimeters.  As a tube it would have formant frequencies around 500 Hz for the first, 1500 Hz for the second, 2500 Hz for the third and 3500 for the fourth. The larynx-dependent singer’s formant would be around 2800 Hz. High tenors might have a tract short enough to average much higher formants.   Low basses might have tracts so long their first formants average much lower.  Sundberg quotes research which indicates that the singer’s formant can range anywhere from 2.3K to 3.6K in male voices, depending on the voice category. The size of the larynx might be very important.

Do women have the singer’s formant?  Bartholomew’s older studies indicate that  the female larynx resonates around 3200, although the newer studies Sundberg quotes make this figure arbitrary if not in doubt.  The graph of the female’s spectral energy looks slightly different.  There is not as much amplitude in this “peak” area  as there is around 2800 Hz for the male, due to the fact that the upper formants cluster differently for the female voice.   Larynx resonance may still cause a peak in that area for some women who produce the singer’s formant. 

Sundberg says that studies of sopranos show that it is POSSIBLE for them to sing with a high larynx.  But it has not been shown that the quality of such singing equals low-larynx singing. In Jerome Hines’ book, GREAT SINGERS ON GREAT SINGING, all of the female stars mention directly or indirectly that they sing with a lowered larynx.   Because of the anecdotal nature of the book, references are not always direct, but there is no mistaking the communication that the women as well as the men believe they sing with a lowered larynx. In my own singing I cannot remember working with a leading female singer who did not believe that she sang with a lowered larynx, and since the physical sensations of a lowered larynx are fairly overt it is difficult to believe they were all mistaken. On the other hand, the extent of the lowering may be very slight, especially compared to a male with a large larynx. And it does seem that some light sopranos sing quite well without generating a singer’s formant.

The question is whether there is a spectrum envelop peak in the soprano the way there is in men. Some dramatic sopranos and mezzos seem to have it. Most lyric and coloratura sopranos do not. There certainly needs to be more research done on this. Laryngeal positions too would be important to study. For example, a mezzo might indeed utilize a lower laryngeal position that a high coloratura soprano. As regards the question whether sopranos can sing with a high larynx or a mobile larynx, the jury is still out. One researcher indicated that he had conducted tentative tests in his lab which indicated that premiere sopranos sang with a relatively low and stable larynx throughout the range whereas the ones which utilized a mobile larynx tended to sound like “hysterical chickens.”

The expertise of the singer and the resultant sound quality are factors that need to be weighed.




The Voice Organ 21

Area functions

04      8      12     16
Distance to lip opening (cm)



Vocal tract profiles





Area functions for some vowels describing the shape of the vocal tract in terms of its cross-sectional area as function of the distance from the lip opening. This shape is profiles: the lip and jaw openings, the tongue shape, the velum, and the larynx.(After Fant, I960.)


So what do the formants do? We know they take the crude voice source from the larynx, which is a spectrum, and give it character by the process of vowel formation or enhancement of certain partials. The actual sound coming out of the larynx by itself is undifferentiated. Appleman indicates that the sound of the larynx without the vocal tract is scarcely more than a “buzz.”   In fact, Vennard’s film, “The Vibrating Larynx”, shows us an excised larynx which is made to “sing.” The sound is only a buzz. The positions of the vocal tract then, during coupling of the pharynx and the buccal cavity by tongue shape create the vowel and radiate the sound towards the mouth opening with a character we call timbre, defined by the vowel and particular characteristics of the sound spectrum  as the partial receive their particular enhancement from the formants of the tract.


This slide shows the positions of the vocal tract which create the spoken vowels. The tongue is the major instrument for the coupling of the pharyngeal and buccal cavities.  Notice in the spoken vowels the great variety of pharyngeal space. The pop singer utilizing the larynx in a speaking position has a different resonance-quality for each vowel since the pharyngeal space changes so greatly for each vowel. The serious singer with the larynx in the low posture begins phonation with the base of the pharynx more expanded and can therefore learn to try to sing all the vowels more evenly. The widest pharyngeal space is in /i:/. The teacher may use the pharyngeal space of /i:/ as a reference for the other vowels, thereby giving the essential impression that all the vowels are resonating in an even fashion. Such pharyngeal evenness can never be fully attained, of course, or there would be no vocal tract shaping to create the other vowels, but using the goal as an illusion in teaching can help to take out the constriction in the pharynx during production of the other vowels. Thus the pharyngeal space of /i:/ can be a great helper for shaping the other vowels with a lesser degree of  constriction than the spoken vowels employ. Pant’s X-Ray profiles are of SPOKEN  Russian vowels. Notice also, that the tongue MUST be free to create the coupling necessary for each vowel. The attempts to keep the tongue flat and against the back of the teeth are holdovers from a previous era in which the role of the tongue in vowel creation was misunderstood. (There was a time when vowels were thought to be formed within the larynx and simply amplified by the vocal tract. The tongue was thought of as an “unruly member” which needed to be out of the way.)  For a singer to create the vowels /u:/, and /o:/ with their high arches and still keep the tip of the tongue touching the back of the teeth, may be difficult for some, especially if the tongue itself is short.  Try it. Keep the tip of the tongue against the back of the teeth and try to sing all of the vowels purely with the larynx in its lower position.  You may find, if you have a short tongue, that the coupling is so altered that the vowels are very impure.  Furthermore the “holding” and “forward stretching” of the tongue in this manner may pull upward on the larynx, in some cases so much that it takes OUT the larynx 2800 resonance by cancelling the necessary “one-sixth” condition in the pharynx.

I tend to agree with Fant’s pictures for /u:/ and /o:/ although I see many singers who are capable of keeping the tip of the tongue touching the teeth for all vowels without impurity. Appleman’s tongue positions differ somewhat from Fant’s, especially for /u:/ and /o:/, where Fant shows the tip of the tongue well retracted. Appleman shows almost as high an arch for /u:/ and /o/ as Fant but wants the tongue tip firmly against the teeth. Sundberg studied the formant differences in spoken and sung (low- larynx) vowels, using professional singers in Sweden. He found, as we would expect, that the formants of all sung vowels shift downward towards their darker sides  due to the lengthening of the vocal tract by the lowered larynx. He also found that his subjects, three basses, used the tongue in a way contrary to the Fant models. Sundberg’s basses used less high arch for /u:/, /o:/ and /i:/, so that the tongue picture for all vowels is closer, but not identical, to Pant’s /a:/. The tongue tip was retracted for /u:/ as in the Fant drawings. Apparently, Sundberg says, singers select a variety of ways to use the tongue during singing with a low larynx.  We should add, however, that it is perfectly possible to sing the vowels with the tongue coupling Fant shows!

In 1971, Sundberg and Lindblom developed an articulatory model for tongue coupling, finally giving us a working model to use in correcting students. A close look at the model will disprove many anecdotal teaching references to tongue posture.


Vennard recounts: “An X-Ray study by Hope and Frommhold of singers ranging from beginners to stars of the Berlin Opera shows that it is the beginners who pull the tongue forward.”


In teaching voice in the International School of Voice, the tongue is not consciously manipulated. Of the five articulators (jaw, lips, tongue, velum, larynx), only the jaw and the lips are consciously manipulated. One can move the tongue, of course, since it is used for chewing as well as speaking, but one could not consciously position the tongue during speech even for one sentence! The nice thing is that the five articulators work as a unit to form a particular configuration of the vocal tract. Therefore if the jaw and lips are accurate, and the concept of the vowel is accurate, the tongue , velum, and larynx will be accurate. Tongue exercises almost always end up in constriction and rigidity. The reflexive nature of the system has been inhibited by conscious intervention. In the International School, attention is places on the breath, the concept of the vowel, the jaw and the lips.


Sundberg says, “The closer a partial is to a formant frequency the more its amplitude at the lip opening is increased…The vocal tract is constricted and expanded in many rather complicated ways, and constricting it in one place affects the frequency of all the formants in different ways.  There are, however, three major tools for changing the shape of the tract in such a way that the frequency of a particular formant is shifted in a particular direction.  These tools are the jaw, the body of the tongue, and the tip of the tongue.”  The two lowest formants are the most important for strength and color of tone, while the singer’s formant gives the distinctive “bloom” of the voice.

If the fundamental pitch of the sung tone is not amplified by the first formant, the sound will be less intense in proportion to the frequency distance from the formant. Sundberg shows that when we sing pitches whose fundamentals are higher than our first formant frequency we must raise the frequency of our first formant, otherwise the  sung tone will lack intensity since the fundamental pitch will be higher than the first formant. Since the Jaw is the main way we have of changing the first formant frequency, the procedure is not hard. The wider the jaw space, the higher the frequency of the first formant.   We can shorten the tract by stretching the jaw so that the lower mandible slightly narrows pharyngeal space, or we can draw back the corners of the mouth, as when we smile, to shorten the tract from the front end maximally. That gives us the maximum shortening without raising the larynx. However, if you smile for the whole voice, all of the formant frequencies will be raised, so all of the voice except for the very highest notes will be under-resonated.


Sundberg asked why it was that singers originally desired a low larynx.  He wondered if it developed out of the need to be heard over large orchestras.  The answer to that must be “No,” since the large orchestra is a modern phenomenon. The “Camerata”-those Florentines who began opera and serious vocal training—and their kinsmen, have left us vocal pedagogies and descriptions of singers (e.g. PRACTICAL REFLECTIONS ON FIGURED SINGING, MANCINI, 1777) so that we know they were after a rich sound, if not necessarily the fully resonant sounds of the latter nineteenth century. Those singers of the First Golden Age were not competing with large orchestras. Nevertheless, their desire for quality and beauty of tone does not suggest that they would have tolerated the kind of tones Sundberg describes as “high larynx.”   Sundberg says: “According to experiments with synthesized singing the auditory impressions of an elevation of the larvnx is promoted by a voice source having a weak fundamental, which is typically affiliated with ‘pressed phonation’, as we know. This suggests that a raised larynx is typically associated with a general muscle tension in the voice organ, while a low larynx can be associated with ‘flow phonation (abduction) and a general relaxation of the voice organ.”    Therefore it is virtually impossible NOT to conclude that the quality sought and described by the earliest teachers (the Camerata) is consistent with the lowered larynx. It was developed by singers and singing teachers because of the QUALITY of tone it engenders.  QUALITY OF TONE PRODUCED THE LOW LARYNX.


The questions then, are how do you lower the larynx and how far? Women have a lesser laryngeal drop than men simply because they have shorter vocal tracts and smaller organs. Vennard tells us that in speaking, the larynx is high, the top of the hyoid bone near the third cervical vertebra usually, and the angle of the thyroid does NOT show any forward tilt. The mylohyoid muscles, which form the floor of the mouth, or “the diaphragm of the mandible,” do NOT show significant activity in the raised larynx. Rather most of the work is being done by the swallowing muscles which arise from the palate, and other suprahyoidal muscles (those muscles ABOVE the hyoid bone which pull UP on the larynx).   Most of the suprahyoidal muscles are constrictors. Thus, when the larynx is high, the throat constricts. Furthermore, the larynx rises higher with higher pitches during speaking or singing in the high position. Such constrictor tension makes the pharynx an extremely poor resonator.  Since the “one-sixth” condition Sundberg described as necessary for the larynx to have its own resonance cannot be met with the larynx high, the singer’s formant is not in the graph of the voice.
It would be essential to look at the pharyngeal width of such singers for a complete explanation. (Sundberg’s later examination of Wang’s studies showed that his singers did NOT produce a singer’s formant.)

The photos below show X-Rays of Vennard’s throat when the larynx is said to be in the high, low, and depressed positions. Each position is distinct, as shown by the muscular activity recorded on the electromyograms and the sonagrams of resonance patterns.  This is very important since some people claim the low larynx and the depressed larynx are the same thing.  Usually they are teachers who are afraid to teach low larynx for fear students will falsely depress, although in some cases one suspects some bias due to tonal preferences. In such cases they seem to disparage the low larynx by equating it with the depressed larynx.



TECHNIQUES OF SINGING, The depressed larynx and the low larynx are equated in a six-page article devoted to disparaging laryngeal lowering.  He says: “Low-1aryngeal positioning is the logical consequence of techniques which induce sensations of pharyngeal enlargement. As Nagel pointed out, the low position of the larynx follows from the yawn posture.  It remains the favored laryngeal position in the German School because it contributes to the production of sound demanded by the aesthetic ideal of that particular national school.  Further, it induces the kind of placement sensations associated with that ideal…Unless a singer is versed in acoustical principles and has some prior knowledge of laryngeal function, the system appears logical and ‘scientific’.”


At first this sounds like an amazing statement! Surely Vennard and Sundberg qualify as knowledgeable scientific investigators whose work is known by Mr. Miller. The beginning of the yawn and pharyngeal broadening Miller disparages speak precisely to the events—if words mean anything—which Sundberg talks about as NECESSARY FOR THE LARYNX TO HAVE ITS OWN RESONANCE.  INDEED THE EXPANDED BASE OF THE PHARYNX HAS BEEN FOUND TO BE THE SINE QUA NON OF SERIOUS SINGING IN SUNDBERG’S STUDIES.   Professor Miller quotes Sundberg’s conclusions in his THE STRUCTURE OF SINGING, but dismisses them by saying Sundberg has studied too “narrow” a field of subjects (p.152).   Could it be that this is more necessitated by Miller’s aesthetic than an accurate appraisal of Sundberg’s studies? Indeed, in Sundberg’s collection of studies found within THE SCIENCE OF THE SINGING VOICE, a wide selection of studied singers confirms the need for the expanded base of the pharynx.


The seeming contradiction between Sundberg and Miller can be explained.  Sundberg is talking about what the organs must do in order to generate the singer’s formant. Miller is talking about what the singer must consciously attempt. In the system known as Appoggio, which is the trademark of the International School of Singing, the singer consciously attends to the breath and the vowel, not the organs themselves directly. The proper breath lowers the larynx to a position relatively low in the throat by cueing the sternothyroid muscles to contract. To attempt to voluntarily lower the larynx by an exaggerated yawn is as foreign to Appoggio as it may be common within provincial German techniques. When Sundberg speaks of the similarity of the yawn and the pharyngeal broadening necessary to produce the singer’s formant, he is speaking only in terms of measurements, not the singer’s conscious attempts at manipulation. Appoggio will produce the necessary pharyngeal broadening for the singer’s formant without requiring conscious manipulation which almost always results in constriction.

Miller agrees that Appoggio utilizes a larynx “relatively low” and stable throughout the singing range.


Sundberg says that during rising pitch, the area of the pharynx tends to decrease, and the side-wall tissues pile in on one-another. To avoid that, the singer may expand pharyngeal broadening further with the rising pitch. Does this fit the definition of “localized action”? No, such pharyngeal broadening, if it happens, would still be reflexive rather than conscious.


Miller says in discussing “schools of singing”, “In fact it would appear to be nearly a physical impossibility to maintain “low-laryngeal posture and pharyngeal enlargement together with the thoracic posture of the classic Italian-trained singer.”  That conclusion needs examination.


Vennard made a big distinction between the low larynx and the depressed larynx. We will try to look at that distinction. Vennard believed he showed that the low-larynx which expands the base of the pharynx, which gives Sundberg’s “one-sixth” condition, and the singer’s formant, is attained by relaxation of the suprahyoidal muscles. “ApparentIy,” Vennard writes, “no effort “in the sternothyroid “is needed to Keep it there…When the pupil has learned how to sing, not while forcing the larynx down, but while LETTING IT REMAIN DOWN, he will have cured both symptom and the disease. He will have the vocal lips functioning freely, and he will have the best possible resonator for them.” Look at the electromyograms again and you will see that the activation of the sternothyroid muscles is insignificant in Vennard’s X-Rays of the lowered larynx.  In Vennard’s view, the student learns to maintain the requisite relaxation of the suprahyoidals. We will come back to this. But first, what is the experience of great singers in this matter?


