Registers—The Snake Pit of Voice Pedagogy

PART 1: PROPRIOCEPTION, PERCEPTION, AND LARYNGEAL MECHANISMS

Christian T. Herbst

[This article is a slightly revised English version of the manuscript series,

“Register—die Schlangengrube der Gesangspädagogik,” published 2019–2020
in Vox Humana.]

A
TERMINOLOGICAL CONFUSION
ccording to Thurman et al., the first attempts to define
vocal registers date back to the thirteenth century.1 Since then, a
large variety of register definitions and categorizations have been
Christian T. Herbst proposed (see, e.g., excellent historical overviews by Henrich,
Stark). A limited selection of noteworthy register classifications is provided
2

in Table 1.3
A review by Mörner et al. found more than 100 different terms for registers
used in scholarly writing.4 Some thirty years ago, Sundberg commented on
this situation, stating that, “Unfortunately, there is no generally accepted clear
definition of the term register.”5 While recent decades saw some excellent
scientific contributions, some of which are mentioned in this article, there

TABLE 1. Exemplary overview of proposed register classifications.

Author Year Count Terms

Zacconi 1596 2 voce di petto, voce di testa
Garcia 1847 3 voix pointrine, registre de fausset, registre
de téte
Behnke 1886 5 lower thick, upper thick, upper thin, upper
thin, small
Van den Berg 1963 3 (+2) chest, head or mid, falsetto (main registers)
Strohbass, whistle (further registers)
Vennard 1967 3 chest, middle, head (females)
Chest, head, falsetto (males)
Hollien 1974 3 pulse, modal, loft
Miller 2000 4 chest, middle, upper, flageolet (females)
Chest, full head, falsetto (males)
Roubeau 2009 4 M0 (fry), M1 (chest), M2 (falsetto), M3
Journal of Singing, November/December 2020
Volume 77, No. 2, pp. 175–190
(whistle)*
Copyright © 2020 * The terms in parentheses were added by the author in order to increase readability; see Table 5
National Association of Teachers of Singing in Roubeau for a complete list of synonyms.

November/December 2020 175

Christian T. Herbst

remains some controversy concerning both (the lack Rather, it should become clear from this article that
of) unanimously accepted definitions and pedagogic the terms themselves are not important, but rather the
application. proper descriptions and definitions of underlying phe-
The variety of seemingly incongruent register defini- nomena. Once the latter becomes clear, the terminology
tions suggests that the subject matter might be more may even become irrelevant, as long as everyone who
complex than sometimes assumed. In particular, some partakes in a discussion about registers has the same
confusion may stem from the fact that various register basic understanding.
definitions target different aspects of reality. Janwillem
van den Berg, for instance, critically observed that the 1. PROPRIOCEPTION
different register terms
Probably the first explicit mention of vocal registers was
are a hodge-podge arising from such divergent sources as: made by Zacconi at the end of the sixteenth century.
secondary clues (resonances in the chest with the chest In his Prattica di musica, he distinguished voce di petto
voice); misconceptions (non-existing resonances in the (literally translated as “voice of the chest”) from voce di
head with the head voice); acoustical illusions (with the testa (“voice of the head”).8 Since these register terms
falsetto); acoustical resemblance (the rattling sound of clearly target distinct regions of the body, they most
trodden straw with the Strohbass voice); and similarity of
likely have been established in close relation to the con-
origin (eddies generated in a narrow opening and subse-
cepts of appoggiarsi in petto und appoggiarsi in testa. For
quent cavity resonance with the flute or whistle register).6
instance, the Enciclopedia Garzanti della musica explic-
The notion that vocal registers might be assessed itly states that appoggio refers to both the torso and “the
on several levels provides the conceptual framework part of the facial cavity where the cervical resonances of
for this study, which will discuss registers against the the sound are perceived.”9 Surprisingly, this description
following backgrounds: 1) proprioception; 2) psycho- very closely matches a current definition of resonant
acoustic perception; 3) laryngeal mechanisms; 4) vocal voice; “a voicing pattern involving oral vibratory sen-
tract effects; and 5) individual didactic systems. The sations, particularly on the alveolar ridge and adjacent
aim is to show that some apparent discrepancies between facial plates, in the context of what subjects perceive as
register definitions/classifications can be resolved by ‘easy’ phonation.”10 This impressively demonstrates that
considering the different backgrounds against which proprioception is an important common denominator
they have been established. Such an acknowledgement of various notions of voice pedagogy. However, it also
of the increased complexity of the topic of registers (or, hints at a certain degree of convolution of the involved
rather, the removal of unjustified oversimplification) concepts, that is, proprioceptively defined registers,
may eventually lead to increased clarity and better appoggio, and resonant voice. This may potentially
understanding in both science and teaching. exacerbate respective discussions, particularly when
To conclude these introductory remarks, a crucial (currently partially lacking) objective empirical evidence
disclaimer must be made. Unfortunately, all available is substituted by subjective impressions of propriocep-
vocal register terms come with certain connotations tive experience.
and interpretations; this is true even for the apparently The central question is whether the historical proprio-
neutral terminology of M0 to M3.7 However, discussing ceptive register definition of “head” (relatively higher in
registers and their classification unfortunately requires pitch) vs. “chest” register (lower in pitch) is supported by
an a priori auxiliary terminology; thus, an arbitrary empirical evidence provided by modern voice science,
choice for preliminary register terms had to be made. objectively documenting vibrations of the head or chest.
During parts of this text, the terms (vocal) fry, chest, In a highly significant review, Sundberg summarized the
falsetto, and whistle register are used as proxies, because results of nine different studies, coming to the following
they seem—at least in the opinion of this author—to be conclusions.11
the most widely utilized terms, without making a conces- • Vibrations in the chest and head regions do indeed
sion that these terms are appropriate or even “correct.” typically occur during singing. However, they do not

