Monday, August 4, 2008

First formant high voice: The virtues and dangers in Alfredo Kraus' Legacy

Few opera singers in the last century have as much of a cult following as Alfredo Kraus, and with good reason. The term tenore di grazia was made manifest in the art of Alfredo Kraus. He was capable of great facility in the high tenor range, great range of dynamics throughout the voice and a sense of the musical line that made for very "graceful" singing as the moniker, tenore di grazia, suggests. Kraus's voice and exceptional musicality made him the ideal vehicle for the great works of the Italian bel canto repertoire as well as the higher French tenor roles.

The paradox in Kraus' legacy is the following: the singer who is the greatest example of longevity in our times has an inherent flaw that makes his example dangerous to tenors who do not understand how he managed his voice. This may seem controversial and disrespectful. But I assure you, I am a devoted fan of the art of the great Alfredo Kraus. This is not an attempt to blemish the reputation of the great tenor, but rather to bring to light a point of fact that accentuate both Kraus' extreme awareness of the possibilities and limitations of his instrument, and the dangers when those who follow his example do not understand the limitations of the approach.

The issue I will bring up in this article has great relevance not only to lighter voiced tenors, but to the way that those who hire in the business misunderstand vocal weight in general.

The controversial characteristic of Kraus' technique is his approach to the passaggio and above, and what the resultant vocal quality is. Kraus, unlike his contemporary, Luciano Pavarotti, and more like his predecessor, Giuseppe di Stefano, sings what we often all an "open" top range. The term "open", which has multiple definitions has been dealt with in a previous article. In this case, I am speaking of resonance strategy. Kraus sings a high voice that can be described as "uncovered" or "non-girato", and in more objective terms, first-formant dominant.

The term first formant dominance refers to the natural resonance of the tenor range below F#4.
In the case of such singers as Di Stefano, Kraus and currently Juan Diego Florez (another very successful tenore di grazia) this resonance strategy is taken higher than F#4 and in fact throughout the full range of those singers. The question therefore is: "Is there in inherent danger in first formant dominance in the high male range?" The answer is a qualified no!

Kraus' longevity is enough of a testimony that one can sing in first formant dominance throughout the range and excell. Diego Florez is another singer in the style of Kraus who examplifies grace and flexibility. The caveat is the following: first formant tuning above the passaggio requires raising the larynx above its natural at-rest level in order to be resonant with the rising pitch. The raising of the larynx reduces the lower pharyngeal area which is necessary for the strength of lower harmonics of the sung pitch. The natural lower laryngeal position (we are not speaking of depressed larynx here) is also absolutely necessary for the resonance of lower notes as well.

In other words, a high larynx high range, takes away the naturally darker colors of the voice in the subtle chiaroscuro (bright/dark) balance. This is crucially relevant to repertoire choice and this is where Kraus' awareness of his own vocal limits made him the icon of longevity but simultaneously restricted in his range of repertoire. The lack of strength in the lower harmonics give the voice a brighter more lyrical color, consequently inappropriate for roles that require richness of color. The closing off of the darker colors of the voice makes it virtually impossible for the first formant tenor to sing a fuller resonance, balanced by both sides of the chiaroscuro spectrum. In the first formant strategy, the only way to achieve darker colors is by singing a heavier sound, which over time would compromise the tenor's ability to sing high notes.

This is the crucial difference between Di Stefano and Kraus. Both Kraus and Di Stefano might have had enough vocal power to address the heavier side of the tenor repertoire if second formant tuning was a part of their resonance strategy. Kraus respected the limitations of first formant tuning in the high range and sang a lighter repertoire. Di Stefano sang heavier repertoire and paid the price. Indeed, over time he could not trust his upper range, although he had a impressive and consistent high C in his younger years. Kraus on the other hand, never violated the basic weight of the instrument and accepted the limits of the brighter first formant strategy. It is really interesting to compare Kraus' voice throughout the many years of his career. His voice had a lot more depth in the early years. Here I compare Kraus' 1977 and 1989 renditions of the aria Salût, demeure chaste et pure from Gounod's Faust.