Discussions with GREAT singers from a variety of national backgrounds show that they consider themselves to sing with a lowered larynx.


In Jerome Hines’ book, GREAT SINGERS ON GREAT SINGING, I grouped the artists according to their statements concerning breathing, laryngeal positioning, vowel formation, passaggio technique and placement. If the artist did not give clear cues I did not list them.   Using terms or concepts which Mr. Miller gave in his book to cue “national schools” I found that all but one would be grouped within his “German School” for laryngeal positioning. Almost all of them were in the German School for breathing, vowel formation, and passaggio technique. Placement showed as many undecided or eclectic as in the French, German, or Italian School. Since most of these artists are Italian or sing the Italian repertoire—Indeed not one native German singer was interviewed—either the idea of strict national schools is faulty or Mr. Miller’s analysis of them would seem to be incorrect. Indeed Gigli’s teacher, Cotogni, appears to have taught him what Mr. Miller calls German breathing technique, and Melocchi—one of the more famous Italian voice teachers—taught Franco Corelli and Mario Del Monaco what Miller seems to call German laryngeal positioning!  Perhaps the concept of strict national schools is too rigid. Hines’ interviews tend to suggest that all these international singers have much more in common than would be suggested by the idea of national schools. Many used concepts drawn from two or three of Miller’s schools, while their singing is considered by critics, audiences, and other singers to be indicative of a particular repertoire. After all, the art of voice teaching spread from Italy to Germany shortly after the creation of opera. Ideas were passed on and exchanged, and modifications made in a manner we would expect from essentially empirical approaches.

Stockhausen was a pupil of Manuel Garcia II but is associated with the “German School.” (Fischer-Dieskau gives credit to Stockhausen and Garcia for his technique.)  Emma Seller, who taught singing in Heidelberg, was highly influenced by Helmholz’s discoveries and “German” teaching, but also studied the pedagogy found in Italy at the time. The strict idea of national schools simply does not do justice to the great deal of interpenetration of thought. The differences in language account for many of the apparent differences in so-called “national schools.” It may be that Miller is discussing national tendencies in these “national schools.” Many singers may reach international fame and be eclectic in their technique. The more a singer reaches the goals of the International School of Singing, the less he seems to be confined to national tendencies. There is indeed an “International School.”


National tendencies appear more distinctly within more provincial singers. The better a singer becomes, the more he/she approaches an international standard which is less defined by a concept of strict national schools.


Another concept of laryngeal lowering appears to be offered by Appleman, Husler, and Reid. They indicate that the larynx is maintained in any position by a suspension system of the extrinsic laryngeal muscles.  Appleman writes: “Although the movements of the post-dorsum and root of the tongue are responsible for altering the size and shape of the pharyngeal cavity, the suprahyoidal and the infrahyoidal muscles of the throat and neck create a state of muscular suspension through their action as antagonists, which firms the pharyngeal walls and stabilizes the larynx in the phonatory tube at any chosen position…The sensation of finding the proper open throat and laryngeal position in the middle voice is that of the first stage of the yawn… However, the expanded throat sensation must always be accentuated as the pitch is raised.  The singer and the teacher must remember that the suprahyoid and infrahyoid muscles always act as a unit in their antagonism and never as independent muscles.”   Reid reiterates this.


This is not necessarily a “different” concept expressed by these teachers. They are again describing organ function rather than the singer’s attention, which as we have seen should be to the breath and the vowel. If these teachers state or imply that consciously attending to yawning is integral to their systems, then they would indeed

be stating something different, but I do not see that as a necessary conclusion.

An extensive explanation of this antagonistic system comes from Frederick Husler.

Husler’s book SINGING, THE PHYSICAL NATURE OF THE VOCAL ORGAN, is a mixed bag. I have had to revise my opinion of his work. Much of it is just plain wrong. However, even Richard Miller told this writer that he admired “one thing” about Husler, that he really wanted to know how the mechanism worked.  Unfortunately he seemed to have a penchant for backing the “wrong horse.”  He backed Goerttler and believed in the existence of the ary-vocalis, and even thought he had found a way to isolate the muscle; amazing since the muscle doesn’t exist! He back Husson and believed in the neurochronaxic theory of vocal fold vibration; amazing since it is utter hogwash!  On the issue of laryngeal suspension, however, he seems to have been more accurate. Therefore I have left in a portion of his material on this subject.  He details the “elastic scaffolding” in which the larynx is suspended. “A properly functioning singing organ consists of one vast interplay, a widespread cyclical process, in which all parts cooperate to support and help each other. In short, ‘each muscle, as antagonist, regulates the action of another'”(p.24).


 The Suspensory Mechanism (Elastic Scaffolding)!


When singing takes place in the correct physiological manner, the larynx—at the precise moment in which its own activity begins—is strongly inspanned in a net formed by a number of paired muscles. These muscles stand in close reflex connection to each other, jointly forming a separate mechanism, a unit of great significance to the voice. The inspanning process takes place as follows: out of the network in which the larynx is flexibly suspended certain muscles, ‘Elevators’, pull upwards (slightly forward and slightly back as well), while others, ‘Depressors’, simultaneously draw downwards (also somewhat forward and somewhat back). This general counter-play (opposition) brings about the necessary inspanning of the larynx.


The following muscles draw the larynx upwards:

(1) The shield cartilage-tongue bone muscle, thyreo-hyoideus (Fig. 31).

(2) The muscle palato-laryngeus-tensor veli palati (Fig. 32).

(3) The muscle stylo-pharyngeus (Fig. 35)4


There is not much to be found in specialist literature about the action of the palate in singing. But all good singers are strongly aware of their muscles in the region of the palate, especially when producing their most beautiful top notes. (They can also feel the powerful manner in which the larynx is simultaneously drawn downwards: the counter-play between Elevators and Depressors.) That is the reason why singers sometimes practice ‘yawning’.


(It has been said that the pharynx muscle: constrictor-pharyngis superior, is a
serviceable Elevator in singing. This is scarcely credible, however, because it is a definite throat-constricting muscle used in swallowing. The vocal result would be a form of ‘tight’ or ‘squeezed’ tone.)


The muscles that act as opposers to the Elevators by depressing or, to be more
accurate, drawing the larynx downwards are:

(1) The paired chest bone-shield cartilage muscle, sterno-thyreoideus (Fig.
33). To be exact, the pull is forward and down. (This muscle also
renders the necessary opposition to the Stretcher of the vocal folds,

  • K. Goerttler.i3

+ This term was coined by Goerttler to describe the ‘passive tensing’ of the vocal folds through the crico-thyreoideus, ‘which, basically, does no more than tauten its elastic scaffolding’. We have adapted it for its descriptiveness, giving it a broader meaning.

t According to V. E. Negus.32



(2) The crico-pharyngeus muscle that pulls downwards and back. Physiologists call it a ‘powerful muscle’.* It springs from the lateral lower border of both sides of the ring cartilage, encircles the gullet (at the junction of the pharynx and oesophagus) where it anchors the ring cartilage, sometimes known as the ‘base’ cartilage. This arrangement enables the inferior horns of the shield cartilage to move freely on the ring cartilage. Without the help of the crico-pharyngeus, the larynx is easily obstructed by the cramping action of the upper muscles of the tongue and tongue bone which try to compensate for its deficiency. (Fig. 34).

(3) In addition, wind-pipe and oesophagus can exert a certain pull on the larynx; the windpipe draws it down and the gullet (being connected to the Santorini cartilages, and therefore to the pyramids) pulls it back and down,  


      Out of this muscular network, the crico-pharyngeus is, as a rule, the weakest and the least developed (in speaking it is hardly used). It is, nevertheless, one of the most important of the singing muscles. Singers of exceptionally strong physique are well aware of it: ‘Sing from the nape of the neck’, or as Caruso said: ‘Place the voice low
down at the back of the throat’ (meaning that the action of the muscle behind the gullet should actually be felt).                                


      One reason for the great importance of the suspensory mechanism is that it forms a connection (and partly a direct one) between the throat and the trunk. Moreover, if set in action by a strong enough impulse, it will automatically bring into play all the respiratory muscles needed in singing. The proper erection of the vocal organ seems to depend chiefly on this mechanism.


The requisite tensing of the muscles of the throat—Tensors, Stretchers and Closers—is also occasioned by it. In other words, its co-operation first brings the functions of the larynx into the proper condition for singing.           .:  

The various paired muscles of the suspensory mechanism pull on the larynx in four different directions. Each of these directional pulls is able to alter substantially the shape and the degree of tension of the vocal folds, and of the laryngeal cavity above. Each one changes the tonal character of the voice, its possible variations being almost unlimited.                                       

Some of the muscles of the suspensory mechanism, however, are scarcely needed or actively used by modern man in his present way of life so that, if they are not completely stunted (atrophied), they are generally badly innervated and correspondingly impotent. As a rule, it is their predisposition to connect up as a unit, as an individual mechanism, that is seriously impaired if not totally extinguished. Their under-development is often perfectly visible, shown by the chronic retraction of the larynx in the normal person (this is also the reason why he cannot sing). –

Science has come to the conclusion that the action of the suspensory muscles in the production of voice has been ‘somewhat too greatly underestimated’.*

  • V.E.Negus.32


We will go further and assert without the least exaggeration that the suspensory mechanism constitutes a cardinal factor in the formation of the singing voice (though not, of course, of the spoken tone). Were the inspanning, stretching and tensing of the vocal folds through the suspensory mechanism to be entirely lacking—an imaginary case—the result would be, either a weak form of ‘falsetto’, without any possibility of strengthening it and so passing into the full voice, or a raucous ‘chest voice’ with inaccessible high notes, with no possibility of modifying its volume, and the like. The reason for this is that one of the most important links between the two extremes would be missing. At its

 fig. 36. Diagram showing the direc-
tional pulls of the individual suspensory
muscles. (Simplified after V. E. Negus.)




very worst the result would be the ugly noises made by someone who is said to have ‘no voice*. Naturally, every variation exists between the absolute zero point of the non-singer, the physically highly-gifted natural singer, and the ‘made’ singer (i.e., one whose voice has been completely ‘unlocked’, regenerated): the suspensory mechanism is capable, needless to say, of partial functioning in any number of ways and combinations. The importance of this mechanism in singing can be observed time and again in the voice trainer’s practice, simply through the fact that, thanks to the efficient functioning of the suspensory mechanism, vocal folds considered to be fairly seriously damaged by the physician, are still able to produce a far better singing tone than the perfectly healthy ones of an organ which, in this sense, has collapsed upon itself. In this respect the proverbially pillar-like necks of singers endowed with magnificent voices are instructive, displaying, as they do, the presence of highly developed suspensory muscles. Those at the back of the neck, which can be considered peripheral
antagonists, are particularly striking.

* M. Nadoleczny. M


indirect inspanning muscles

An indirect but equally important contribution to this inspanning process is
rendered by a paired muscle that reaches from the tongue bone to the shoulder,
M. omo-hyoideus, and by another running from the tongue bone to the chest bone, M. sterno-hyoideus. These powerful muscle-bands draw the tongue bone downwards and, as the larynx is muscularly attached to it (M. thyreo-hyoideus), it necessarily shares in the downward movement (that these muscles take a strong and active part in the production of a sung tone can be felt quite easily with the fingers, especially the omo-hyoideus). Though voice physiology realizes that the meaning of these suspensory or inspanning muscles ‘still needs precise investigation’,* nonetheless it emphasizes their importance ‘in the formation of high notes’. (Fig. 38).

It is easy to conceive that, if the tongue bone is drawn downwards, it has the
effect at least of relieving awkward tenseness in the upper muscles of the tongue bone.

complications DUE TO deficient functioning OP THE suspensory


On pages 19 and 26 two hypothetical cases were mentioned to show how the
voice can split into definite registers (‘falsetto’ and’chest register’). Such a falling-
apart of the voice is due, basically, to the failure of the inspanning mechanism.

Here are some further examples, showing how the same cause can produce quite
different results:

If the Depressors fail to function, so that the larynx is not properly inspanned
while singing, muscles of the tongue and tongue bone, as well as the large number of swallowing muscles (which through constant, life-long use are always perfectly innervated), take charge to compensate for this deficiency. Without the necessary resistance from below, the swallowing muscles pull the larynx back and up and fix it there, while the tongue does its share by pressing it down and back (for a more detailed description, see illustration and text, taken from an X-ray picture, page 28).
This, more or less, is what gives rise to the ‘constricted’ tone.

The voice has a certain ‘falsetto-head tone’ content, because the specific Stretcher
(crico-thyreoideus) is active, and there is no special difficulty in singing up to a
moderately high range.

The variations in this type of voice equal approximately the number of tongue,
tongue bone and swallowing muscles (the ‘squeezed’, ‘constricted’ voice is one ol

It is extremely difficult to remedy this particular form of vocal distortion, because
it invariably means that the singer’s instincts are equally distorted.

The most usual complication caused by inadequate suspension of the larynx is of a different nature, and is less difficult to cure.

* G. E. Arnold.1                 + G. E. Arnold1 quoting Ammersbach.



Tongue, tongue bone and swallowing muscles are not at fault in this case, but the
Elevator, thyreo-hyoideus (the muscle that suspends the larynx to the tongue bone) which, by becoming independent, so to speak, draws the larynx up too high. The Closers of the vocal folds work to excess, which enables them to keep the glottis closed up to the highest pitch; the Elevator, thyreo-hyoideus, evidently acts as external antagonist: to the Closers (see also the chapter’Falsetto—Head Register’).
The chink of the glottis and, apparently, the vocal folds themselves, are considerably shortened. In this situation, there is not only no co-operation from the Depressors of the larynx, but the necessary active participation from the organ of breathing is also lacking.

The vocal product is the so-called ‘white voice’ the ‘voce bianca’. In a milder
form this kind of voice is heard in large numbers, especially in tenors and coloratura sopranos.

The so-called ‘pressed tone’ is equally attributable to the deficient functioning of
the suspensory mechanism—primarily of the Depressors. The air accumulates
beneath the throat and presses, it upwards. To parry it, the action of the Tensors and Closers is forcibly increased, while the work done by the Stretchers is seriously obstructed.

fig. 40. This illustration has been drawn from an X-ray picture taken during production of a so-called ‘squeezed voice’. The accompanying text reads as ‘follows: The larynx lies exceptionally high. The  horns of the shield cartilage press against the pharyngeal wall. The tongue bone is also tightly pressed against the wall of the pharynx, partly disappearing beneath the jaw. The tongue is strongly drawn downwards and back and the X-ray taken in profile shows no space between the tongue and the posterior wall. The voice channel is completely displaced. The pressure on the larynx is so strong that no appreciable outlet remains between pyramids and epiglottis.’ (Quoted from: Experimentally Phonetic,  Panconcelli-Calzia.)          :



It has been observed that this has the effect of tilting the pyramid cartilages forward* which, in turn, considerably reduces the laryngeal cavity (see
‘Stiff Throat’ in Chapter XIV).


