Sunday, July 27, 2008

Why falsetto (flute voice) is important in vocal pedagogy: An issue of muscular balance

The title of this article is obviously paradoxical. Falsetto or flute voice are equivalent states in the male and female voice respectively in which the Crico-thyroid muscle group is active without a balancing antagonism from the vocalis muscle group. So why then do I call this an issue of balance?

It is my experience in the studio, particularly in recent years, that the ability to produce a falsetto is dependent upon achieving the "at-rest fold length" that is conducive to the most efficient use of the vocal apparatus. What do I mean by this? The answer will take a historical as well as a theoretical scientific perspective.

There was a time that the development of a classical singer's voice could be compared to the development of the physical skills of an Olympic athlete. The level of proficiency of a singer was not solely based on what range of pitches the voice is able to execute with relative ease, but also the sound quality of each note. We read from the writings of many of the great Italian teachers that vocal training could last 8 to 10 years of daily exercise before the singer was allowed to make a professional debut. Contrary to our time when expedience is the primary concern, singers often develop superficial skills that meet with a quick demise commensurate with the training time. With the knowledge and equipment we have available to us in the 21st century, it is conceivable that the training time could be cut considerably without loss of quality. Unfortunately very few teachers use the knowledge and equipment that science has made available.

The issue at hand is quality. It is crucial to understand that the specialization of voice types has reached such extremes that a classical singer's skills can no longer be compared favorably with the ideals of the Olympic athlete. It must be said that the modern operatic soprano, other than the the dramatic coloratura rarely develops a complete range. If we look at the works of Rossini and Mozart and consider the types of singers that they composed for, we will find that modern categorization makes no sense. Were Malibran and Colbran Sopranos or Contraltos? Was Mozart's Constanze or Queen of the Night a light voice coloratura or an anomaly in the Sutherland mold? Are the voices we consider unusual like Callas, Sutherland and others truly the exception or should they be the rule (in certain respects)? Every soprano I have taught in the past few years is capable of at least a three octave range, accessing regularly the Queen of the Night´s High F. This does not mean that every one should sing coloratura roles. It only means that when the soprano voice is fully developed those extreme high notes are accessible and this phenomenon is a sign of healthy flexibility and adequate strength.

This brings us to a discussion of registers. The terms falsetto and flute voice have been associated with "fake" sounds in modern time. It is important to know that the term falsetto originally referred to an uncoordinated, breathy production that tended to manifest in the male passaggio. This uncoordinated passaggio eventually took on a new name, "voce finta" and equally pejorative term meaning "feigned" or faked voice. Consequently, the term falsetto was used more and more in reference to the light quality sound in the high range that is devoid of vocalis muscle activity, an uncoordinated sound in a sense.

So why then would I consider uncoordinated sounds to be of benefit in vocal pedagogy?

First of all, we must understand that not everyone can produce a falsetto or a flute voice, and that this inability is a sign of muscular imbalance. This is a question that is not asked enough: Why can some singers not produce a falsetto or flute voice?

Second, falsetto is not inherently breathy if we consider the natural tendency for full closure in the higher falsetto or flute range. This full closure phenomenon tells us at least that at high pitch levels the falsetto achieves full closure quite naturally (as fundamental frequency increases, activity in the external thyro-arytenoid increasingly adjusts the folds medially). Furthermore, once full closure is achieved, there is a build-up of sub-glottal pressure, which induces a vocalis response. By definition, the production is no longer falsetto but modal (an antagonistic relationship is established between Crico-Thyroid muscle group and the Vocalis muscle group). Therefore, is it not logical to develop the coordination of the "light head voice" or "full-closure falsetto"? From such a production, as pitch level decreases, vocalis activity increases. If such a coordination is developed dynamically (as opposed to abruptly), the muscular dynamics necessary for an easy high range would be achieved. Why not follow this logic?

As previously said, not every singer can produce a falsetto and this is a sign of malfunction, at very least weakness in Crico-Thyroid and lateral Thyro-arytenoid muscles. These muscles mainly are responsible for the lengthening of the vocal folds. It could also mean that the vocalis is relatively muscle-bound, so rigid as to resist Crico-thyroid activity at a given pitch range. The ideal at any pitch level should be a possibility of variable interaction between the two muscle groups in question; that one is not limited to a single state on any given pitch. If the vocalis is so inflexible as not to allow falsetto, this must be seen as a relative dysfunction. This lack of falsetto ability is observable in many tenors referred to as possessing "robust" voices, and often in baritones who train as basses, and natural sopranos who train as mezzos. There is indeed a relationship between "pushed" down voices and inflexibility of the vocalis.