Kraus 1989 Faust

Alfredo Kraus high C 1977 and 1989

This spectrograph shows the 1977 version on top and the 1989 version at the bottom of the high C5 with the green cursor going through the first formant and the white arrow pointing to the second formant. The vowel modification is [E] as in "met".

The crucial observation here is that Kraus' natural abilities point to a tendency toward second formant tuning in his early years. His technical philosophy, very much based on "forward placement" encouraged a high larynx production resulting in first formant dominance in the high range. We find confirmation for this in other comparative instances from the clips, for example the high G# at "enveloppant son ame":

Alfredo Kraus High G#4

This spectrograph also shows a clear difference between the spontaneous approach of Kraus' early years and the change to first formant strategy in the middle and later years.

This is one of the reasons why discussing technique is often a frustrating quagmire. It is difficult in this case to distinguish between Kraus' nature and his technical nurture. I believe that Kraus' nature reflected the traditions of the time. The quality of his voice was not unlike the young Pavarotti. In fact, upon hearing Kraus' early recording of the Duke in Rigoletto opposite Robert Merrill and Roberta Peters, I mistook him for a young Pavarotti. In later years, I could not possibly mistake one for the other. The change to first formant strategy happened over time as a consequence of his technical philosophy.

As previously stated, the first formant strategy gives a brighter more lyrical sound. Compared to Pavarotti who as similar attributes (e.g. full lyric tenor with easy top notes, efficiency of phonation, dynamic flexibility), Kraus sounded considerably lighter in weight. Did Pavarotti have a naturally heavier voice? No doubt. But I submit: not by much! The two tenors sang similar repertoire in their early career and given Kraus' tendency in the direction of second formant tuning, the voices had much in common. The divergence in repertoire, Kraus' continual leaning out and Pavarotti's deepening can be easily tracked through the difference in their resonance strategies. Did Kraus' repertoire force a first formant strategy or did his technique limit him to a repertoire that required lyricism and grace rather than power? This is not so easy to answer, but given the continuous gain in higher partial and loss in lower throughout his career, I submit that his technique dictated the nature of his repertoire.

Pavarotti by contrast follows what would be considered a logical progression. Second formant tuning in the high range conserves his chiaroscuro balance. His voice never loses depth. Even with his naturally brighter voice, richness in the middle grows with time. With time the vocal musculature gains in strength and is able to endure greater breath pressure. Therefore Pavarotti was able to assume more dramatic roles with negligible loss to his attributes. He was able to expand his repertoire, while Kraus was limited.

These two clips show Kraus and Pavarotti at 31 and 36 years of age respectively singing "La donna è mobile". They are similar in sound quality here. The next two clips show the difference in quality after 30 years of singing.

The later two clips show both tenors remarkably fresh at age 61. However, to my ears, Kraus' voice has thinned out resonance-wise. The top is secure but a touch strained. Pavarotti by contrast remains full-toned throughout and his voice gained in depth and substance with time. One could say that we cannot compare two such individual voices, and on the one hand, that may be valid. On the other hand, it could be a cop-out.

It is important to analyze the imperfections of our heroes in this art form. I find it totally consistent that I enjoy Kraus the artist and gifted vocalist and still be able to make critical observations about his technique. If the discussion was about musicianship, I might have a word or two against Maestro Pavarotti's contributions. In Richard Wagner's words, Opera is a gesamtkunstwerk, a total work of art. No opera singer has been perfectly gifted in all areas. We learn best when we can make sense of the strengths, weaknesses and especially the idiosyncrasies of our operatic heroes. The art of opera is traditional and cultist to a fault. I am preempting a possible backlash from Kraus devotees. Paradoxically I am one.

However, the final word I wish to leave in this long edition is that this first formant strategy in the high range has become so chronic among professional tenors that those who are hiring cannot distinguish clearly wether someone is "turning" the voice or not. Consequently, they hear dramatic tenors with a brighter sound than their predecessors, because of a crucial misunderstanding about what produces the ring in the voice, and they believe that this is a correct new trend. And when they hear a lyric tenor who achieves an excellent chiaroscuro balance (which necessitates second formant strategy in the high range), they mistakingly think that such a singer is a dramatic tenor because of the darker color he achieves. We will continue to have a crisis of tenor voices if a clear distinction cannot be made between a voice that is bright or shrill because the larynx is high and one that is brilliant because the voice is balanced in terms of resonance. The nature of vocal tissue has an effect on color. Lighter voiced tenors may have dark voices, and dramatic tenors may have very bright voices even when the voices are well-balanced. We have the tools to make these distinctions and we should begin now, and help this art form develop rather than stagnate.