The first step to be taken in all these cases is thoroughly to rouse and mobilize the entire organ of breathing. It eventually co-ordinates with the inspanning muscles and so will free the throat. Those familiar ‘tongue-relaxing exercises’ are useless: they do not strike at the root of the trouble and so cannot eliminate it.*


(*My Note: This is a good statement of the vocal technique of Appoggio. By attending to the breath, all the muscular processes are coordinated. Husler’s narration of these muscular actions should not be interpreted to mean that singers are to try to exercise direct control over or that merely learning about the muscles will help singers to become proficient in their art! )


If the muscles in which the larynx is slung. Fig. 35, are inactive while singing,
then upper tongue bone muscles (see Fig. 39), and perhaps others as well, used
in swallowing, take charge to give the larynx some sort of support. This draws
the larynx up and stiffens it, which inevitably narrows or otherwise contorts the

fig. 31. Elevator. Shield cartilage- tongue  bone muscle (M. thyreo- hyoideus).

fio. 32. Elevator.  Palato-laryngeal
muscle (M. palato-laryngeus) seen from
the back.



fio. 33. Depressors. The paired chest
bone-shield cartilage muscle (M. sterno-

fig. 34. Depressor (M. crico-pharyngeus)



Fie. 35. Suspensory mechanism—Elastic scaffolding, (a) M. Thyreo-hyoideus: elevator.
(b) Muscles of the palate: elevators, (c) M. stylo-pharyngeus: elevator, (‘raises and widens the pharynx’—Quiring), (d) M. sterno-thyreoideus: depressor, (c) M. crico-pharvngeus:

fig. 38. Indirect inspanning muscles; M
omohyoideus,  M. sterno-hyoideus.





It could be that Vennard’s analysis of laryngeal lowering is incomplete. The operation of the suspension system might be concealed by the nature of his study, since we do not have X-Rays and electromyograms of the process of continuing phonation in Vennard’s analysis or EMG’s of all the musculature. Furthermore, we have no auditory impressions of the QUALITY of the tone he produced under these circumstances. The X-Rays and electromyograms found in his book, SINGING, THE MECHANISM AND THE TECHNIC, are based on the lower middle voice, E3 on the bass staff, for only 2.2 seconds.  The general auditory impression of Vennard’s singing voice around the same time as these studies suggests that his lack of sternothyroid action during phonation may have been part of the cause of his vocal wobble and register separation, Vennard’s singing for the film, “Voice Production: The Vibrating Larynx”, shows a poor quality of tone, including a wobble and mobile larynx for the higher tones.  In other words, Vennard’s EMG’s measured his vocal defect.

As Husler describes, the sternothyroids assist the cricothyroids in tilting the thyroid cartilage forward and would therefore show increased action with raising of pitch.  Sundberg also says that the sternothyroid muscle is a prime candidate for the muscle responsible for LOWERING PITCH.  Therefore, it would seem to be erroneous to state that there is no infrahyoidal activation during singing. Close-up videos of great singers show sternothyroid action. Robert Merrill’s appearances on the Firestone Hour show a large thyroid cartilage which seems to be pulled down during phonation, apparently by the sternothyroid muscle.  It appears to lower over 20mm. on inhalation and descend slightly lower for the higher notes.


(Yes it does indeed seem that Vennard was wrong in stating that the  sternothyroids should be relaxed during singing. His mistake is understandable however since in the singer’s proprioception, there should be no direct awareness of their contraction, and it is true that the mylohyoids need to be as relaxed as possible. In analyzing the causes of Vennard’s wobble we would look first at the correctness of the breath, then the vocal tract shaping for the vowel. Vennard was wrong about several aspects of vocal tract shaping, including the consciously grooved tongue. Generally constriction is generated by such techniques as Miller has shown.)

Vennard’s statement that the student must learn to “let it (the larynx) remain down,” is phrased passively, but since the infrahyoidal muscles are essentially INVOLUNTARY muscles they need just such images and concepts to activate them.

Dr. Van Lawrence has something pertinent to say about the activation of the sternothyroid muscles during singing. “In the last several years, we’ve become increasingly aware of the vital importance of the neck strap muscles (sternothyroid, sternohyoid, omohyoid) in high-range voice production—thanks in part to some work done at the Haskins Laboratories. These muscles seem to be necessary for stabilizing the main firm structures of the larynx so that the smaller and more delicate intrinsic laryngeal muscles can function optimally” (VOCAL HEALTH AND SCIENCE, NATS, p.49).

Here we appear to have confirmation of Husler’s assertions. The low-larynx does  stay low as the result of the “well managed breath cycle” (Miller), but reflexively, the strap muscles exert a considerable force downwards.  It seems to be true that the “tracheal pull” from the breath CUES the contraction of the extrinsic muscles which lower and support the larynx.  Coordination between the two systems must be learned and many of us believe that the best system for that is historic Appoggio.

(Yes Merrill’s larynx is quite low but Merrill’s vocal technique was generally  excellent. The size of his larynx and the dimensions of his pharynx are responsible for the long laryngeal drop. Proper inhalation lowers the larynx by a tracheal pull and that tracheal pull reflexively cues the stemothyroid muscles to contract.)


(cf.Sundberg, THE NATS JOURNAL, January/February 1993).   Dr. Van Lawrence cautions singers who may have thyroid gland surgery to tell the surgeon NOT to cut across the straps and re-suture. To do so may cost the singer his/her high notes. Indeed this connection was seen when singers who had thyroid surgery had their straps cut, to allow the surgeon more room to operate. The sternothyroid, and sternohyoid muscles were cut and later sutured back together. The result was that the singers lost some of their high notes due to the loss of some muscular function.

Sundberg’s studies also show that at least some professional singers allow the larynx to descend further down with high note production. This would be impossible without sternothyroid action.*


(Note: Both Van Lawrence and Miller are saying the same thing but from different perspectives. The “well-managed breath cycle” is all that the singer consciously attends to in Appoggio. But the effect of that is to coordinate with the depressor muscles which stabilize the larynx.)


Images, both tonal and extra-tonal, are the primary tools for activating this muscular system. Once activated, however, there are definite sensations of singing with a lowered larynx which may be used as biofeedback.



Reid outlines  exercises for coordinating the balanced use of the suspensory muscles in his book, VOICE, PSYCHE, AND SOMA. At least one of them is mentioned by Vennard as well.  They are: “the staccato, the octave jump with reiterated upper note, arpeggii without tonal pulse, the trill on major thirds, the octave jump with a rapid triad added to the upper note,” and an exercise Reid calls “the trombone slide.”  Many of these I use successfully, especially the octave leaps and the staccato arpeggii in a staccato 1,3,5,8 rising, combined with a legato 9,8,5,3,1 descending.


Coordination of the suspension system of the larynx is not too complicated to learn. It is a part of the international system of  Appoggio.


We should come back and ask at this point, if an efficient usage of the laryngeal suspensory system is truly at odds with what Miller calls “the thoracic posture of the classic, Italian trained singer.” Apparently great singers like Caruso and Corelli have never had any difficulty with it.  However, thoracic breathing can be carried to such an extreme that it is divorced from deep breathing, which Husler says, “the majority of experts nowadays are in favor of” (p.48).  It may be divorced from the co-contracting strategy which Sundberg proposes to maximize tracheal pull. Boris Christoff was trained by Riccardo Stracciari and uses thoracic breathing. Close-up videos also show that this breathing method is carried to the extreme: high shoulders are thrown back at each inspiration. The body is so occupied with this clumsy breathing process that there is no bodily involvement in the character. In fact, one must close the eyes to appreciate the voice! At times, the auditory impressions are that the high breathing interferes with stable laryngeal suspension. In fact, the classic symptoms of CLAVICULAR breathing are unstable laryngeal posture and poor intonation. One can easily see why from Husler’s descriptions of the sternothyroids’ role in working with the cricothyroids, both for pitch changes and laryngeal stability. Because thoracic breathing can so quickly turn into clavicular breathing (the upward heaving shoulders are also observed among many Italian singers) I personally teach deep, diaphragmatic beathing, combined with expansion of the rib cage. I later discovered that this was Sundberg’s co-contracting strategy. The expansion of the rib cage does indeed give the feeling of a broad chest, slightly raised at the sternum. Reid refers to this as “intercostal breathing.”   One begins inspiration with diaphragmatic breathing, then expands the rib cage. To reverse the order, raising the chest first, expanding the rib cage, and giving any left-over attention to the diaphragm, is dangerous. The breath is shallower and clavicular breathing may not be far away. Therefore, in response to Miller’s question, we would have to say that good laryngeal suspension (which creates a relatively low laryngeal posture) is only at odds with inefficient breathing. The breathing is inefficient precisely because it interferes with good laryngeal suspension.

Sundberg’s co-contracting strategy gives a strong tracheal pull which cues the extrinsic muscles to lower the larynx and stabilize it by a suspension system of opposite pulls. The proof of the pudding is not only in Caruso, but in every great singer. All  I have said is but another description of the classic breath in Appoggio  which Miller and I both teach. There is a risk in explaining things so mechanically, as Husler does, to think that there is conscious attention to muscular action in the throat. There is not. It is also quite possible that a provincial German singer who excessively yawns and locally broadens his pharynx, and employs breath damming, will think that he is following Husler’s descriptions to the letter. That is why all of the written word must be demonstrated!

Corelli’s description of his sensations in the upper range is significant, especially since this artist had the greatest high notes since Caruso. He indicated that “You are only conscious that, as you sing higher, the larynx doesn’t rise”.
Electromyograms have been done on students and lesser singers, but these are not sufficient for judgments on crucial issues. We need electromyograms of GREAT SINGERS like Corelli, in addition to those made by students and lesser singers. This would make a good follow-up book by Jerome Hines. If anyone could get great singers to submit to EMG’s, it would be Jerome Hines!

I think the auditory evidence suggests that Franco Corelli’s upper tones were efficiently produced!  Great singers all agree that the larynx must not rise during phonation, which suggests that the infrahyoids are active as antagonists to some elevator stabilizing muscles. (Husler and Quiring suggest palatopharyngeus and stylopharyngeus.)  Great natural singers may not be aware of what the muscles are actually doing. They may only be aware that as they sing higher, the larynx does not rise. Good laryngeal suspension is the product of Appoggio, not foreign to it. Only overly tensed and consciously manipulated strap muscles are at odds with



fig. 39. False elevators. Upper muscles of the tongue bone. (a) M. mylo-hyoideus

(b) M. digastricus. (c) M. stylo-hyoideus. (M. genio-hyoideus is hidden by the


(We will soon discuss the “elevators” in this suspension system. Note here in the slide on the previous page that Husler lists the “false elevators” which should not be utilized within this antagonistic system. The most common “false depressor” used is the tongue.)

Studies of laryngeal postures were conducted by Frommhold and Hoppe using a cross-section of

“outstanding international artists” and students. They found that “international artists were conspicuous without exception for a constant (low) posture over the entire vocal range” while students showed increasing postural change upwards during phonation of higher pitches.

I have commonly observed that the great singers with whom I have performed, frequently tilt their head back slightly for high notes as though they were looking at the first balcony. Indeed, in my own singing, the highest, full  voiced, extension of the baritone voice, G4,A4,B4,C5, may be sung with the head slightly tilted back, at least when singing full voice.  Many voice teachers disparage head tilting, believing it puts the vocal instrument out of alignment.  Certainly any excessive raising of the chin which also causes the larynx to rise would be deleterious. We are not talking

about the sword swallowing position. But teachers must seriously ask why virtually all great singers slightly tilt the head for high notes.



Vennard assumes that all head tilting  causes laryngeal elevation, and calls such action “reaching for the high note.”  Why do great singers do it? We have pictures of Caruso with the head slightly tilted for the tones above the second passaggio. It cannot be simply to sing to the balcony. Sometimes there isn’t one and singers still tilt the head. Certainly the auditory evidence indicates that such famous head tillers as Caruso and Corelli did NOT raise the larynx.


Berton Coffin gives an answer which bears on the question of laryngeal suspension. He says: “…unless the head is tilted back there is not enough room for the depressor muscles (sternothyroid, et. a1)  to make a downward pull on the larynx, which assists in bringing about the tensing of the vocal folds necessary for high notes. Persons who bow their heads have difficulties with high notes because there is not room enough for the depressors  to work and the cavity of the throat gives a pitch which is too low. This can be easily detected by thumping the throat with vocal cords adducted while raising and lowering the head…If more teachers would listen  functionally with their eyes, when hearing operas and  concerts, there would be more understanding and less fear of exploring the techniques of singing.”


It is true that the slightly tilted head gives a higher cavity pitch. It also seems to be true that the suspension svstem is working more efficiently.  BY THE SAME ACTION, the low posture of the larynx is maintained. Therefore, the almost universal observation of slight upward head tilting among great singers tends to strengthen the positions of Appleman, Husler, Reid, Van Lawrence, et a1., regarding infrahyoidal muscle activation within laryngeal suspension. LUCHSINGER AND ARNOLD confirm greater sternothyroid efficiency during slight head tilting (VOICE, SPEECH, LANGUAGE, 1965)! The small head tilt up for high notes is permitted within the noble posture of Appoggio while the sword swallowing position is not. Appoggio as a system is not denying infrahyoidal action for the depressor muscles. It is only denying that conscious manipulation of them is appropriate.


Look at videos of some of the great singers (e.g. Corelli Bjoerling, Del Monaco, Merrill, Warren, Tebaldi, Albanese, Siepi, Hines, Tozzi,  Christoff, Bergonzi, Cossotto, Price, Yeend, et a1.) and you will see the slight upward head tilt for high notes. Indeed, some of them show MORE than a slight tilt. Drawing from the experience of another science, psychology,  the lessons of Maslow and Freud should show us that the science of voice must be derived from studies of the “peak experience” of singing, rather than problematic singing. Professional singers have been studied for some things, but they are usually entrance level professionals or singers of less-than-great stature. Stanley studied Melchior and Martinelli, but only for intensity readings. How informative electromyograms would have been! 


A more serious objection to laryngeal lowering may be stated: If you ask a student to lower his larynx he MUST depress it, since the infrahyoids are involuntary. Even Stanley seemed to say that at times. Teachers of this persuasion give exercises designed to create laryngeal suspension, but do not talk to students about laryngeal height.

I think this objection goes too far.  Control over the jaw and the mylohyoids is possible, and exercises can coordinate the usage of the suspensory muscles. Ingo Titze says he prefers to use terms in teaching based on function because he believes that “the mind can control the body better if it understands how it works” (Titze, VOCAL SCIENCE AND HEALTH, p.Z88).


In order to help young students learn to relax the mylohyoids and allow the larynx to drop, Vennard mentions the beginning of the yawn, and the drop which happens after a swallow.  I used to try to get my students to remember bodily those times when they have to shout to make themselves heard, as in an emergency, or a call to the hot-dog vendor at the ball park. The low-larynx shout is an experience we a11 have. One of my former colleagues tries to get his students to remember what it is like to have to do a yawn when in church or class and not be able to do so out of politeness.  The pharynx broadens but the jaw doesn’t stretch. The “church yawn” still puts the jaw in some tension (keeping it from opening) but the sensation is similar to the relaxed jaw, low larynx, we want.


There may be some value in these suggestions still, but I believe Miller is correct in saying that the yawn is a dangerous image since it moves so quickly to its stretch component. It is also a lethargic endeavor and is inappropriate for singing. It is better to talk of the breath of surprise or astonishment. The usage of the shout has to be very cautiously used. It is too easily associated with “breath damming,” which is to be avoided at all cost. The fact that there is some similarity between a strong tone and a brief shout does not mean that shouting will bring about a good tone.