176 Journal of Singing

 Registers—The Snake Pit of Voice Pedagogy, Part 1: Proprioception, Perception, and Laryngeal Mechanisms

significantly contribute to the radiated sound but may being notably relevant for the quality of vibrations
(only) serve as a feedback for the respective singers. in the head. While particularly the pitch dependency
• The strength of the vibrations occurring in the supra- of vibrations and their perception may explain why
glottal regions is dependent on the sung vowel, notably Zacconi introduced the kinesthetically oriented con-
influenced by jaw opening which has a causal influ- cept of “head” vs. “chest” voice, this can not be taken
ence on the center frequency of the lowest vocal tract as evidence or even proof that laryngeal registers (see
resonance, and thus the first formant. In contrast, the Laryngeal Mechanisms below) are the cause for these
strength of the vibrations of the subglottal region is vibrations; rather, these phenomena might simply be
independent of the sung vowel. linked to fundamental frequency/pitch. Along those
• The strength of the vibrations is greatly dependent on lines, a consensus report produced by the Collegium
the sung pitch or, in physical terms, the fundamen- Medicorum Theatri maintains that “while the sensations
tal frequency of vocal fold oscillation. In particular, felt by singers, of course, are valid sensations (indeed,
only fundamental frequencies up to 300 Hz (D4) can even the non-singer can experience them), they have
be perceived in the chest.12 Surprisingly, that upper nothing to do with vocal registers!”15 It would appear
frequency limit of ca. 300 Hz coincides more or less that a series of rigorous empirical experiments is needed
precisely with the zona di passaggio, constituting the to shed further light on the matter.
main register transition in classical singing. Along From a pedagogic point of view, it is remarkable
those lines Sundberg suggests that that vibrations of the chest and the head only margin-
ally contribute to the radiated sound.16 It is thus safe to
as the sensitivity to vibrations decreases above about assume that they are actually more or less inaudible.
300 Hz and as phonatory vibrations decrease with Rather, these vibrations may serve as a proprioceptive
rising frequency, phonatory vibrations cannot be per- feedback for singers and, indirectly, for their teachers.
ceived at high pitches. A bass singer should be able to In this context it is important to realize that vibrations
sense vibrations of his sternum throughout his range.
in the supraglottal region are greatly dependent on the
A tenor should be able to sense them for the lower and
sung vowel. This would suggest that proprioceptive
middle part of his range, the notes below the pitch G4.
An alto is not likely to sense vibration of the sternum feedback is relatively stable during vocalises with con-
except for her lowest notes. This may be the reason stant vowels, at least over a limited pitch range, but may
why the register used in the female voice for the lowest greatly vary when singing lyrics made up of multiple,
pitches is called the “chest” register.13 varying vowels. Furthermore, assuming that cervical
vibrations are created by acoustic standing waves in the
Whether facial/cranial vibrations are actually linked supraglottal vocal tract,17 the respective proprioceptive
to different voice registers remains unclear, according to sensations may not be comparable across singers, due to
Sundberg’s review. A recently published pilot study by intra-individual anatomic differences. For these reasons,
Kitamura and Othani, investigating three female singers concepts of “voice placement,” like the one proposed by
with laser Doppler vibrometer scans, suggests that the Lilli Lehmann,18 should, in the opinion of this author,
amplitude of facial surface vibrations is dependent on not be treated as absolute truth, because there might be
fundamental frequency of the emitted voice. However, limits to generalizing them. Rather, such concepts are
in that study’s methodology vocal registers have not useful in illustrating proprioceptive principles. They
been explicitly controlled. Therefore, it remains unclear may be utilized individually. There potentially exists a
whether the effect is caused by fundamental frequency “body map of voice placement” for each singer, which
or by the chosen vocal register.14 is dependent on personal anatomy and sung vowels and
In summary, the available empirical evidence base pitches. Such maps need to be developed carefully and
seems to suggest that there is some systematic correla- individually in voice training. Once reliably established,
tion (if not causal relation) between the fundamental they are likely most useful in situations where the singer
frequency of voice production and vibrations in the does not have good acoustic feedback, as, for example,
head, neck, and chest, with the choice of sung vowel when singing in the midst of a loud orchestra.