I will never isolate the voices of working professionals for analysis on this blog except to point out a positive element. It is hard enough to manage a career without having side-liners making commentary. However, there is nothing that keeps me from attempting to make sense of the careers of prominent singers who are either retired or have passed to the choir of angels. We owe it to ourselves to understand the great stars fully, and not be only dazzled by their obvious achievements. The kind of criticism that we too often read about the vocal technique, indeed about all aspects of the performance of opera singers, is too often superficial, subjective, arbitrary and even mean-spirited. When critics have little or nothing to say, they give in to sensationalism which impresses the less knowledgeable reader with shock value. It is my hope that we can begin to have a discussion about great singers that goes beyond idol-worship by fans and singer-bashing by sour critics.

© 08/12/2008


Martin Berggren said...

I have a question! (Tenor voice assumed,)

When the first formant dominates below the passagio, the second harmonic is typically enforced. If a covering a la Pavarotti is applied above the passagio, the second formant enforces the third or fourth harmonic. I hear this very clearly as a sort-of whistling sound (quite exciting!) in a few tenors like Pavarotti and Corelli, to a lesser extend in old-schools like Björling, and not so much in somebody like del Monaco.

However, what I do not understand is how singers like Kraus and Florez, who does not apply cover in the above sense, approach their high range! The first formant is located (depending of wovel) roughly around 400-600 Hz, right? This corresponds roughly to the fundamental for the tenor high notes. Thus the "first formant singer" needs somehow to shift resonance from the second harmonic in the middle register to the fundamental at the high notes. This implies some kind of passagio even for the first-formant singer, or have I misunderstood something?

As "summer homework" from my teacher, I am trying to get through Una Furtima Lagrima. I have therefore listened to a few versions, and I have to admit (somewhat reluctantly) that a tenore di grazia voice like Florez has some advantages in this aria. Since the whole piece wanders through the passagio, there is a less noticeable sound-quality shift in such a voice than in for instance Pavarotti's version. Well...

Toreadorssong's Vocal Technique Blog said...

Dear Martin,

Thank you for another well thought out question. The first formant of the [a] vowel or better said what we are willing to accept as [a] in context goes up as far as 850 Hz. And although the bandwidth of the vowel formants are relatively narrow, at best 40 Hz, it is their proximity to the relevant harmonic that matters.

As for the first formant tenor's high range (400-520 Hz), it rides between the fundamental ( F0) and second harmonic (H2). This is the inherent difficulty in first formant tuning that high. Quite a bit of contortion is necessary to affect the formant resonance in the first formant strategy (e.g. excessive lip rounding, closing of the jaw, laryngeal migration, etc). That is why I think it is not the ideal strategy.

But the high tenors have an advantage, they have significant crico-thyroid dominance and have therefore very easy phonation in the high range. Even if the resonance is not adequate, it will be close enough to get the sound out. What we lack from such singers is the "scuro" part of the chiaroscuro balance.

As for Florez, his voice is unusually high by nature. He is a true tenorino. His extreme musicianship and musicality and the purity of his phonation add to his charm. Florez's high-lying voice makes all of his repertoire extremely easy for him to sing. Him singing 9 high Cs is the equivalent of Pavarotti singing 9 high Bbs if not As.

Toreadorssong's Vocal Technique Blog said...

A correction! I received an off-blog comment that made me aware that I had the soprano wrong on the Rigoletto recording that I referred relative to mistaking Kraus for Pavarotti. The recording I listened to was the 1963 RCA recording with Kraus, Merrill and Moffo (not Peters).

Hemichromis said...

Very interesting.
A quick question,
I have been singing along with David Phelps, a classically trained tenor who sings gospel music, He, oftens sings up to C# or even D in a 2nd harmonic dominance.
The interesting thing is that his larynx doesn't seem to rise (aural evidence only)

I have also found that when i (try) sing along with him it is also the 2nd harmonic that dominates.
My theory on this is that it is the 2nd formant on the 2nd harmonic on the very highest notes.