I go from those familiar sensations to other images designed to help the student relax the mylohyoids. Some of them can be quite humorous, like imagining you are a frog with a bag of air under your chin. I actually have some photographs of Richard Tucker singing with  fantastically relaxed mylohyoids which give just such an impression!  Images are essential as the servant of mechanics.  Images divorced from mechanics are meaningless and confusing. This does not mean, however, that the lesson time is primarily devote to image creation. Exercises for Appoggio are always primary. The relaxed jaw does not mean the hung jaw. A mobile jaw and lips are essential to Appoggio.


By now we have disparaged the “depressed” larynx so much that it hardly seems worth discussing. There are some issues worthy of mentioning, however.  First is the confusion in the name itself. The very term “depress” sends shudders through some teachers. We have already mentioned the case of Miller who equated low-larynx and depressed larynx in a disparaging way, although I think I have explained that he was speaking of an excessively lowered larynx, accomplished by localized actions.  Further adding to the confusion is the fact that the infrahyoid muscles are known scientifically as “the depressor muscles” (Sternothyroid, Sternohyoid, Thyrohyoid, Omohyoid, Cricopharyngeus {Husler}). Their function is not disparaged because of their nomenclature! In speaking of their usage as antagonistic muscles to the elevator support muscles we may correctly say they “depress” the larynx,  which is their function as “depressors.” Voice teachers, however, commonly use the term “depressed larynx” in a variety of confusing ways to refer to a poor articulation. In Vennard’s photos, the articulation is unclear, although his depressed larynx might be one which is PUSHED down from above by downward pressure on the back of the tongue, even though the sternothyroid is activated. Such an articulation could give the narrowing of the base of the pharynx shown. The root of the tongue is, after all, in the hyoid bone. Pushing down from above– a contrary usage of the suprahyoids–appears to result in a lower larynx but a narrower base of the pharynx. This is the reason the falsely depressed larynx should not be favored, as well as the tension in the pharynx resulting from such a muscular movement.
It makes sense that since the “one-sixth” condition cannot be met, sonograms do not show the singer’s formant in the depressed larynx of Vennard’s studies. It is the expansion at the base of the pharynx which is essential for the singer’s formant.

Teaching by using a real-time spectrometer can confirm this. I find that my students progress faster when they watch the spectral display in real-time during vocalization. One can experiment with many different articulations and watch the changes within the strength of the singer’s formant. With the high larynx there is no significant singer’s formant. With nasality, there is some strength in the 2k to 3k Hz area but the peaks below 1k Hz are blunted.  With the low larynx the strong singer’s formant is immediately observable. Letting the student experiment by pushing down on the tongue will show him how the singer’s formant strength drops out and also let him feel the difference between the lowering which expands the base of the pharynx and tongue depression.


In Jerome Hines’ book of interviews, the distinction between the lowered and the depressed larynx seems to be blurred. In interviewing Pavarotti, Hines looked at Pavarotti’s tongue during an “Ah” vowel and said, “I distinctly saw the back of his tongue depressing and his Adam’s apple lowering.”   While the back of the tongue must lower, it is something quite different to say that the back of the tongue depresses.  Vennard says, “When the larynx descends, it pulls down on the tongue at the back…” and earlier he defined the “Ah” tongue position as “reasonably low in the mouth, lower in front than in back, making the buccal cavity somewhat megaphone shaped.”  Therefore the slightly lowered back of the tongue does not imply any depression of the larynx. Certainly in the case of Pavarottt, the auditory evidence is overwhelming that the singer’s formant is present, the base of the pharynx expanded.  Pavarotti lowers the larynx. He does not depress it falsely. (How far does he lower it? How much do his sternothyroids contract? That would be interesting to measure.)

This is the sort of confusion in terminology regarding depression and lowering which we must clarify, as long as voice teachers continue to use the term, “depress”, as indicative of an improper articulation. The tongue is neither a legitimate elevator nor a depressor muscle.

It is generally believed that the “Russian School” teaches lower voices to “depress the larynx” by pushing down on the back of the tongue. The auditory evidence does not suggest any such generalization. Certainly, the evidence by the ears suggests that great Russian singers have the singer’s formant in the voice.
Russian vowels require different coupling, and consonants may be sounded longer. This is one reason pharyngeal space appears to be less in the Russian language. Nevertheless, the great Russian singers who take their place within the international circuit of opera seem to have “learned” how to increase pharyngeal space during singing, just as European and North American singers have.

If we use the term “depress” in its correct reference to the contraction of the strap muscles Dr. Lawrence talks about, then the low larynx is a “depressed” larynx which creates the one-sixth condition, the singer’s formant, and great high notes.
Maybe voice teachers should start using the medically correct terminology! If we equate the terms “depress” and “lower” out of scientific accuracy we must also clarify the more common usage of the term “depress” which suggests a usage of an illegitimate muscle for the laryngeal suspension system, usually the geniohyoid.









Or “To Twang or Not to Twang, That is the Question”
Further Discussions on Vocal Ring

Joseph Shore

We have discussed at length the 2800 (3200 {Vennard}) resonance of the larynx itself, resonating within the “collar” and the ventricle of Morgagni. We have seen how this laryngeal resonance is “independent” (Sundberg, “Scientific American”, 236, 81-91) of the vocal tract resonance which creates the vowel. We have discovered that this laryngeal resonance IS the source of the singer’s formant, and we have observed how its amplitude increases the amplitude of the partials near it.


We have placed this larynx-dependent “so-called extra formant”, between the normal third and fourth formants of spoken vowels, as Sundberg has shown us. The effect on the formants of sung vowels is two-fold. Since the larynx must be low (in most males of average physiology) to create the Singer’s Formant, the resonating tract is lengthened and the vowel formants are lowered towards their darker sides. The 2800 resonance itself, inserted between what would seem to be the third and fourth formants of the speaking voice, boosts their amplitude and gives us this peak envelope which has been called the “ring” or “ping” of the voice, once again, the research of Sundberg, Fant, Cleveland and others, shows that the 2800 is larynx-dependent, largely independent of the vocal tract. The number “2800” is arbitrary. Sundberg quotes studies which show that the actual frequency of the formant can range from 2.3K (in low basses) to 3.6K (in high tenors). Changes in laryngeal and pharyngeal sizes would be causative of the frequency changes in the formant. For the sake of convenience, however, we will just continue to use the term, “2800.”

It is really not accurate to talk about creating an “extra formant,” since it would be impossible to generate extra formants within the vocal tract. “What IS possible is to change the frequencies of already existing formants in the vocal tract. Thus, what we call the extra formant actually exists also in spoken vowels, even though it is tuned to a higher frequency, where voice source overtones are so weak that the formant does not appear clearly in an analysis of the spectrum.” (Sundberg, THE SCIENCE OF THE SINGING VOICE, P. 117)



Thus, what appears to be an “extra formant” inserted between the normal third and fourth formants of spoken vowels is actually the fourth of five formants where the frequency of the fourth is lowered towards the “2800” mark.

“If the pharyngeal cross-sectional area is considerably wider (more than six times) than the cross-sectional area of the larynx tube opening, then the fourth formant frequency is almost exclusively determined by the area function in the larynx tube, particularly the volume of the laryngeal ventricle which is the cavity located between the vocal folds and the ventricular bands.   Particularly if the lowering of the larynx expands the laryngeal ventricle, the fourth formant frequency can be lowered from its typical value of 3.5K in adult males  all the way down to, say, 2.8kHz. There are good reasons for assuming that the extra formant..is identical with the formant that in the model experiments showed a strong dependence on the larynx tube. Thus we may say that the larynx tube seems to be an important tool for obtaining the clustering of higher formant  frequencies needed for generating a singer’s formant.” (Sundberg, IBID. pp 101, 121.)

Vennard’s studies took place many years prior to Sundberg’s and do not take into account Sundberg’s conclusions. Particularly important is Vennard’s assumption that all 2800 is created equal, or so it seems from his statements. Vennard assumed that the 2800 ring of the voice could be achieved by putting “twang” in the vowel. Therefore, he called “twang” a necessary ingredient in the singing voice, sort of like salt in a cake, and gave his students exercises on what he called the “nasty,” “twangy” sounds in attempt to put this quality into all the vowels. I am using his description here. It is not really possible to add ingredients to vowels. What is possible is to “color” the vowels. There is not one set of formant frequencies for each vowel. Rather the vowels can be represented by “islands” of frequencies. Singers can color the vowels towards the lower or higher frequency side of the islands and still get the same vowel. “Coloring” outside of the islands gives another vowel.

It has been shown that if the third formants of vowels in men approach 2700 Hz there is an element of “twang” to them. It would seem, then, that any effort to “add” the “ingredient” of twang would entail a raising of the third formant of the vowel towards that frequency. Since the third formant of /i:/ is around 2700, some people have called this “twang” approach   “putting /i;/ into all the vowels.” Vennard gave his students exercises, as we have discussed from his major book, on the nasty, twangy vowel, /a~e:/  in attempts to give all the vowels some of this quality.
The effect of some nasality on the vowel would seem to be to boost the third formant frequency. In effect, he was attempting to teach the vocal tract of these students an articulator technique which raises the third formants of all vowels towards the “twang” mark of 2700 Hz. Since 2700 is close to 2800, he seems to have assumed that the events of “twanging” vowels and “2800 ring” were one and the same. In other words, “all 2800 is created equal.”  Pedagogues who had a desire for “bright” vowels liked this idea and these exercises are still used by some teachers. What can we say about this today? First, the “chiaroscuro” image, which was the functional aesthetic test for BEL CANTO for three centuries, matches the acoustical events described by Sundberg’s and Fant’s studies. The lengthened resonating tube, from the lowered larynx, shifts the singing vowels towards their darker sides (oscuro) and the resonance of the larynx itself, then set apart from the rest of the vocal tract, gives “ring” (chiaro). In the attempt to “twang” vowels the “oscuro” seems to be decreased. That is the empirical auditory impression. Let us consider twang in speech. The formants of spoken vowels are higher than sung vowels (Sundberg). The larynx is high in speech, and the velar seal may not be sufficient to eliminate all nasality. All this means that “twang” is easily achieved in the speaking voice. If you could hear my Ozark twang, you would hear that one of its characteristics is a low velum. If I were to open and shut my nose during speech you could hear how much of the sound is entering the nose due to the low velum. Vennard’s x-Rays show the low Velum during “twang” articulation. The larynx is also high. Raising the larynx higher, retracting the corners of the mouth, while lowering the velum is associated with the extreme forms of twang found in “Hillbillies.” Since I come from the Ozark Hills  I can demonstrate this fairly easily.

Vennard quotes Paget’s studies which show that twang resonance is circa 2700-2732 HZ. But Paget went on to say that twang is produced by “A CONSTRICTION OF SOME PART OF THE PHARYNX, SO AS TO PRODUCE AN ADDITIONAL RESONATOR OF HIGH PITCH, THOUGH THE PRESENCE OF NASAL RESONANCE (I.E. NASALITY) SEEMS ALSO TO BE indicated.”  Lets look at that. “an additional resonator” is formed by this articulatory technique. And that “additional resonator” could not be in the larynx itself since a “broadening of the pharynx”, not a constriction, is essential to create that. Therefore, the 2700 of twang is not the laryngeal 2800!! We are not quite sure WHERE Paget’s additional resonator is located, but “twang” experts claim it is in the area of the pillars of the fauces. Vennard agrees and says that the “twang” comes into the vowel just about the time the posterior pillars of the fauces are taut. The problem with this theory regarding the pillars and twang is two-fold. First, it does not detail the relationship between twang 2700 and the larynx 2800, and second, some singers claim that their pillars are relaxed, not taut, during twang. Both twang groups agree that the velum is low. Vennard discussed the 2800 resonance of the larynx itself but still thought twang was necessary. Why? Can we have a strong singer’s formant AND twang at the same time? If we can, do we NEED both for the functional aesthetic of Bel Canto, the chiaroscuro tone, and a good vocal hygienic tone created by flow phonation?   In answering these questions let us first look closer at the vocal tract shaping for twang. The key statement from Paget is that constriction of the pharynx is necessary.  We can outline the articulatory technique responsible for “twang”, collating Paget, Vennard, and Bartholomew:

  1. Tenseness in the walls of the vocal tract.
  2. Lowered velum.
  3. Narrow jaw opening and retracted corners of the mouth.
  4. Raised larynx.

We would also add a fifth factor from our understanding of the role of the tip of the tongue in changing the third formant values.

  1. The tongue tip must be kept touching the teeth.


Using this articulation, the third formants of all vowels will be raised towards Paget’s mark of 2700 for twang. The vowel /i:/ will be the most twangy since its third formant is already near or at that mark. However, everyone agrees that aspects of at least 1,2, and 4, if not ALL five of these articulatory elements, are bad for singing. Therefore an attempt was strenuously made to correlate the necessary requirements for “twang” with a better usage of the vocal tract. Vennard attempted to minimize these features, i.e., to keep the larynx as low, the jaw as wide, the velum as high, the walls of the vocal tract as unconstricted as twang would permit.

To make sure that I am not improperly stating Vennard’s position, I am going to show his extensive passage on twang and ring from his major book.


“When the tensors work alone the soft palate is stretched from the level of the hard palate to the level of the upper teeth, and the pharynx is separated from the oral cavity. In some animals to whom the olfactory sense is indispensable, the velum and the epiglottis actually overlap, so that the odor of food which may be in the mouth will not prevent the animal from smelling the approaching danger which may be downwind. When the levators contract they are opposed by the tensors (another case of muscular antagonism) and the velum forms a double arch extending from the hard palate to the upper constrictor. This causes the throat and mouth to be one continuous cavity, and closes the nasal port; that is, it separates the pharynx from the nasal cavities. A band of fibres in the superior constrictor (called Passavant’s ridge) completes the closure. The soft palate is thus a two-way valve, opening the pharynx either into the mouth or into the nose. When it relaxes it is in an intermediate position. I am convinced that it should be up.

   “Attached to the aponeurosis of the soft palate are three thin flat muscles whose fibres twine together like ropes at the sides. The ones in front go into the sides of the tongue. You may call them palatoglossus or giossopalatinus depending upon which way you think they pull. When the larynx is low and the palate high there is considerable stretch in the palatoglossus muscles and they stand out from the walls of the throat, forming an arch over the back of the tongue. They are called the anterior pillars. The other two muscular layers of the palate enclose the slender (and apparently useless) muscles of the uvula, which arise from a point at the middle of the edge of the hard palate. These same layers also enclose the insertions of the levators, and their fibres rope together to form the palatopharyngeus, or pharyngopalatinus muscles. They form another arch and are called the posterior pillars. Between the front and rear pillars are the well-known tonsils, which are only two of a whole ring, including the adenoids, and similar glands on the back of the tongue and in the pharynx.

409  There is likely to be stretch in both pairs of pillars, increasing as the pitch rises and the extrinsic musculature comes into greater play along with the mounting activity of the cricothyroids. This is the one exception to my reiteration of the word “relax* throughout this discussion. I think this is what is referred to by those teachers who prefer “throat or pharynx firmness*
(Par. 333). A good many refer to a feeling of stretch, (Lehmann, p. 98). Lawson tells us, “It will feel almost as if the vocal cords were attached to an elastic stretched from the hard palate to the diaphragm,* (p. 34). If we remember those longitudinal fibres of the esophagus his statement is borne out almost literally.

