Kashu-do (歌手道): The Nose Should Be In the Voice but the Voice Must Not Be In the Nose

I had heard this seemingly paradoxical axiom, years ago, and back then it made no sense. With time I understood there was a difference between nasal singing and the sensation of efficiency that produces a feeling of vibratory intensity in the region of the face, commonly called the mask. Yet the exact function differences that create the two results only recently became completely clear. The unsatisfying quality of nasal singing is often confused with equally irritating quality of pressed phonation. Both sounds seem to suppress low partials in favor of higher ones. At least that is the illusion from hearing.

The functions are however the opposite. A truly nasal production seems to be an attempt at remedying a lack of brilliance (squillo) in the sound that normally would occur when phonation is balanced and the vocal tract is adequately adjusted. There are three different types of fold posture/laryngeal depth possibilities: the balanced set up that produces true squillo and two others that produce an illusion of brilliance.

1. Balanced Phonation: We have already discussed the mechanism of phonation here. I wish to remind that phonation is at its simplest a two-dimensional occurrence driven by pitch. Pitch is essentially driven by the length of time it takes for one vibratory cycle to occur, which in turns determines what the listener perceives as a given pith in vibratory cycles/per minute (Hertz). Ideal fold posture includes a specific fold depth that requires ideal closure whereby the vocal folds close in such a way that breath is neither wasted nor held back (the latter would cause excessive sub-glottal pressure). Some scientists suggest that a 50% close quotient produces the most efficient results. Much higher close quotients have been recorded as acceptable among more robust voices. Indeed the distance between optimum vocal production and what is considered professionally acceptable is considerable.

2. Pressed Phonation: Pressed phonation (over-adduction of the vocal folds) occurs when the fold depth is insufficient. Too little fold depth speeds up the rate of vibrations, which means pitch would increase. To prevent a rise in pitch, the folds press together to slow down the opening of the folds. In this case the folds remain closed for long (higher close quotient) and the amount of air released is less. The result is less sound with a discernibly thin quality. It is important to remember that the shallow fold posture is not a random occurrence but rather a result of the singer’s conception of his/her own sound. Muscular coordination is a result of desire. In other words, the singer produces his/her own dysfunction without being conscious of it. When the singer then discovers that the sound is dysfunctional (whether because others do not like it or s/he begins to experience discomfort), the muscular balance had become so ingrained that spontaneous correction is usually not possible. In most cases of imbalance, a certain amount of time is required to achieve balance.

3. Nasal production: Nasal production is more complex because it is an attempt at remedying a perceive imbalance, namely a deficit in brilliance (i.e. a lack of strength in the high partials due to loose phonation). In such cases, the singer experiences a type of brightness that has no foundation in balance, but feels much more satisfying than the dullness of an unaltered breathy phonation. This kind of production gives the singer a false sense of correctness, because most of us have been told that the sound needs to be bright and forward. As I explain above, when an imbalance in the phonation process has been produced for a long time, spontaneous correction is usually not possible. However a spontaneous pacifier is possible and that is nasality. To the singer’s ear, the tone is brighter and feels more “forward” because of the nasal vibration, which the singer, who has not experienced the intense frontal sensations that accompany proper phonation, mistakes for mask resonance.

Another element in this process is laryngeal height. A high larynx is most often a result of pressed phonation, which produces increased sub-glottal pressure that the laryngeal depressors cannot sustain. However an imbalance between laryngeal depressors and levitators can produce a high larynx in the case of balanced phonation or even loose phonation. This is to say that nasality is not dependent upon a high larynx.

The remedy for nasality is better glottal closure because nasality is usually introduced to make up for the lack of brilliance that would be a normal result of balanced phonation. To induce better closure it is common knowledge that nasal consonants like [m], [n] or “ng” help bring the vocal folds closure together. A nasal vowel, for some reason does not. My guess in this case is that the nasal consonants cause an occlusion of the vocal tract, which has proven to influence greater glottal efficiency. To have an influence on phonation, it makes sense that the consonants that occlude the vocal tract must be continuants. For that reason, voiced fricatives like [v] and [z] have proven particularly effective in increasing efficiency of phonation. In the case of the nasal continuants, it would seem that their efficiency factor trumps whatever ill effects might be otherwise produced by nasality. It should be clear therefore that nasal continuants are not used for their nasality but rather for the positive effect of their occlusion. In this way, we can understand why nasal vowels do not help phonation.

It is also my observation that nasality is more common with vowels that are prone to tongue tension. When there is glottal inefficiency (either pressed or loose phonation), the tongue is often recruited (pulled back) to help reduce the resulting instability in the larynx. Because the vowel [i] is produced with a considerably elevated back-of-the tongue, it has been very helpful in inducing more efficient phonation. In other words, the muscular action innate to the production of the [i] vowel counters the retraction of the tongue that is associated with inefficiency. I should caution that the [i] vowel is not a full-proof tool. The statement above regarding a requirement of time to change imbalance is important. The [i] vowel may produce the quickest difference in the direction of efficiency, but it does not guarantee efficiency. That is why a singer can sing a very good quality scale on the [i] vowel and has a terrible time doing the same on [a]. Nevertheless, the [i] vowel over time can stabilize efficiency such that production of other vowels also become more efficient. The [u] is particularly prone to inefficiency. When dealing with breathy phonation, I would employ vowels in the sequence [i], [a], [u].