If we take merely a linear consideration between Vocalis and Crico-thyroid the problem could be more easily understood. At a given pitch level, fundamental frequency can be maintained at variable fold length-thickness-tension. If the shortest fold length is selected (i.e. greatest amount of vocalis tension) for a given pitch, the result is also the loudest, most pressurized possible result. One could conceive of gradually increasing CT activity and gradually decreasing vocalis with the rise in pitch. However, the entire strategy is based on vocalis hyper-function and CT hypo-function. There will indubitably develop excessive vocalis strength and inadequate CT strength over time, resulting in an inability of the CT in a CT-dominant posture to withstand the kind of pressure that can be endured in a vocalis-dominant scenario. At the extreme, second formant dominance which coincides with CT dominance in the male voice becomes impossible. Additionally, the hyper-function of vocalis over time curtails the possibility of complete vocalis passivity which is necessary for falsetto. And even if vocalis passivity were possible, the underdevelopment of the CT muscles also precludes the possibility of the kind of fold lengthening necessary for falsetto.

From this premise, the ideal fold length is the midpoint between maximum vocalis activity and maximum CT activity for a given fundamental frequency (pitch). The true timbre of the instrument would be identified when such a state is achieved and all other muscles are functioning adequately to produce unforced full closure. Strength in both muscle groups in balance can then be increased by gradually increasing volume (breath pressure). What we observe in "induced" or fabricated robust voices by and large is the scenario described above, whereby vocalis activity is increased to bulk up the folds, which in turn induces an increase in sub-glottal pressure. That induced robust quality trains a gradual hyperactivity in the vocalis group, which then throws the ideal balance of the muscles off. Lyric tenors singing dramatic repertoire is the plague of our time. Dramatic baritones singing bass, or dramatic sopranos singing mezzo is just as rampant. I have in fact observed many coloratura sopranos who were trained as mezzos. More personally, I believe that my long career as a baritone is an example of this gradual destabilization of the instrument. Developing and maintaining a healthy falsetto range is one of the necessary characteristics of healthy vocal production. I had a very healthy falsetto in my youth, even singing some counter-tenor. I had all but lost the ability during the last two years of my baritone career. Now not only do I have it back, It is nearly an entire octave higher than before.

Theoretically, there is a finite maximum fold mass and therefore a low pitch limit for any given voice. However, there is an infinite possibility of mass reduction and therefore an infinite possibility of high pitch production, "theoretically". The following singer is not a freak, but simply someone who probably had unconsciously developed a facility in the falsetto range which got him attention. As a result he continued to develop this ability:

The singer sang nearly an octave above his previous record, which proves that coordination improves with training. His final note was C#8, two octaves above the soprano high C. The coordination is referred to as damping, whereby only a portion of the vocal folds is vibrating (this production is also called whistle voice). In essence, the smaller the vibrating portion, the higher the singer can sing. The vibrating portion can be made infinitely smaller and therefore, the pitch level can be theoretically infinitely higher.

Many respected teachers and scientist take the stand that falsetto does not help in vocal training. I disagree for the reasons given above. I believe the problem with the assessment of scientists in this specific case stems from an over-concentration on acoustic factors. Acoustics are crucially important once the phonation is of a high quality, but understanding how to bring the instrument into muscular balance is where science is lagging. It is virtually impossible to get professional singers to submit to laryngeal electro-myography. The idea of inserting needle-electrodes in the laryngeal muscles scare singers to death. Therefore, empirical data cannot be gathered relative to the function of those tiny muscles at work. Additionally, there is a prejudice in the field of vocal pedagogy against the necessity of long-term training to achieve superlative results. Teachers and singers and scientists by association, by and large believe that a singer's abilities can only be improved to a certain point, that someone is specifically vocally gifted or not. This kind of limitation discourages the thought that a voice that appears imbalanced and does not response to superficial remedial exercises could improve to a professional level. If this is accepted, then there is no real reason to study muscular dynamics, for it is assumed that the coordinated singer is gifted and the less coordinated one has to chance of improvement in that regard.

Consequently, the research has focused more on acoustics because we have immediate control of resonance factors which are principally vowel production and vowel modification. Acoustic research is yields very immediate results, the equipment is graphic and flashy and the layman can follow on a superficial level. Such research is much more easily funded and the information gets a lot of mileage.