410  When the desired adjustment has been achieved the overtone I have called •2800″ comes into the voice just about to the extent that the posterior pillars are stretched. Since they are a port of the soft palate, the “2800” formant is often miscalled “nasal resonance.*

411  The sound to which I refer is “twang,” which is a characteristic of “Yankee” speech. Nasality may be mixed with it, because if the velum drops enough in achieving it, the “honky” quality of the nose will sound, but this is only a byproduct. Paget’ reproduced this overtone by constructing “a resonator formed of a rubber tube (about 1 inch in diameter) attached to an organ reed and fitted with a cork tongue.” His account continues:

It was found that if, while the reed was sounding, the tube was suitably pinched, near the opening from the reed, … an appreciable twang was added to the vowel-sound. This experiment indicates that a part, at least, of the so-called nasal quality … is probably due to a constriction of some part of the pharynx, so as to produce an additional resonator of high pitch, though the presence of nasal resonance seems also to be Indicated, (pp. 95, 96).

Paget estimated the frequency of the resonator to be 2732, which certainly agrees with Bartholomew’s “2800.” Russell also tells us that exaggerating partials above 2500 in frequency gives typical New England “twang,” (p. 172). There is a resemblance between the experiment just described and those of Miller and Russell in squeezing pipes, (Par. 306, 307). Perhaps “twang” can be explained in terms of muscular tension (including the posterior pillars) without locating a new cavity. Nevertheless, there are two cavities to be mentioned, formed by the pillars. Paget also reports that according to Eljkman a pair of small resonance chambers is formed between the palatopharyngeal pillars and the back of the pharynx, which produces the frequency of “twang.”

412  This quality of “twang” is a needed Ingredient of tone, like salt In a cake. Too much is ruinous, but none at all is equally undesirable. This is probably the real reason for the pedagogical success of the hum and such consonants as m, n, and ng, which exercise the velum. They produce nasality, but they also call into play the pillars, and once the “twang” is induced, the “honk” can be eliminated by arching the velum. Between “twang” at one extreme and “honk” at the other we hear the sounds of the French nasal vowels (Par. 513).

413  A somewhat contradictory theory is offered by Bartholomew. He calls attention to the fact that the pillars are part of the group of swallowing muscles which raise the larynx, making the pharynx a poor resonator. Relaxing the velum tends to relax the swallowing muscles, and hence, producing nasal sounds tends to drop the larynx, which everyone agrees is essential to good singing.












X-rays from van den Berg0 and Vennard, NATS Bulletin. Fleshy parts have been retouched
from visible evidence In the negatives. HI, third cervical vertebra. In ‘twang* openings to
mouth and throat appear equal, throat larger in diameter. In “honk* larynx Is lower, but open-
ing to mouth Is almost closed. See Par. 513.

Sonograms from van den Berg8 and Vennard, NATS Bulletin, corresponding to the adjustments shown above in x-rays. Pitch, C3 on bass staff. Time, 1.5 sec. Scale of kilocycles at left. Note that “twang” shows formants up to 3500 cps., “honk” shows little above 1000 cps.
Sonagraphy is discussed in Par. 458.

Bartholomew continues,


It is quite possible, after the ability to secure the enlarged throat has been acquired, to learn to lift the velum lust enough to close off the naso-pharynx and destroy the head resonance while still maintaining the enlarged throat and the relative relaxation of the swallowing group. And, yet, one of the most effective ways of first learning how to secure that throat setting is by use of some such imagery as “head resonance.” Thus, one should not condemn this Imagery.
(p. 153, see also Bartholomew2).






This sounds like a hodge-podge of fact and fiction, all due to the confusion regarding the location of 2800. First of all, the pillars of the fauces belong to what may be one of the most important elevators, palatopharyngeus. At least some researchers and teachers believe that palatopharyngeus must contract to act as antagonist to the infrahyoidal muscles. This action is thought to stabilize the larynx. Dr. Van Lawrence certainly hints of this in discussing the role of the “strap muscles,” and Husler makes bold assertions regarding its antagonistic function. Furthermore, it has never been shown that nasality lowers the larynx. Usually a low velum is associated with an elevated larynx (Garcia). The extrinsic laryngeal musculature works by antagonistic pairs. If palatopharyngeus relaxes, its antagonist, sternothyroid relaxes and the larynx remains high and flaccid. Therefore, contraction of the velar-palatal muscles, which raise the velum, does not mean the larynx has to rise. To the contrary, an arched velum, and active palatopharyngeus muscles may be part of the antagonistic system of the low larynx described by Sundberg as necessary to generate the singer’s formant. This would be the case if the sternothyroid muscles responsible for lowering the larynx to the one-sixth posture respond as antagonists to palatopharyngeus (and stylopharyngeus [Husler]).  If Eijkman is correct, and a pair of small resonance chambers is formed between the palatopharyngeal  pillars and the back of the pharynx which produces twang, then we can see that twang and the singer’s formant are two different events, the former dependent on “Eijkman’s cavity” and the latter dependent on the laryngeal cavity, the ventricle of Morgagni, as Sundberg has described. Vennard’s description of the two-way stretch involving the pillars would correspond to the muscular activity when the larynx is inspanned by the suspension system. The “2800” comes into the tone, then, because the lowered larynx “has a resonance frequency of its own” (Sundberg, “Scientific American”), not because there is twang in the vowel. The attempt to put “twang” into the vowels, raising the vowels’ second and third formants, seems to be an attempt to attain the peak effect of the singer’s formant by radically raising (coloring) the upper vowel formants towards their brightest side. It would not seem that this radical articulatory technique could match the quality or the efficiency of the larynx-dependent singer’s formant, at a frequency between 2.3K and 3.6K for adult males, appearing to be BETWEEN the normal third and fourth formants of the spoken vowel. In fact, it appears that the lowered larynx lowers all vowel formants. The third formant, associated with “twang” when raised, LOWERS about 11% when the larynx is lowered to produce the singer’s formant. (Sundberg, IBID. p. 115). Therefore it would not seem possible to get “twang” and a strong singer’s formant.


It would be very interesting to see X-Ray profiles of the vocal tract during twang articulation. The X-Rays we have of Vennard singing with a “twang” and a “honk” show important changes in the vocal tract. In twang the larynx is relatively high, the velum depressed enough to give a larger entrance into the nose than to the mouth. The tongue has some bulging towards the posture the Sundberg—Lindblom model identifies as /e;/. The sonograms for C3 on the bass staff, seem to show only four formants. The third is around 2700 Hz and the fourth is around 3400Hz. A fifth is not clearly shown, nor does there appear to be a visible clustering of the third, fourth and fifth that we get from the singer’s formant. In the “honk” the velum is very low, the larynx high, and the opening to the mouth has been almost closed. The energy peaks of 700 to 1000 Hz have been blunted and while there is some energy above 2K it is relatively weak.

We would have to conclude from all of these considerations that attempting to add “twang” quality to all the vowels is something very different from articulating the singer’s formant.

Requesting singers to add twang seems to be requesting them to raise the third formant of the vowel. The other formants will likely be affected in the process. It is highly doubtful that singers can raise the third formant of vowels independently.

“How can one determine the articulatory changes necessary to change the frequency of a given formant in a specific way or to determine the formant frequency consequences of a shift in the positioning of one specific articulator? It would not be a good idea to ask a subject to move one articulator and try to keep all other articulators constant, because the subject has no “possibility of obeying: we do not have that kind of command over articulation. If we move one articulator, we can take it for granted that all others will move as well..Also our awareness of our articulatory activities is generally low. We think of articulation in terms of the sounds we produce, rather than in terms of positioning articulators” (Sundberg, IBID. pp96-97).


Vennard distinguished between his “twang” technique and the overt addition of nasality, but many teachers do not.   The effects of nasality on the vowels are thought by some to be identical to “vocal ring,” and therefore they teach students to mix nasality with their vocal tract tone. They may call it “nasal resonance.” Some even identify it with the singer’s formant. Ingo Titze writes: “Confusion arises when perception (and particularly autoperception) is linked to production without careful scrutiny of the acoustic signal…For vocal ring, an increased concentration of energy in the 3000 Hz region has been explained acoustically and physiologically by Sundberg…For nasal vowels, the energy in the low frequency region (0-1K Hz) is spread out…This reduces the prominence of the peak of the low-frequency energy in relation to the peak of the high frequency energy. If we now make the assumption that perception of vowels is based primarily on the energy of the peaks of the spectrum, it is reasonable to draw a tentative conclusion that nasality may bring about a perceived increase in brightness of quality, similar to vocal ring. The mechanisms are completely different, however, in one case (nasality) we lower the low frequency prominence of the oral and pharyngeal resonance by creating an acoustical leakage through the nose; and in the other (vocal ring) we raise the high-frequency prominence of the oral and pharyngeal resonance by creating AN additional resonator  IN THE LARYNX.” (Ingo Titze, VOCAL HEALTH AND SCIENCE, p. 244-245) Clearly then, the effects of nasality on the vowel are NOT the same as vocal ring.

Titze continues: “Returning to nasality..it is known that nasalized vowels have an enhanced widened first formant region, but the effect of the higher partials varies a lot from person to person. Perceptually, there seems to be a ‘honk’ and a ‘twang’. The honk often heard in speech of the deaf and those having cleft palates, is related to the low portion of the frequency spectrum. This is not the quality the singer is trying to achieve.” (Ingo Titze, IBID, p. 243)

“Returning to the pedagogical issue of using nasality to achieve a brighter quality, we are faced with a situation where a little bit a deception might lead to the right perception. Teachers may use nasal sounds to trick the student into experiencing some of the key sensations in the facial region, but then quickly point out the alternate (and preferred) ways of obtaining those same sensations. Otherwise, the singer may learn to interpret excessive nasality (twang) as vocal ring. Pinching one’s nose during vocalization is one easy way to test for excessive nasality.” (Titze, IBID,.p.245) 


Pedagogically, this technique seems a tangled mess!. Why not concentrate on laryngeal suspension exercises in order to “create the additional resonator within the larynx”?  It is not just the student who gets confused. The teacher does also. As we mentioned, Vennard called “twang” a necessary ingredient like salt in a cake and thought that twanging vowels brought about vocal ring. The difference of vibrations in the nose during nasalization from vibrations in the head as a result of vocal ring in the larynx can be experienced as soon as the laryngeal resonance has been established. I have never had a student who did not think the two were distinctly different once the larygneal ring had been established. The 2800 energy of the laryngeal ventricle may give vibrations in the cervical vertebrae behind the larynx and in the so-called mask area of the nose and face. Like the sounds of noisy neighbors passing through apartment walls, the high energy of the laryngeal 2800 passes through the walls of the vocal tract and creates vibrations singers become aware of. Those vibrations turn out to be much too small in amplitude to add anything to the sound which radiates from the mouth. If those vibrations are used as biofeedback monitoring by the singer it is important to realize they are laryngeal byproducts. Otherwise singers are apt to make articulatory changes, especially in the jaw and lips, designed to increase their awareness of “something” happening in the masque area. By and large, those changes drift far away from the simple jaw, lips and tongue postures Sundberg and Lindblom show in their articulatory model for vowels. These changes are rightly called “radical reflex postures.” Much of the time, the singer’s formant strength is not at all increased by these changes, although the resulting muscle tension in the jaw, mylohyoids, and the neck, seem to communicate to the singer that “something” is happening in the face. In some cases, these radical postures may even effect the larynx height– after all, the mylohyoids and digastric muscles, as well as the lower, middle and upper constrictors, pull UP on the larynx– and negate the essential one-sixth condition between the area of the larynx tube and the area of the base of the pharynx, depleting or defeating the singer’s formant. Lots of SPEAKERS claim to SPEAK in the masque. Since the singer’s formant is clearly not in the average speaking voice, masque sensations are usually caused by different processes than true vocal ring within the larynx, giving the singer’s formant. All pedagogy “tricks” should be directly geared towards establishment of the one-sixth area relationship between the larynx tube and the expanded base of the pharynx. Otherwise, both the student and the teacher may confuse the singer’s formant with mixing nasality with the tone.

The singer’s formant is not in the nose, behind the nose, in the sinus cavities, in the teeth; it is in the larynx. This basic understanding would seem to be clarified now by studies of laryngeal resonance and vocal tract resonance. All 2800 is NOT created equal. Singer’s might be well advised: “Ring”, don’t “twang.” Or we may use Vennard’s own words when he said, the nose “is a poor means of taking a breath quickly, and it is a poor resonator for either improving or building a tone. For these reasons, most singers both inhale and sing through the mouth”






















Vennard says, “It is now recognized that the sound vibrations cannot be directed at all. They enter every nook and cranny possible, and also set all the bones into vibration. Where they reach a mass of the right natural frequency, be it air or be it bone, it will be agitated by sympathetic resonance, and may produce a sensation of which the singer will be conscious. But these vibrations add nothing to the tone that reaches the ears of the audience.” We have already noted the experiments of Wooldridge, Vennard, et a1., which showed the fallacy of nasal and sinus resonance. Ideas of nasal and sinus resonance emerged in an era before scientific investigation of the voice. But even in empirical circles, singers show very different experiences and attitudes.  It has already been noted that in Jerome Hines’ now oft quoted book, about half of those great singers were undecided about placement or did not believe in it.

This seems to be an emotional topic. Singers and teachers become very combative and defensive when discussing placement or its Italian equivalent, “impostazione”. The reason seems to lie in the fact that singing is a very personal event, and to call into questions one’s “feelings” is to risk “hurting feelings.” Hopefully, we can avoid that and look at this “illusion” more closely. Certainly, there have been many “images given by voice teachers which have been confusing. An image based on fact can help. An image denying fact almost surely hurts.


The first thing we must acknowledge is that, since sound waves come out from the voice source like concentric circles, they cannot be placed. A straight line can be directed, an object placed, but circular waves, like the ones caused by a pebble thrown into a pool of water, cannot be placed. Therefore the term itself is highly inadvisable. Doubtless, it emerged from the assumption that the singer’s feelings—in the head, skull, face, chest, etc.–contributed GREATLY to the sound heard by the audience. We now know that is not true. In measuring such vibrations it was necessary for Lindsley to use an “electrical stethoscope” which greatly amplified them. As Vennard says, “Obviously they could be no part of what is heard by others.”


In a very informative study in 1964, Kirikae et a1. measured vibrations caused by phonation in forty different locations on the body.  The results are interesting for singers’ “feedback signals”, or, what is thought of by some as, “placement”.  They found that all vowels produced vibrations with very small amplitude on the forehead and cheek. The vibrations on the JAW were very strong on all vowels, strongest on /i:/. Vibrations were strongest on the larynx itself, virtually vowel independent, which is not surprising, since “vowel differentiation is achieved in the vocal tract” (Sundberg). All vowels also produced strong vibrations on the sternum and on the lips. As you would expect, vibrations in the chest wall, while too weak to contribute to the radiated sound at the lip opening, were felt stronger at lower frequencies, strongest around 180 Hz. Above that, they decrease in sensation. It is not possible to sense them at all above 300 Hz, or D4. This is very interesting, since the so-called “chest voice” is supposed to end around that area.