© 11/07/2009


6 thoughts on “Kashu-do (歌手道): The Nose Should Be In the Voice but the Voice Must Not Be In the Nose

  1. I think I always heard this axiom the other way around. The tone should be in the nose, but the nose should not be in the tone. As in, you should feel the voice vibrating in the nasal passages but you should not hear nasality in the voice. Either way, your point is correct! This is one of those classic cases where proprioception seems to conflict so much with scientific reality. Caruso swore that when he sang the sound invaded the nose and sinuses, and was so convinced that when x-ray imaging was done of him singing with a completely sealed naso-pharyngeal port he refused to let it be published with his name attached to it.

  2. Thank you Miriam for the very kind comments. I hope to get to meet you in person.


    I was wondering if someone might catch this. This is similar to the post I wrote called: Si parla come si canta! I have heard it said both ways, which makes for even more confusion. The meaning in both cases is clear, that we are trying to avoid nasality. Thanks for mentioning the Caruso fiasco. Fascinating isn't it!

  3. This is the second time I see the 50% number on closed quotient, and I'm curious. Can you give a little more information on this? From what I can tell, especially in great tenors, especially on high notes, the closed quotient is estimated to often be in the 70's. I am always skeptical when theory doesn't match what actually happens. Why do you think this is? Pavarotti cannot have been employing pressed phonation, can he?

    Also, in your experience, can closed quotient change over time with exercise? Does a habitually “headier” production lower it? Does proper appoggio affect it?

    Finally, recent research suggests that the voice doesn't function on a source-filter theory basis, but that the configuration of the vocal tract during phonation may have a sort of “feedback” effect and alter the vibratory pattern and closed quotient of the vocal folds themselves. Perhaps this can help explain the lure of nasility with further investigation. I personally have found that nasality isn't only an aesthetic choice, but that it can feel easier to sing, too. (Don't get me wrong–I agree that it is undesirable, I'm just proposing other reasons for its prevalence.)

    I recently read a dissertation in which a majority of professional tenors surveyed were found to employ a (very small) amount of nasality in the passaggio. Again, a mismatch between theory and apparent fact. Why do you think this may be?

    At any rate, thanks for a very interesting post.

    Klaus Georg

  4. Dear Klaus,

    Thank you for another well thought out comment.

    1) My post admits there is a substantial divide between the ideal (50%) that Sunberg's team proposes and the observations (70%) recorded by the likes of my dear colleague, Don Miller with the Groenigen team. And yet I believe that Pavarotti at times, particularly in the extreme top range did press. This explains his preference for for the narrower vowels in the Bb-C area whereas other great tenors like Gedda, Bjoerling, Wunderlich, Corelli, Gigli, etc achieved a fuller top with the possibility of more open vowels.

    My support of the Sunberg model comes from helping singer's develop ease in the top voice. My experience in most cases is that they undersing (sing a thin production)requiring more medial closure. Singing a fuller sound (i.e. greater fold depth), increases vibration time which then releases the excess medial pressure (without this the pitch would go flat when they sing fuller). I bet that Pavarotti's phonation rarely approaches 70%, perhaps in the extreme top, but I am sure it is 60 or slightly above. Singing a high Bb or B or C is not the easiest thing to learn. It has to be developed. So any tenor who develops the ability to consistently sing those notes will tend to be successful. It is not surprising that the majority of tenors out there are not so close to what might be ideal.

    And yes the CQ can be changed with exercise. I am proof. When I was singing bass-baritone, Don Miller recorded my CQ well over 70% around F#4. What I discovered is that I was singing too thin and masking it by rounding the sound. The last CQs I recorded of myself before my computer and analysis equipment was stolen shows CQs of 58% around A4 and in between 58 and 64 on Bb4. That is a big change. My F# was in the 50%.

    Now what do we mean by headier? If by headier we mean a feeling of release, I would say yes, this would lower the CQ. That feeling of release, which I get more in more in the Bb-B4 range (C4 and above not yet consistently so for me) is accompanied by greater vocalis participation creating a fuller sound. Some would call this more chest voice. What I have attempted to describe here is that the “feeling” of release called Head Voice is achieved by accomplishing correct fold depth/closure. For the top range, the necessary correction is usually in the direction of greater fold mass because the average singing sings a thin sound on top. (continue in other comment)


  5. Dear Klaus (continuing),

    As for the source-filter theory, you are correct. My feeling is that new findings do not take away the fact that overtones are suppressed or encourage based on the acoustic nature of the vocal tract. I believe that modern developments show that the glottis and the vocal tract have an “inter-dependent” relationship, that they effect each other rather than the one-sided idea that the the vocal tract is only a filter. What I write here regarding supra-glottal inertia is in recognition of the acoustics having a direct effect on the glottal vibration.

    As for tenors using a little nasality in the passaggio, it is a well known practice. It is a strategy for dealing with the the un-dynamic coordination of the TA-CT in that region. Phonation tends to go from too thick (and loose) to too thin (and pressed). To remedy this, tenors allow the upper passaggio to be a little looser and correct that approach by being a little nasal. This in fact does promote better breath flow in the top range (i.e. induce a slightly thicker production resulting in appropriate fold depth/closure). It is nevertheless not the ideal way to go. The slight looseness in the passaggio tends to promote cracking in the high range if the singer has to jump from the passaggio to the top. That is why Pavarotti suggests narrow in the passaggio. Pavarotti's suggestion is practical because singers tend to sing a little too heavily in the lower voice. Therefore narrowing the passaggio corrects the heavy production. However, if a singer sings a balanced tone in the lowe end, there will be neither a need to narrow nor a need to loosen and nasalize.

    Thanks again for a superbly thought-provoking comment.


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.