Although we have less empirical data about the behavior of the intrinsic laryngeal muscles with regards to the professional voice, we can however extrapolate logical expectations from what we know about the basic function of this musculature. For my part, I have based my teaching on these premises and indeed my own retraining as a tenor is based upon expectations of how the mechanism would respond as muscular dynamics change.

Verdi Di Provenza.mp3



How I sang as a baritone and what I do here in the Strauss Cäcilie practice clip are worlds apart. There is even a significant quality difference between the Cäcilie clip and the Deposuit clip that I posted here one month ago. For better listening pleasure, I will end this issue with a clip of Leonora's Tacea la notte placida sung by Mexican Soprano Jessica García Gonzales, a current student who sang as a mezzo-soprano until April of this year.


© 07/27/2008


Martin Berggren said...

What you write in this article strongly resonates (!) with me. The vocal tract is acoustically very short, compared to wind instruments, so the role of the vocal tract resonator, although important, cannot be compared with the role of a trumpet, for instance. (The exception perhaps being the very high soprano register or overtone singing.)

The muscular actions in the larynx associated with the generation of sound has felt almost completely like a mystery to me, so I have recently tried to catch up on this issue by reading and experimenting. There is actually some research done on so-called "vocal fold posturing". I have, for instance, recently browsed Titze and Hunter's article "A two-dimensional biomechanical model of vocal fold posturing" (Journal of the Acoustical Society of America 121 (4), 2007, pp 2254-2260) where they attempt the ambitious goal of modeling the movement of the muscles affecting the vocal folds with measured electromyographic signals as inputs. So existing research does not only cover vocal tract acoustics!

Personally, I think I am finally able to feel the antagonistic actions between stretching of the vocal folds (CT, I guess) and the opposing length reduction (vocalis portion of the TA muscle). It is actually like a revolution to actually be able to distinguish these motions!

Finally, I believe it is very brave of you to expose your development to the world the way you do! It is also inspirational for us that try to develop our voices, although it was a long time since we were teenagers...

Martin Berggren

Toreadorssong's Vocal Technique Blog said...

Hello Martin,

Thanks for your commentary. There have been several studies done on the two-dimensional model and Titze has been on the forefront of using physics and fluid dynamics to understand how the vocal tissue would respond. I am still boning up on my physics in order to understand completely what he writes. What I meant is that acoustic research is so much further along the way. We can take a lot of information for granted acoustically. Muscularly we are comparatively behind. I was very busy in 2007 and was not aware of Titze's 2007 article and have not had access to it yet. Guess what my next reading assignment is?

As for exposing my own development, I believe in putting one's money where his mouth is. I believe that much can be learned by demystifying the process. Singers do not like to expose their less than perfect sounds to the public, but one thing I am certain of is that no real vocal progress is achieved without some difficult moments of imbalance. We need to know the in-between stages of development. And I have recorded myself throughout the process. There are some things I have not put out earlier on because I don't believe anyone would respond well to them. But when I have tamed this beast, I will bring them out. People are more willing to accept listening to in-between stages when they hear the final results. What I post here is promising enough that I don't think I am in danger of undoing my reputation.

My principle message here is that an unbalanced voice can be brought to ideal balance and can sound pretty amazing in the right repertoire.

Many great singers are kept out of the game because their true voices are hidden. The future of the art depends on relieving the frustration of such interesting artists rather than banking on the superficial product of some singers who are naturally coordinated. Singing is first and foremost a musical art form. Too many well-coordinated singers are not musicians and that belittles the art form. Also too many excellent musicians are vocally uncoordinated. That can be remedied, and that is my goal ultimately.

Keep the wonderful interaction coming. I enjoy your well-informed commentary.

Adam said...

I agree wholeheartedly about the importance of the falsetto to vocal training. I wonder, though, about the goal of achieving the "midpoint between maximum vocalis activity and maximum CT activity" for each pitch. What if the goal were to achieve only the minimum necessary vocalis activity (that required for full fold closure) for each combination of pitch and dynamic level, as a sort of "brace" against the pull of the CT? That would seem to me to allow great flexibility, ease of production, etc. Assuming, of course, that the singer has sufficiently developed CT function - which is where the falsetto comes into play. What do you think? Is there a scientific reason for seeking a "midpoint"?

By the way, your Caecilie clip shows some exciting improvement. The top of the voice is really starting to "spin". It's very inspiring to witness your transition in a public forum like this.

Thanks for another informative and intriguing post.