Kirikae also measured the amplitude of vibrations on the top of the skull and found that the lower the first formant, the greater the amplitude of vibrations (Sundberg). That is pertinent, because the lowering of the larynx lengthens the resonating tube, thereby lowering formant frequencies. This shows us that the lowered larynx also gives stronger biofeedback singer-signals in body vibrations. The strange thing is that most teachers in this century have not attempted to cultivate the singer’s awareness of body vibrations where they are strongest, in the jaw, the larynx itself, the sternum, and the top of the skull. One could make a good case for using such vibrations as auditory feedback. It could be said that the voice is “placed” in any of these bodily locations, but the very usage of such a term shows its limitations. What would it mean to say the voice is “placed in the sternum”? No more or less than to say it is “placed” in any other area!  It is better to use the word “placement” to describe the “positioning” of the larynx itself in the scaffolding of the suprahyoids and infrahyoids. The larynx is “placed”.  Air “flows”. The vocal folds “vibrate” or “oscillate”. The sound stream “resonates”, and the vocal tract “transfers” sound. The correct verbs are important to describe function.

Bodily vibrations are the RESULT of resonation happening in the vocal tract. They are NOT causes of resonation, or evidence of any addition of resonance to the vocal tract resonation. Contrary to many 20th century advocates of “nasal resonance”, this knowledge goes back to Bel Canto pedagogy, of which Lamperti was considered an heir. In his VOCAL WISDOM, Lamperti said:

“Always remember that what goes on above the throat are illusions, no matter how real they may feel and sound…Although you may acquire a wide range of voice, you cannot modulate the sounds until the resonance of your tones becomes round and rich, ‘chiaroscuro’.  The pitch (vibration) of your voice seems to emanate from the back of the mouth (pharynx) spontaneously. The resonance of your voice seems to originate in the front of your mouth (lips) voluntarily. These two together seem to make the chiaroscuro (clear-dark) tone…There is vocal resonance like unto a composite vowel-sound containing all of them. The reverberation of the voice is felt as an elastic solid, filling head, throat, and (in low tones) chest. When the composite sound is focused in the middle of the skull, it can be molded by the lips into any form or shaded into any color, changed from one vowel to another and made to open or close at will…The vocal bands then tune this chiaroscuro tone without push or pull.”

To be fair I should have continued Lamperti’s quote because he says that even though vibrations above the throat are illusions, these illusions provide the singer with a means of controlling singing!


It is certainly true that vibrations in the head give the singer biofeedback information that is very useful. However, different voice types among singers experience these vibrations differently so as to make it impossible to systematize them for a generalized pedagogy. Asking a basso profundo to try to experience the same vibrations in the head that a light lyric tenor feels is doomed to failure.



These images are closer to reality. They do not talk about nasal resonance, or ask the singer to increase vibrations in the face. Most importantly, he speaks of a vocal resonance which is a composite vowel sound containing all of them, a “chiaroscuro” tone.  Notice that this is not a “bright” tone, but a “clear-dark” tone. He also seems to agree with Manuel Garcia II that “the pharynx is the mouth of the voice”. (Garcia also called “placement” feelings, “mere appearances”. His image of a “composite sound” seeming to be focused in the middle of the skull might correlate with Kirikae’s findings. The presence of a low first formant frequency, associated in most people with a low larynx, produced the greatest amplitude in top of the skull vibrations (Kirikae, Sato, Oshima, Nomoto, VIBRATION OF THE BODY DURING PHONATION OF VOWELS). The tone Lamperti is describing has a very strong lower partial—low first formants amplify pitch fundamentals more generously —and a good 2800 larynx resonance. Hence, the image of “chiaroscuro”, a tone, in his terms, round and rich, dark and clear.  None of this sounds like the pedagogy found in 20th century studios which teach that the “bright” sound is the most resonant, and that the voice must be “placed in the masque”! There, we have finally used the phrase!  Where and when did the concepts of nasal resonance and masque placement originate?


Witherspoon comments that ‘nasal resonance’ is not to be found in any of the


writings of teachers from Tosi to Lamperti, i.e. from the 17th to the beginning  of the 20th century. The inventor of the term and concept of nasal resonance, “singing in the masque”, was Jean de Reszke (1850-1925), a Polish, French tenor who taught singing in Paris and Nice from 1902 until the time of  his death. He teamed-up with his laryngologist, Dr. Holbrook Curtis, to popularize his concepts, especially nasal resonance. 



Even Richard Miller, a critic of laryngeal lowering, says:


“It  is well known to what extent concepts of chanter dans le masque’ have dominated vocal pedagogy in France in this century. De Reszke is credited with bringing the system to prominence and acceptance in France, but it would seem more logically to be the result of national linguistic tendencies and tonal ideas…If the term ‘forward placement” is admissible as a description of resonance concepts in singing, it surely more appropriately matches the resonance goals of the French School than any other.”


Notice that these concepts are NOT a part of historic Appoggio. Miller says that in the historic Italian School (the International School) known as Appoggio, placement sensations are not isolated in one place but are permitted in all parts of the vocal tract.

As we have already mentioned, the concept of strict national schools is problematic, but it may come as a surprise to discover that nasal resonance, singing in the masque, and forward placement, are all concepts which link together, in De Reszke and French singing. They spread like wild-fire throughout the singing world of the 20th century, largely because De Reszke was such a famous singer. Caruso did not live long enough to teach or we would doubtless have been deluged with Caruso’s influence in pedagogy, perhaps for the better.


William Johnstone-Douglas, De Reszke’s assistant for many years, writes: “He (De Reszke) placed great importance on mask resonance, which was used to obtain maximum resonance and ring in the voice. This was gained by allowing just enough room for the tone to be thrown up into the resonance spaces behind the nose, using the sound of the French ‘an’,(a~) in ‘souffrance’. I gather that he was obtaining a vibration of air in the upper turbinate of each nostril, which lies almost between the eyes.” 

This is a precise example of misunderstanding of acoustical events. First, it is obvious that De Reszke thought that the bodily vibrations added greatly to the tone–“mask resonance to obtain maximum resonance and ring”.  We now know that this is not true. The space behind the nose is the naso-pharynx which is shut off by the raising of the velum against Passavant’s cushion. Therefore the naso-pharynx is not a resonating cavity except in nasalized vowels. Vennard says: “The closure of the naso-pharynx should be complete. As a demonstration that nasal resonance is of negligible value, I sometimes sing a tone for my students while alternately closing and releasing my nostrils with my fingers. There is practically no change in quality.” He also mentions Passavant’s famous experiment in the 19th century in which tubes were placed in the nose to keep open the space between the velum and the cushion on the pharyngeal wall which bears his name. He discovered that the seal of the velum did not have to be hermetic to avoid nasalization. “An opening 12sq. mm in cross-section HAD NO REACTION UPON VOWEL QUALITY, BUT ONE 28SQ. MM RESULTED IN DISTINCT NASALITY” (Vennard).  Not everyone has wanted to admit what that means. It should be obvious. The only resonance of the nasal cavities is nasality. There was no gradual change in vowel quality. It simply means that the seal of the velum does not have to be hermetic. However, Bartholomew wanted to put it another way, as quoted by Vennard: “If the nasal resonance is only a small part of the tone, it is not objectionable and is called ‘head resonance’.”  Firstly note that in historic Appoggio, “head resonance” is never confused with even a little nasality. Secondly his conclusion needs examination   Passavant’s experiment showed there was NO CHANGE in vowel quality—not some small addition of nasal resonance—unti1 an opening into the naso-pharynx, 28sq. mm in cross-section, resulted in distinct nasality.  Put another way, either the vowel is resonated in the vocal tract or it is nasalized, although vocal tract resonance may be mixed with nasality voluntarily and many do.

Reid and many others point out that a cavity must be adjustable to act as a resonator and the nasal passages are not capable of adjustment for pitch. Furthermore, as Sundberg has pointed out, their mucous membrane linings are more suited for DAMPING tone than enhancing. We have already quoted the scientific investigations of Wooldridge, Vennard et a1., which indicate that the nasal cavities cannot add anything to the tone which radiates from the mouth opening in non-nasalized vowels. One wonders why certain pedagogues continue to talk about “coupling” the nasal cavities with the other resonators.


Richard Miller’s book, ENGLISH, FRENCH, GERMAN, AND ITALIAN TECHNIQUES OF SINGING was published in 1977. In discussing “the material of voice placement”, he quotes H. Lowery, A GUIDE TO MUSICAL ACOUSTICS, published in 1966: “…the richness and quality of the voice is determined in the main by the head resonances arising from the various cavities, nasal and sinus.”

One wonders why such a statement should be quoted. Lindsley’s, Wooldridge’s, Schaeffer’s, and Vennard’s contradictory studies were surely known to Professor Miller. Kirikae’s findings were known as of 1964. True, obviously Mr. Lowery seems not to have known of these studies, but Professor Miller must have. Then Professor Miller continues: “Almost all singers believe some part of the distinguishing characteristics of timbre and quality are determined by the changing shapes of those resonators which lie above the larynx, specifically the cavities of the mouth and the pharynx, and BY THE EXTENT OF NASAL CONJUNCTION WITH THEM.”
There we have what appears to be a planned presentation for introducing respectability to the concept of nasal coupling!

In his vocal pedagogy textbook, THE STRUCTURE OF SINGING, Professor Miller says that there may still be a possibility of coupling the nasal cavities with the pharynx and buccal cavities, and as proof quotes one sentence from Sundberg’s now famous article, “The Acoustics of the Singing Voice,” published in 1977. I find that amazing since the whole article deals with the VOCAL TRACT as the resonator for the voice. Sundberg simply admitted he had not studied the nasal cavities for the model in this article. In the VERY NEXT paragraph he does say: “Resonances outside the vocal tract, such as in the head or the chest, cannot contribute appreciably to the singer’s acoustic output in view of the great extent to which sound is attenuated as it passes through tissues.” To quote Sundberg in order to prop up nasal resonance is absurd and misleading.

RESEARCHERS THE CONCLUSION, “THE TERM ‘NASAL RESONANCE’IS WITHOUT VALIDITY IN DESCRIBING VOICE QUALITY IN THE SINGING VOICE” (Wooldridge, Warren B. The Nasal Resonance Factor in the Sustained Vowel Tone in the Singing Voice, Indiana University, 1954).  As we have seen, subsequent research has confirmed Wooldridge’s results. Dr. Van Lawrence comments: “One of our greatest living sopranos has had extensive nasal and sinus surgical procedures done with no significant alteration in resonance…I have had the principals in one entire production of LA BOHEME all ill with purulent maxillary sinusitis on one or both sides, yet singing without audible impairment or difference”(VOCAL HEALTH AND SCIENCE, NATS, p.48).  So much for nasal resonance. Therefore, even if you could “place” your voice in the “mask,” and you can’t really, it would not make your voice more resonant! 


Richard Miller told this writer that he remembers distinctly how he felt when he first learned that the sinuses contribute nothing to the tone that exits the mouth. He was dumbfounded, he said, because his own singing voice gives many anterior sensations.


DeReszke’s mask resonance was simple nasality as Johnstone-Douglas’ account of his teaching shows:

” Because of its numerous nasal syllables, a very effective sentence to obtain this resonance was ‘Pendant que 1’enfant mange son pain, le chien tremble dans le buisson,’ monotoned on each successive syllable of the scale…In a course of purely mask singing ‘o’ and “a” would become ‘on’ and ‘an,'(i.e. nasalized), the open ‘e as in “etre became ‘ain’…”

This is a simple account of nasalizing  vowels, nothing more. The “feelings,” the bodily vibrations, in the naso-pharynx, during nasal phonation, De Reszke erroneously believed to be a resonance which would give “maximum resonance and ring in the voice.” Reid writes on this:


“In recent years there has come into vogue a belief in the doctrine of ‘nasal resonance’. The first enthusiasts for this method of tone production were Dr. Holbrook Curtis and the tenor Jean de Reszke. Between them they share the responsibility for establishing and then, despite having witnessed its obvious failure, helping to perpetuate one of the most ugly and destructive forms of voice production imaginable. An example of the futility of nasal resonance as a basis for correct tone production has long been, and is continually being, demonstrated by French singers, or those who employ the French style, whose ‘white’, ‘thin’, ‘pinched’, ‘non-resonant’, ‘inexpressive’, off-pitch singing has long been the rule rather than the exception. Only in the rarest instances does a singer trained in the French style acquire a technique which keeps the voice fresh and youthful after many years of arduous singing. Contrary to belief, the French nasals can be reconciled with the principles of Bel Canto.” (BEL CANTO, PRINCIPLES AND PRACTICES, p.186)    The position of Dr. Curtis in the musical world was that of a laryngologist who for many years had specialized in relieving singers of throat troubles brought on by incorrect or over-extended singing. Toward the close of his singing career Jean de Reszke was one of his patients. Between them they ultimately arrived at the conclusion that the vocal cords were too weak and fragile to withstand the rigors of operatic singing. In order to relieve the cords of all pressure built up by tension De Reszke and Curtis struck upon the idea of letting the head cavities assume the burden of effort.





“If the results of his work had not been such a tragic blow to the art of singing, in many respects Dr. Curtis’s book is amusing. In his analysis of the muscular operation of the vocal mechanism Dr. Curtis finds it necessary to devote almost two hundred pages to describe the complexity of its action in singing. He shows how the vocal cords are stretched and brought into tension to establish pitch: he demonstrates how the breathing muscles should and should not work, and how they co-operate with the stretching of the vocal cords to determine pitch; he explains exactly how the resonance cavities are shaped, and so on. After many highly informative sections have been given over to the anatomical, acoustical and physiological phases of tone production, Dr. Curtis then proceeds to reveal the practical implications of his argument. The section of Dr. Curtis’s book showing the practical application of the material discussed is truly incredible. The voice should be brought ‘forward’ and ‘placed’ in the masque. The means suggested for accomplishing this end are simple. Merely commence each vowel with the lips closed, as ‘mah,’ ‘moh,’ ‘mee,’ etc. This immediately transforms the tone quality and presumably readjusts the entire mechanism so that it is brought into a more correct co-ordination. Apparently the vibrations, reinforced by the action of the nasal passages, relieve the vocal cords of all tension and strain, and all subsequent singing is free and effortless.” (Reid, Bel Canto)
The idea that every sound begins with a hum and that humming or using nasalized sounds places the voice forward in the mask is a pure invention of Holbrook Curtis, which continues to be used in some 20th century studios. Lamperti frankly told his students, “Don’t hum”! His biographer, William Earl Brown says, “When Lamperti said, ‘Singing is humming with the mouth open’, he did not mean that humming with the mouth closed would bring this about! ‘If you cannot sing with the mouth open, you cannot with it closed’, he would exclaim. ‘You cannot hum right, until you can sing right…Don’t hum!'”

Today, the vestiges of De Reszke’s misconceptions continue in the pedagogies of teachers who think that by humming or using nasalized sounds, the voice can be “placed in the masque”. I should add, however, that humming may indeed be used to show some pressed voices the sensation of relatively smaller amounts of medial compression and greater degrees of cricothyroid stretch. Towards that goal humming may give a useful illustration to the singer.

Dealing with “appearances” instead of realities conies from teaching by sensations. Witherspoon commented aptly:

Sensation is responsible for much of the confusion in teaching, because teachers try to induce correct sensation in the pupil through imagination, imitation, or suggestion, in order to get the correct tone, instead of asking the pupil to “do” something to cause correct action which produces correct tone, and which in turn will cause the correct sensation. That is, sensation is an effect and not a cause of tone. Correct sensation may be a guide after it has once been experienced by correct singing; it cannot be obtained except by correct singing. We may ask a man who has never eaten an olive what an olive tastes like, or what is the real “taste sensation” of eating an olive. He will promptly voice his ignorance, and say, “Let me eat an olive and I will tell you.” The sensation then becomes a guide for future eating. One of the most pernicious and dangerous obsessions in the teaching of singing is the belief that pupils can be taught directly to “feel” the voice in some part of their anatomy! From this have come more ridiculous fads and fancies than from any other illusion.