Could this be accurate?

If so, why shouldn't a tenor do this?

Thank you

Jean-Ronald LaFond said...

Very good question James and there is nothing wrong with this approach. F2 on H2 sounds normal to me on the [a]-Schwa vowel spectrum. Second harmonic would be a little over 1K Hertz which coincides with the 2nd formant of [a]. So there is no problem with this. The problem would be if F1 was raised to meed the 2nd harmonic. Both coordinations are possible. The later would require a high larynx for sure.

This question is particularly interesting in that it is not possible in your average spectrogram to distinguish whether H2 is boosted by F1 or F2. We could probably tell by the nature of the sound coming out. One would sound balanced while the other would sound very strident.

Hemichromis said...

Thank you for your quick reply

I suppose my question would then be:
Why don't Tenors do it?

Fore a High C with a written vowel in the A-U spectrum it would be the simplest most natural sounding vowel modification, would it not?

I can only think that maybe using an open vowel would perhaps loosen phonation up there?

Jean-Ronald LaFond said...

For the high C, the [o] vowel can work and would bring most of the energy of the spectrum on H2 but lip rounding effects the entire spectrum and has a dampening effect on the singers formant. That is why I think the majority of tenors open the vowel to [ae] as in cat tracking F2 on H3. It encourages higher partials whereas lip-rounding emphasizes the lower part of the spectrum. It is theoretically possible to do what you are suggesting and modify the pharyngeal part of the vocal tract to a more neutral vowel like a schwa. Meaning the lip rounding can be made to effect F2 while the singer seeks to sing a schwa in the lower part of the throat. It is possible and there are singers who can do it well. Carlo Bergonzi uses this approach and there are some Italian schools that prefer it. I find it works better for baritones. Bergonzi had problems with notes above Bb because of this (I believe) Like every thing it takes practice, but it is not the most obvious adjustment from a resonance tracking perspective.

Hemichromis said...

I can see that acutally i just looked at your post about format tracking in tenors where you show begonzi using an 'U' strengthening the 2nd harmonic surprisingly Corelli did the same thing sometimes. Bjorling started on 'U' then ended on 'eh'(tuning the third)
Pavarotti went straight to 'eh'

Interesting, It does seem that perhaps the 3rd harmonic strategy lends a certain intensity to the voice.

Martin Berggren said...

I have a question!

Lately, I'm becoming somewhat more consistent in the change to F2 dominance on F#4 and up, at least when vocalizing... I check this by letting the Mac Voice Analyzer run. On Aah, I can usually accomplish a dominating H3 now. On Eh, it seems to be a higher partial that dominates. But the freaking Oo (Swedish Å) is almost impossible. I seem to get F2 dominance if I don't completely change the vowel. Any thoughts? To weak source tone?

Hemichromis said...

Mr Berggren,
I know you're not asking me but thought i would answer anyway!

I find that the 'eh' vowel tends to gravitate toward the 4th harmonic between F and A, infact it was the first vowel i could turn!

I had problems with overly thin phonation before but it's getting better now.

I Used to get very annoyed when Mr Lafond would say: when phonation is efficient the resonance change is automatic. He has now been proven right in my experience!

(I am assuming that the second time you mention 'F2 dominance' is a mistake)

As discussed above and in the post which featured Bergonzi. F2 dominance for the oo vowel may well show as a 2nd harmonic dominance.

Look on youtube for 'O paradiso' and you will see how different tenors handle the oo in the first Bb.
some for for the oo and enhance the 2nd harmonic
others go for eh enhancing the 3rd
still others start oo then move to eh(bjorling)

I suspect I am only telling you things you already know!

Hemichromis said...

on a closer look at the spectrum of myself singing scales i have noticed that if i do not 'cover' the 1st formant will stay where it is, it will go no higher than it's postion at F# and yet the width of it's effect allows it to continue enhancing the 2nd harmonic up to G or even Ab. the tension builds and then it has to turn or else become spread.

This to me confirms what i am feeling; that my larynx is not going up as i sing higher.

I would expect it means that my larynx's defualt position is a little higher than ideal?