Toreadorssong's Vocal Technique Blog said...

Dear Adam,

First of all, thanks for the supportive comments. The tenor thing has reached that stage whereby I am over the hump. Every day is easier than the previous and for me the greatest thrill is that I have a sustained high B every day. All my tenor friends told me that the B is another register all together, but I have a hard time believing this because all the greats do not sound like they go to another register there. In other words it is a matter of building strength in the CT over time while maintaining the other elements in balance. That means the C is around the corner. I must not stop now but continue to practice to develop flexibility and higher notes. Soft singing up to High A is coming in. Another good sign.

As for your very logical deduction about minimum vocalis activity, I feel that it makes sense if one is only concerned about the top of the voice. Under-training either of the muscle groups will cause problems in the passaggio and in the case of your suggestion eventually weakness in the lower range. For any given note it is a balance that must be found. The ideal balance requires activity from both muscle groups that makes transition to the next adjacent note as natural as possible. If one note has an overactive CT, then the next lower note may end up being weak, and with time the vocalis weakens and the CT overactive, but also weakened because of the lack of counteraction from vocalis. This is more an ideological conclusion. On any given note we must have choices. Sing a note mezzo forte requires the balance I suggest, while singing it piano may require the one you suggest. The middle ground however seems the point of balance that we must routinely exercise so we have access to colors on both sides (lighter and heavier) of that ideal.

CD said...

This is a very informative site. I'm glad I stumbled upon it.

Martin Berggren said...

I will, if I may, make an update of my own late-night experimentations with the "flute voice", inspired by your blog. As I said in a previous comment, I believe I have learned to feel the antagonism between stretching of the chords and the thickening, the latter being relaxed in the "flute voice" (falsetto). By tuning that balance, I am now able (sometimes at least) to crescendo to full voice from falsetto, say on a passagio F4, something I did not think I would be able to do!

What I have noticed is that I can achieve a different kind of phonation compared to my normal one by working from the flute voice, a phonation that feels like it operates at a lower static lung pressure but with a more energetic flow of air. To get this feeling (easiest in a confortable range, say B3), I start with a quiet "flute voice" and crescendo into forte, making sure that the lung pressure gently increases and the flow of air continues. If this goes well I can feel that when the loudness increases, there is an acoustical response in the pharynx (and also in the chest) that seems associated with overtones forming. The feeling is different from my normal phonation (that uses a higher lung pressure together with more resistance from the vocal folds I think), in that the overtones seems milder and more "wood-like" compared to my normal sound. This is still quite difficult for me to control, I must add.

I am still not sure whether this is a good phonation approach; maybe I am overdoing it and being to loose, but I cannot help but relate these feelings to what I have read in a quite recent paper by Titze, where he summarizes and extends his research on nonlinear coupling between the vocal fold vibration and the vocal tract (Nonlinear source--filter coupling in phonation: Theory, J. Acoust. Soc. Amer. 123:5 (2008)). The subject of the article is how the vocal tract loading actually can alter the waveform that the voice produces. This is a nonlinear effect, different from but related to the effect of lowering the phonation threshold that an vocal tract inertial loading can achieve.

Titze makes, I believe, an important claim towards the end of the article: the level of possible interaction with the vocal tract depends on the type of vibration pattern (or "mode") in the vocal folds. I will try to summarize the findings below.

An vibration pattern associated with a chest voice (a surface wave propagating upwards involving deep layers of he folds) does not interact strongly with the vocal tract in its production. That is, the spectrum produced by the vocal folds is pretty much given no matter what, and the vocal tract just filters the sound, producing the vowel sounds and the squillo.

In contrast, a thinner type of vibration pattern of the vocal folds, involving only the upper portion of the folds, interacts strongly with the vocal tract loading. A favorable loading affects the waveform of the produced tone and produces a richer spectral content.

It feels like my experiences with a more head-voice oriented phonation is consistent with this reasoning. And now some speculation: Perhaps this stronger interaction that happens when using a thinner type of phonation can explain the particular kind of sound that can be heard of some of the old-school tenors like Björling?

Toreadorssong's Vocal Technique Blog said...

Thank you CD. Glad you found us. Feel free to participate in the discussion.

Toreadorssong's Vocal Technique Blog said...