The attention of the singer and pupil for the past two or three decades has been directed toward the acquirement of “nasal” resonance. To this end all kind of tricks have been invented to induce the pupil to “feel” something instead of to “do” or “sound” something. The correct sensation of the resonance of the head or “masque” can be gained only by correct singing, therefore by correct coordination of all of the breath and voice organs, for the simple reason that resonance is the result of sympathetic action, not separate action. As a matter of fact, the endeavor to “feel” some unknown unexperienced sensation is generally productive of tension and interference. It is the old habit of putting the cart before the horse. Sensation may be a guide after it is obtained by correct singing, but it is evident that one cannot get the right sensation unless he makes the right tone.




Vennard quotes Bartholomew on another usage of nasals which is supposed to help the student lower the larynx. Bartholomew thought that a relaxed velum with relaxed pillars of the fauces helped the singer lower the larynx. Once the larynx is lowered, he thought, it was possible to raise the velum just enough to cut out the nasality while keeping the larynx low.  This theory has a hundred problems and I have never found it to work in the studio. Bartholomew thought that the 2700 Hz of twang was the same as the “2800” of laryngeal resonance. He wanted the velum lowered to get twang, the larynx lowered by the relaxation of the velar palatal muscles, then a slight raising of the velum to shut off nasality while keeping the twang.   The 2800 laryngeal ping is not located in the velar area! Researchers are now fairly well in agreement on that. If there is a cavity for twang in the velar area (Eijkman’s cavity) then it can be shown that twang and ring are two different events, since ring is now known to be dependent on the laryngeal ventricle.  Lowering the velum does NOT seem to help people lower the larynx! In fact Garcia suggested a relationship of the velum and larynx matching my experience. He suggested that when the velum is low the larynx is HIGH and when the velum is high the larynx is LOW. Nasofiberoptic laryngoscopy performed on ventriloquists shows that nasality is linked to a raising of the larynx.  Yet many teachers use nasals at the onset of a tone because they believe that nasality will relax the throat and lower the larynx. In fact, two different unconnected events are being asked of the student. Even the “sensations” of nasality and laryngeal resonance are different.


Teaching by sensations actually began in the nineteenth century as an attempt to find a short-cut to the arduous six to ten years of study Bel Canto required. In his HISTORICA MUSICA, published about 1624 Bentempi records an impressive description of the daily curriculum set up by his teacher, Mazzochi:

‘One hour in the morning was set aside for difficult passages, another for the practice of shading, another for singing before a mirror, in the presence of the master, in order to acquire a good position of the mouth and a pleasing attitude in singing. In the afternoon a short time was devoted to the study of the theory of music, then an hour was employed to put counterpoint to a canto jermo, and in another, again, the master explained verbally the rules of counterpoint, whilst the pupils put them into practice by writing. In a third hour reading was practiced. The rest of the day was spent in practice of the harpsichord, or in the composition of a psalm, motet, or canzonet, according to the capacity of the pupil. On the days when the pupils were allowed to go out, they used to pass through the tower gate, called Angelus, near the Monte Mario, where there is an echo; there they used to sing, whilst the echo returned their errors to their own hearing.’


It is easy to understand why nineteenth century teachers, prominently Garcia II, would want to find a short-cut. Garcia invented the laryngoscope in hopes that direct observation would lead to such short-cuts in vocal study. Redi writes about Garcia:

Garcia the younger was a unique figure in the history of music.
After a brief career as a singer and impresario he entered the teaching profession in which capacity he labored for seventy-five years. Without doubt he faithfully helped to perpetuate during his earliest years as a teacher those principles of Bel Canto he had learned from his father and other predecessors. No evidence has ever been submitted to indicate that he had ever at any time disagreed with the theories and principles of Bel Canto he had learned as a young student. His only cause for dissatisfaction with the old system was the length of time necessary to obtain vocal mastery. In order to find a short cut to vocal mastery Garcia conducted numerous experiments with the purpose of gaining a direct control over the vocal organs themselves uppermost in mind. Thus, it was he who in a lifetime of teaching introduced many theories widely at variance with the practices of the early masters. Many of the ideas of tone production initiated by him are still in vogue today, and the complete disappearance of the old system of training is to a very great extent his responsibility. These errors of commission he partially rectified in a retraction published in the London Musical Herald, August 1894. In this statement Garcia refutes many of the principles he once supported and which, despite his refutation, have through association with his name since gained an almost world-wide acceptance. This is his advice:

 ‘Avoid all these modern theories and stick closely to Nature. I do not believe in teaching by means of sensations of tone. I began with other things; I used to direct the tone in the head, and do peculiar things with the breathing, and so on, but as the years passed by I discarded them as useless, and now speak only of actual things and not mere appearances. I condemn that which is spoken of nowadays, viz., the directing of the voice forward, or back and up. Vibrations come from puffs of air. All control of the breath is lost the moment it is turned into vibrations, and the idea is absurd that a current of air can be thrown against the hard palate for one kind of tone, the soft palate for another, and reflected hither and thither. !




See the following page for a listing of vocal training aspects from the early days of Bel Canto until today.


of Caccini, Tosi and Mancini

  1. Sight singing
  2. Vowel formation
  3. Establish chest and falsetto registers in all voice types
  4. Develop each register as a separate
  5. Blend registers by perfect mutation
  6. Simultaneous development of fol-
    lowing in order of difficulty:

(a)  Solfeggi

(b) Legato

(c)  Portamento

(d) Easy embellishments
(c)  Difficult embellishments

(f)   Messa di voce

(g)  Agility

  1. Emphasis on pure vowel quality

*A11 exerciKi were performed < cippella


Lamperti and Garcia

  1. Breathing
  2. Resonance
  3. Vowel formation
  4. Sostenuto
  5. Legato
  6. Portamento
  7. Messa di voce
  8. Agility
  9. Subject of registers
    touched upon lightly and essentially in agreement •with earlier procedures
  • Gnduil rnniition to accompanied exerciies


LATE 19th & 20th CENTURY
PROCEDURES of Shakespeare,
Schoen-Rene and other moderns

  1. Breathing
  2. Relaxation
  3. Agility
  4. Voice ‘placing’
  5. Support of tone
  6. Singing on the breath
  7. Open throat.

This chart shows the pedagogical changes in voice teaching from the Bel Canto era of the 17th and 18th centuries through the 20th century. Reid continues, “The first departure from Bel Canto procedure, therefore, began when the channels of pedagogical thought were directed away from clear concepts of registration and vowel purity into the mystical realm of ‘sensations of vibration’…

After the “head* voice had received wide acceptance as a suitable name for the falsetto there soon followed a centering of attention on other phases of vibratory sensations. Singing in the ‘head’ voice created the impression that the tone had been ‘brought forward,’ leading, in its turn, to the assumption that it had been ‘placed.’


Stanley was one of the first major researchers to challenge De Reszke’s idea that increasing mask sensations increased resonance. Stanley writes:

Any reader who is too young to have heard any really great singer may say: “I enjoy so and so’s singing; I think that his voice is wonderful”. Let me point out to him that if Caruso, Destinn, Melba, Sammarco, or any one of the really great singers were to come back and appear at the Metropolitan today, the audience would immediately recognize the enormous difference and go wild with enthusiasm. Perhaps the evil had its first big impetus in Paris, where Jean De Reszke started the “singing in the masque” idea which has done such inestimable harm; although the fantastic imaginings of Lillie Lehman have borne much evil fruit. One of the troubles lies in the fact that people are apt to look to the ex-great singer in order to learn how to sing; but he is an ex-singer—he has lost his voice or he would still be singing, and he has lost his voice because he has abused it. He always does harm, since he teaches the very tricks which have destroyed his own voice. A procedure which can destroy a great voice must indeed be bad. After all, when the voice functions properly, no definite sensations can be observed by the singer himself. It is only when it starts to deteriorate that he is able to observe diverse sensations.

These sensations arise because of muscular interferences, and the worse he sings, the more clearly defined do these sensations become. When he sang really well, he just didn’t know how he did it. As each degree of throatiness became more pronounced, the accompanying sensations also became more apparent. Take Jean De Reszke: the natural fault of a French singer, who is losing his voice, is nasality. The throatier he becomes, the more apparent is this sensation of nasality, or of something occurring in the front of his face. This undesirable transmission of sound waves through the masque is directly communicated by tension on the jaw and constriction of the throat. This is the source of Jean De Reszke’s idea of “singing in the masque”. As the voice becomes throatier and throatier, the action of the muscles of the thorax becomes more and more of a blowing process, hence the idea of “breath control”. Need I develop further? Each school of singing has been derived from the sensations of ex-great singers who have lost their voices. Their theories are always one hundred percent wrong.


Stanley presented two very interesting ideas which need to be examined: (1) When the voice is working correctly there are few facial sensations, and (2) sensations in the face are transmitted by constrictor tension in the throat and jaw. (Now most voice teachers know that Stanley was a heretic, but, as Miller says, “There is some truth even in most heresies.”  Reid seems to agree with the first, claiming that “nasal resonance” seems to be something that poor or mediocre singers have and great singers HAVE NOT. Since sensations were a PERIPHERAL issue in Bel Canto pedagogy we have no head count of great singers who claimed to have a lot of facial buzz. Caruso reported feelings in the head for high notes, but also feelings in HIS LEGS.  Sundberg may have something in noting that the amplitude of the radiated voice at the mouth opening is so great as to over-ride almost all other sensations. The jury is still out on Stanley’s first observation, BUT IT DOES SEEM THAT, THE LARGER THE AMPLITUDE OF THE VOICE AT THE MOUTH OPENING, THE LESS AWARE THE SINGER MAY BE OF ANY BODILY SENSATIONS, at least during registration events associated with “voce chiusa.” As regards the second, since constrictor tension in the throat and jaw make the pharynx a poor resonator, the amplitude of the radiated voice at the mouth opening would be considerably less, probably making it possible for the singer to be more aware of facial (mask) vibrations. I have noticed in my own teaching, singing, and adjudication, that many poor singers claim to have wonderful sensations in the mask, and think, therefore that they are singing wonderfully. The “core” issues of vocal resonance have been totally replaced with this “peripheral” issue of facial sensations.


The speaking voice can give good facial buzz, as Dr. Morton Cooper pointed out  in Jerome Hines’ book. This tends to go along with Stanley’s observation, since the pharynx is frequently constricted while we speak. In fact, Vennard says the pharynx is habitually constricted, which is why it feels good to yawn. The amplitude of the radiated spoken voice is much less at the mouth opening, compared to the sung voice. Thus it does not overwhelm other sensations. The facial buzz, therefore, can be easily felt. Some teachers actually cause the student to raise the larynx higher and higher–and these are teachers who seldom even use the word ‘larynx’–unti1 nothing is left but the speaking voice used in a singing manner. The pupil can feel a buzz in the face, therefore the voice must be ‘forward’ and “resonant’, or so they think.  ( Note: Sundberg has explained that the singer’s formant, dependent on the cavity of the ventricle of Morgagni within the larynx, causes such amplitude in

the high partials around 2800 Hz, that these short wave lengths penetrate the walls of the vocal tract like the sounds of noisy neighbors penetrating apartment walls. They then may excite bone or pockets of air within the head and give the singer sensations. The singer’s experience of  “singing in the mask,” may either be De

Reszke’s simple nasality or the singer’s formant. That is the confusion. If singing with a strong singer’s formant is called  “singing in the mask,” it needs to be thoroughly explained to the singer so he/she will not be so astonished later to discover the truth [as Miller was!]]


All of the great teachers in my knowledge, up until the time of De Reszke, agreed that “placement” feelings were the RESULT of good singing, not the cause. Therefore, the many CORE issues of phonation and resonation were taught first, and PERIPHERAL issues like “placement” took their appropriate back seat. Post-DeReszke, 20th century teachers of “nasal resonance” usually think of “placement” as a CORE issue and teach it first, or at least give it priority after breathing. Frequently, the result is high laryngeal singing, and poor, if any vocal progress. The student fails to have a career because he/she cannot sing well, and decides to teach voice! He/She will, of course, teach the same system which caused his/her own failure. Everyone then throws his hands in the air and asks why there are no more great voices in the world! I concur with Reid, and many others, in saying that the concepts of “nasal resonance”, “forward placement”, and “singing in the masque”, have done almost inestimable harm to the cause of beautiful singing!


In Jerome Hines’ book of interviews the baritone, Sherrill Milnes bristled at the idea that placement is an illusion, and offered as proof of its reality the “fact” that placement can be “changed.” That is to say, bodily vibrations can be changed, and presumably resonance patterns changed with them. In discussing head and sinus resonance, he appeared ignorant of all the studies showing that the vibrations in those areas were too weak to add anything to the tone which emerges from the mouth.  So what IS being changed? We can change the shape, and length of the
vocal tract and thereby alter the formants, raising or lowering their frequencies. We can also change the composition of the walls which define the bodies of air which act as resonators.

People who believe in nasal resonance usually think that their resonantal goal is to INCREASE anterior sensations. In order to do that, some singers drop the velum to allow the sound to be nasalized. Many of them draw back the corners of the mouth, as when smiling, thereby shortening the tract, raising all the formant frequencies for the whole voice, possibly mixing some nasality as well. The voice, then, is under-resonated except for the extreme top. The lower partials are weaker for most of the range since pitch fundamentals will be further away from the average first formant frequency.  Amplitude at the lip opening will be less, and the singer gets more awareness of anterior vibrations!  What a price to pay!


Once, while vocalizing with basso Jerome Hines, this was demonstrated to me. At that time, I was still entertaining the idea that more anterior sensations would correspond to a more resonant voice. Hines was experimenting with resonantal adjustments while I was observing. He chose one in which the voice seemed very bright, one in which I would have expected him to have felt many anterior sensations in the masque. “That’s the one”, I said. “Sing it that way.” He grinned from ear to ear, and said, “But listen, if I sing it this way, I can make a much more
resonant sound that will fill the house”. He then chose a resonantal adjustment which resulted in a chiaroscuro tone which was far more resonant, the radiated voice amplitude far greater.

Individual singers may have different sensations as to where bodily vibrations are strongest when resonantal balance has been achieved.

In my own singing, I have noticed that when my voice is working the way which may be described by the audience as “filling the house”, I am not aware of many bodily sensations. Thus, my own empirical observations side with Stanley.


Milnes’ second assumption, that changing of “placement” sensations would result in a change of resonance, may not always be true in every singer. Since the bodily sensations are stronger in many areas, other than the mask, such as the sternum, top of skull, jaw, a singer may simply switch awareness of vibrations and retain the same resonance. Furthermore, since the vocal tract is constricted in a great variety of ways, there are many constrictions possible which will change bodily vibrations.  The vocal tract can change its resonantal composition without changing its length. We have already mentioned that constrictor tension makes the pharynx a poor resonator. Vennard cites the experiments of Russell with lead-tin organ pipes as resonators which gave a mechanical sound of “ah.” By squeezing the pipe during “phonation” the vowel changed to “au.” Vennard concludes, “If it is possible to make a lead-tin pipe say ‘ouch’ merely by squeezing it, not enough to change its shape, can we wonder that muscular tension in a singer’s throat spoils the quality? We should specify that the resonator is still the body of AIR, but that its vibration is conditioned by the CONTAINER which defines it.”

Singers do a great many things to the vocal tract in the course of colorization and characterization, most of which are too complicated to be performed consciously. All of them are likely to affect bodily vibrations. From this we may gather that singers SELECT which vibrations to notice according to their own understanding of the role of auditory feedback in singing. Singers have been TAUGHT to believe they are supposed to pay attention to anterior sensations in the face or posterior sensations in the oral pharynx or head, when, in fact, their bodies are vibrating in at least forty different places!

The matter is not helped by schizophrenia among vocal pedagogues who are also researchers. Vennard, as a researcher knew perfectly well that “sound vibrations cannot be directed at all” (p.94) and that “neither nasal resonance nor sinus resonance has validity” (p.96). But Vennard, the teacher, said that “the sound MUST be forward” (p. 217). Vennard the teacher said, “I often ask my students whose voices are either breathy or swallowed to wrinkle up their noses  and say ‘angry hangman’, making the tone as ‘nasty’ as  possible” (p. 151). 

Indeed, Vennard the teacher outlined an “ideal” 15 minute vocalizing session, as “five minutes for the’ yawn-sigh’…five minutes of carefree scales and arpeggios for range and agility…and five minutes for the ‘a~e’  exercise” (i.e., the ‘nasty, twangy vowel, p. 218).

This is virtually alchemy. Such schizophrenia in researchers/teachers is partly responsible for the confusion regarding placement. “Wrinkling up one’s nose,” shortens the vocal tract from the front end, since it is associated with a reflex action of the vocal tract, characterized by drawing back the corners of the mouth. It raises the frequency of all the formants, and may indeed give the student anterior sensations. This was all part of Vennard’s confused attempts to teach twang articulation to the student, believing it to be the 2800 of laryngeal ring.

But it has nothing at all to do with learning to sing well, much less with learning to efficiently manipulate the vocal tract. This is vocal pedagogy which confuses “CORE” issues with PERIPHERAL issues. The 2800 resonance is the resonance of the larynx itself when it is lowered in such a way as to expand the base of the pharynx and create the “one-sixth” condition. Recall again, that the cross-sectional area of the base of the pharynx must be a minimum of six times greater that the cross-sectional area of the larynx tube in order to get the larynx resonance
called the 2800, singer’s formant.  Therefore, the CORE issue of laryngeal resonance (2800, “ping”) is laryngeal suspension! Exercises designed to improve that efficiency are the exercises of choice. Unless the song one is singing (coloring) contains the words, “angry hangman”, intoned in a very nasty way, there is no need to practice such an intonation! Doing so will accomplish nothing but confusion! Instead of practicing nasals, we should be looking at the effect of the lengthened vocal tract on vowels and thereby find our clues for exercises.

Sundberg says: “The general impression is that the quality of the voice is ‘DARKER’ in singing (than in speaking), SOMEWHAT as it is when a person yawns and speaks at the same time. For example the ‘ee’ sound of the word ‘beat’ is shifted to the umlauted ‘ue’ of the German, ‘fuer’. The short ‘eh’ of ‘head’ moves towards the vowel sound of ‘heard'”.  This is the simple, straight-forward effect of a lengthened  resonating tube. Most vowel formants are stretched out towards their darker sides. Miller says most teachers don’t want to hear the voice sound like it does when a person yawns and speaks at the same time (p. 153 THE STRUCTURE OF SINGING).  But this is not the issue. Sundberg has simply used yawning as an example of an activity of the vocal tract which lowers formant frequencies, much the same as lowering the larynx does. The TIMBRE is obviously different or yawning would be all that was necessary to create a singer’s formant. Sundberg states that simple yawning will not create a singer’s formant.  Miller continues, “But SHOULD vowels have major differences in quality from speech to singing?”  His answer is obviously, “No”. But simple physics says “Yes”, a lengthened resonating tube will lower formant frequencies. Sundberg tells us that the singer’s formant is achieved by a lowering of the larynx or “vowel darkening,” and that “the darkened vowel sounds deviate considerably from what one hears in ordinary speech.”  The only way to keep the sung vowels the same as spoken vowels is to “sing” them with the larynx in the high speaking position. Even a slight lowering of the larynx from inhalation will cause some lowering of formant frequencies. It follows, then, that an aesthetic bias towards singing vowels with the same formants as in speech will lead to a raised laryngeal posture! To be fair to Miller, the key words may be “major differences.” What does that mean?  Richard is objecting to radical impurities in vowels often found in pedagogies which lie outside of the International School. Excessive darkening of the vowels is often found in Nordic areas. But impurities of a different nature are just as often found in the French and English national schools of singing. The slight darkening of the vowel during singing in “voce chiusa” ( an essential aspect of Appoggio) is what Sundberg refers to. As a scientist, he refers to these changes as deviating considerably from speech. A singer or an audience member might not identify the changes as being that radical. Radical darkening is what Miller objects to and rightly so, but individual voice timbres determine the extent of vowel darkening in voce chiusa. What is appropriate for a dark basso profundo like Jerome Hines is not appropriate for a light lyric tenor like Miller.


Berton Coffin states the matter precisely: “…spoken phonetics are NOT (emphasis his) the same as sung phonetics…The statement of ‘pure vowels’ is true and false. It is true if the vowel formation gives a pure and liquid sound; it is false IF SPEECH VALUES ARE FORCED ON MOST SUNG PITCHES” (Coffin, The Relationship of the Breath, Phonation, and Resonance in Singing). Appleman shows the way “sung” vowels “migrate” away from their pure (speaking) formants, and gives extensive exercises to facilitate efficient migration.  These migrations mean that it is impossible for language sounds to “remain defined” (Miller’s goal)  apart from knowledge of their migratory patterns. Most of us manage to learn to “hear” these migratory sounds and interpret them within the sentences being sung. The human ear is very good at that. Small children may have very high formants and other adults very, very low ones. Our ear is able to interpret both as stable vowels.


Sundberg’s studies on professional singers show that the vowel formants of sung vowels lie OUTSIDE of the islands of spoken vowels.  Does that constitute “major differences”? To a scientist, yes. Those same sounds might well lie within acceptable parameters for good singing.

Just how difficult is it for the audience to learn to hear these stretched out vowels with recognition! Probably not very difficult in low voices, although higher voices, especially sopranos, surely present a problem. It is difficult to understand most sopranos, even those with excellent diction because of the pitch range vowels are subjected to.


We are left with the fact that the great singers of the Second Golden Age of Singing do not sound like the “model” singers produced in 20th century studios which teach “nasal resonance”. In my own “fach”, the auditory impressions of the great baritones of this era–Ruffo, Stracciari, Granforte, Merrill, Bastianini, Warren, McNeil, et. a1.–indicate a lowered larynx, and a rich, dark timbre which also has clarity. Ruffo is the only one who did not sing long and he ruined his natural voice by poor technique, by his own admission. Part of that poor technique was an attempt at nasal resonance! All of the others except Bastianini, who died of cancer, sang well into their advanced years. Merrill sang almost until his death.


The La Scala baritone, Cesare Bardelli gave his students a maxim, “Always approach the vowel from the dark side.”  And he was Italian, taught by Italians, going back to Garcia! Certainly, Bel Canto pedagogy attempted to control laryngeal functions by vowel quality. In chiaroscuro tone, the vowel is stretched out towards its darker side, just as Sundberg says. In most human beings, the singer’s formant, containing the 2800 laryngeal resonance (“ping”), cannot exist without the low larynx, which lengthens the vocal tract and lowers vowel formants towards their DARKER sides. This is the paradox at the CORE of resonation. Effective vocal pedagogy must utilize that paradox. Instead of giving exercises on “nasty”, twangy nasals, one COULD use darkened vowels  combined with laryngeal suspension exercises to help the student develop and experience this paradox. During such time, bodily vibrations may be used for some biofeedback, but usage must be purged from association with TRADITIONAL concepts of voice placement!

At least since De Reszke, many teachers have used placement illusion as a means towards resonantal balance. If that works, fine. But many times the singer gets very confused. Nasality becomes resonance. The location of the singer’s formant is thought to be in the face. The goal is thought to be to increase anterior sensations in the face, etc.


Ingo Titze writes clearly that nasality in vowels is not the same as “vocal ring” caused by the singer’s formant. “Nasality may bring about a perceived increase in brightness of quality, similar to the vocal ring. The mechanisms are totally different, however. In the one (nasality) we lower the low-frequency prominence of the oral and pharyngeal resonance by creating AN ACOUSTIC LEAKAGE THROUGH THE NOSE; in the other (vocal ring), we raise the high frequency prominence in the oral and pharyngeal resonance by CREATING AN ADDITIONAL RESONATOR IN THE LARYNX”(VOCAL HEALTH AND SCIENCE, NATS, p.245).   Why does Titze seem to support, then, the usage of nasals in vocal pedagogy? He says: “Teachers may use nasal sounds to TRICK the student into experiencing some of the key sensations in the facial region, but then quickly point out the alternate (and preferred) ways of obtaining those same sensations. Otherwise, the singer may learn to interpret excessive nasality (twang) as vocal ring”(IBID). 

What a tangled mess! Why not concentrate on laryngeal suspension in order to create the “additional resonator in the larynx”? It is not just the student who gets confused by the usage of nasals. The teacher does also! Vennard himself confused “twang” with “ring,” just as Titze warned. Vennard called “twang” a necessary “ingredient” in the voice, like “salt in a cake,” and thought that twanging vowels brought about vocal ring. Furthermore, the sensations in the masque area during nasality are totally different from the sensations of singing with the singer’s formant.


Sundberg points out that it may well be easier and more productive in vocal pedagogy to explain to the student the actual acoustics of the singing voice than to use these “tricks” that Titze mentions.  Sundberg’s point is well taken, I believe. Shouldn’t voice teachers use facts and images based on facts? Even if we can “trick” a student into good resonantal adjustment using placement sensations, shouldn’t we then explain the real acoustical event? In fairness, I think that Dr.Titze seems to say yes to that.  He also agrees that placement sensations are the result of VOCAL TRACT shaping. The difficulty of “organizing” and “systematizing” these sensations for general usage in teaching is also apparent from Titze’s article on placement in the book, VOCAL HEALTH AND SCIENCE. While it MAY be true that singers’ impressions of placement may correspond to various pressure points within the vocal tract it is also true that there is no way to categorize or describe such sensations so accurately that they become generically usable in pedagogy. Too


Fig. 56. The usual types of placing.




many singers have very different sensations even when the vocal tract shaping  appears to be similar. And too little is still known about this phenomenon. “Due to the presence of multiple formants, there may in fact be several such locations where vibrotactile sensation is maximized” (Titze, Ibid. p.247).



I have excised Husler’s long discussion of placement because it is just too fanciful. When he claims to be able to identify a particular vocal tract shaping that isolates the ary-vocalis (which doesn’t exist!), he has invalidated his teaching. However, the principle involved in his teaching appears to have merit, namely that radical reflex

postures of the vocal tract cause changes in the larynx that are not hygienic. In other words, some placement sensations if associated with certain “schools of placement” may actually cause the larynx to work out of balance. I will just include here his basic scheme of possible placement positions and his conclusions about over-working the larynx. There is a little truth in every heresy.

Laryngologists DO see changes in the larynx during vocal tract positional changes (i.e. “vowels”). The vocal tract and the larynx DO have reciprocal sensitivity. If the vowel is impure (i.e. the vocal tract shaping is incorrect) the larynx pays the price by strain. The principle is sound.


To look at this illusion of placement more clearly and the tragic results that can occur if the phenomenon is misunderstood, let me describe an international baritone to you. I will not use his real name, obviously. Let’s call him “Rick Marshall.” Rick Marshall was a good Midwestern farm boy who made it big. He went to a big name college near Chicago where he had a big, rich baritonal voice with lots of color. Recordings survive of his singing during this era. One such recording of Brahms’ “Vier Ernste Gesaenge” shows a dark, rich tone.   He got chosen to be an “apprentice artist” and started studying to be an opera singer. He was told he must learn to sing “in the mask,” and began changing his technique. The timbre of the voice changed just as we would expect. The rich, dark quality dropped out, and instead he had a snarl and a bite all through the voice. He took out smooth registration and added the “hook” into the upper register because it emphasized the snarl and bite even more. He soon found people of influence and money who backed his career. He toured with a famous opera troupe, then finally got hired at New York City Opera. By this time he was convinced that “singing in the mask” was what had gotten him there. He believed in placement quite literally and thought the “feelings” he had in the head were evidences that he was producing maximum resonance. He got hired at the Metropolitan Opera. He was handsome, tall, physically strong, and eager to cooperate with directors. He had it all. He made his first recording, and what do you suppose the tone showed? What Husler calls, “a narrow voice with an over-metallic timbre or a nasal twang.”  It also showed consistent intonation problems. Some tones were a quarter of a step flat. Still, he was very strong and many people didn’t pay attention to the intonation problems. They were so taken by the idea that a baritone could sing such bright high notes. A few people commented that he lacked the beauty of tone of Robert Merrill, or Leonard Warren, but most people thought he had everything. Each year the intonation problems got worse. Now and then tones were one-half step flat. The quality became less and less distinguished. High notes became more effortful. Still, he continued to “sing in the mask.” Finally his singing deteriorated so that the tone was unrecognizable as the young college student who recorded “Vier Ernste Gesaenge” or even as the singer who first went to the MET. His frequent vocal fold injuries and operations may well be linked to his articulation. The college student was far better! He still has his fame and money but he has lost his voice. What happened?   Husler would explain it by saying: continually singing at number 3a (“in the mask”) leads to “a chronic over-accentuation” of the vocal is muscles.  The stretching muscles (i.e. the cricothyroids and probably sternothyroids) had lost tonicity, “the suspensory mechanism collapsed, and the space in the larynx was reduced.”  Atrophy of the stretching muscles first showed up as intonation problems since they were inaccurate in their stretch for pitch.


In the classic Bel Canto system, “appoggio” exercises are viewed as the best way to maintain tonicity in the laryngeal system. From this perspective. Rick Marshall lost his voice by loss of appoggio, part of which may have involved a misunderstanding of “impostazione” (placement).

 Let me reiterate some of Husler’s conclusions which appear to be true:

“Proper singing consists of a widespread cyclical process which is destroyed by any kind of one-sided specialization. The ‘placing of the voice’ changes according to the volume of sound or the pitch required, varies slightly for the different vowels, and should be altered according to the style of music…If the singer practices one particular ‘placing’ exclusively he over-accentuates its corresponding muscle-system, thereby causing a specialization in the vocal organ that eventually damages it.”  Even if we prefer to say that proper “appoggio” induces this “wide-spread cyclical process,” without further reference to placement, Husler’s conclusion deserves full scientific examination. Even now, as a tentative statement, it should be written on the doorpost of all 20th century studios, especially those believing that good singing means “singing in the masque”. The time has come to utilize a holistic approach to resonantal adjustment, namely, classic Appoggio. We may quote both Miller and Vennard as appearing to be in agreement. Miller has written: “In the historic Italian School resonator coupling permits sensations in all parts of the vocal tract. The resonance balance (placement) relies neither on the pharynx nor on the mouth as chief resonator, but on a combination of both. Sensation centers neither in the throat nor in the face.”

Vennard agrees when he says, “Thus we see that a good tone will seem to be anywhere except in the larynx. The more mellow it is, the more there will be a feeling that it goes ‘down’ and ‘back.’ The more brilliant it is, the more it will seem to go ‘up’ and forward.’ THE IMPORTANT POINT TO REMEMBER IS THAT IT MUST GO IN BOTH DIRECTIONS AT ONCE.”