Thank for the summary of Titze's paper, and I believe you are interpreting the findings exactly right. The "leaner" phonation makes all the difference. What I believe is that this type of phonation requires a different balance between TA and the CT, which in turn affects the response of the other 3 intrinsic muscle sets (CA, both lateral and posterior and the IAs. When I first began my change, I could not produce this sound because my baritone voice was this kind of chest voice that Titze mentioned. The more I strengthen the CT, the more I am able to produce the leaner sound, which without a doubt gets a much stronger influence from the vocal tract adjustments. As a result, the energy I generate in the singer's formant region is exponentially superior as compared with the baritone production. I am working on two posts which might take a couple of days to complete. I am at a friend's wedding and will not have a lot of time to write in the next 4 days, but hopefully I will find a bit of time.

You are correct about the old school tenors and Piotr Beczala, the current Polish sensation is the only tenor at the top ranks I hear consistently with that kind of production. The important thing is that it can be taught, and with awareness of the issues at hand, change can happen relatively quickly.

We must better define the term chest voice. In terms of fold vibration, Titze's definition describes what i call "open chest" or "loose chest, a type of inefficient phonation that require greater breath pressure to sound stable and strong. I also refer to lean phonation with a first formant dominance as "chest voice". That is a resonance based definition which presupposes that phonation across the range should be that lean, like Bjoerling, Gedda, Gigli and more recently Pavarotti, early Corelli and the very excellent Goesta Winberg, who we lost too soon.

Thank you again Martin for a very, very important comment. More on this later.

Martin Berggren said...

[TS]: What I believe is that this type of phonation requires a different balance between TA and the CT, which in turn affects the response of the other 3 intrinsic muscle sets (CA, both lateral and posterior and the IAs.

I can feel the stretching (CT I guess) and (what I believe) the TA contraction of the folds. The rest of the muscles I have not learned to identify yet...

[TS]: ... I am able to produce the leaner sound, which without a doubt gets a much stronger influence from the vocal tract adjustments. As a result, the energy I generate in the singer's formant region is exponentially superior as compared with the baritone production.

This is interesting! Have you noticed this singer's formant difference in the spectral analysis of your voice using different types of phonation? I have not yet done that myself, but my intuitive guess would be that the squillo also is somewhat leaner (lower in frequency) when using a leaner production, even if the intensity can be high.

[TS]: I am working on two posts which might take a couple of days to complete. I am at a friend's wedding and will not have a lot of time to write in the next 4 days, but hopefully I will find a bit of time.

Excellent! Really looking forward to these!

[TS]: You are correct about the old school tenors and Piotr Beczala, the current Polish sensation is the only tenor at the top ranks I hear consistently with that kind of production.

Oh! I immediately checked out again some of Beczala's you-tubes, and I agree, he really seems to work on a nice lean head-type of voice! I wonder, however, a little bit about some of the contributions, for instance the Che Gelida Manina with piano. The production is lean and smooth,, but I do not really hear in this case a very strong resonance response, compared to some of the old-schoolers, or even somebody like Bruce Ford. Well, this is not meant as a criticism; he is a master of course.

[TS]: We must better define the term chest voice. In terms of fold vibration, Titze's definition describes what i call "open chest" or "loose chest, a type of inefficient phonation that require greater breath pressure to sound stable and strong. I also refer to lean phonation with a first formant dominance as "chest voice". That is a resonance based definition which presupposes that phonation across the range should be that lean, like Bjoerling, Gedda, Gigli and more recently Pavarotti, early Corelli and the very excellent Goesta Winberg, who we lost too soon.

Yes, I think this is very true! The Titze chest and head (he calls them modal and falsetto registers) I think should be thought of as being extremes in a continuum. It is likely that the modes of vibration that occur in well supported singing will be more complicated.

Masha said...

I was wondering -- is it possible for someone to teach themselves the flute voice? I'm a female soprano and I would really like to learn how to do it, but there isn't anyone nearby who may be able to teach me.

Toreadorssong's Vocal Technique Blog said...

G'day Masha,

Sorry I did not respond earlier. Yes it is possible to teach oneself the flute voice. If you are not able to produce a flute voice, it is possible that you have an overdeveloped lower voice (vocalis muscle). This means that even when you try to sing lightly, the vocalis remains active. Flute voice requires total passivity from the vocalis muscle. If you cannot do it, we could say that you are vocalis musclebound. With daily practice of trying to sing very lightly throughout your usable range, you will gradually develop the vocalis passivity that will eventually bring the light high voice. I once taught a bass who had no falsetto whatsoever. He was a pianist. We worked for two years. By the time we were done he had developed a very good high falsetto and his bass voice range was much expanded. It takes a lot of patience though. I wish you much success.