November/December 2020 177

Christian T. Herbst

2. PSYCHOACOUSTIC PERCEPTION by a more recent study in the context of quantitative

voice analysis.25 Both the spectral composition of the
One way to describe vocal registers is to perceptually
voice source (i.e., the amplitudes of the individual har-
assess the sound of the voice, typically without consider-
monics created by the laryngeal sound source) and the
ing the underlying physiological and physical produc-
resonance structure may play a role in this respect.26
tion mechanisms. For instance, Titze reported that “The
Upon closer inspection however, the situation is not
term register has been used to describe perceptually
trivial.
distinct regions of vocal quality that can be maintained
• It can be assumed that the ability for perceptual register
over some ranges of pitch and loudness.”19 This defi-
identification is influenced to a certain degree by a con-
nition exclusively targets the psychoacoustic domain ceptual and contextual bias of the listener.27 Therefore,
(with pitch and loudness being the respective percep- care should be taken when formulating the listener’s
tual counterparts of the physical qualities fundamental instructions, because the use of explicit register ter-
frequency and sound level). In light of this definition it minology may have an effect on the perceptual rating.
needs to be established to what extent listeners can actu- • Voice registers theoretically can be distinguished via
ally distinguish between different registers. It appears several acoustic (perceptual) features: fundamental
that two register transitions have been investigated in frequency (pitch); spectral composition (timbre); and
more detail: between vocal fry (sometimes termed pulse sound level (loudness).28 Therefore, the perception of
register) and chest register; and between chest register voice registers is a multidimensional phenomenon,
and falsetto (please recall the initial disclaimer concern- relying on several sound attributes of the voice. This
ing vocal register terms). may explain why the perceptual discrimination of
registers in the zona di passaggio (i.e., a region where
Vocal Fry vs. Chest Register
the fundamental frequency/pitch of registers may
Vocal fry was defined by Hollien as a physiologically overlap) is more challenging, due to ambivalent and
normal kind of voice production having pitches—or, thus potentially unusable fo information.
rather, fundamental frequencies—below those of the • The perceptual discrimination of distinct register
chest register.20 Vocal fry is characterized by distinct glot- changes or even breaks—consider yodeling as an
tal pulses, where each pulse is more or less completely extreme case—is typically straightforward. In con-
attenuated before the next pulse commences. Owing trast, the task at hand is obviously more difficult when
to the relation fo = 1 / T, the long oscillatory period T trained singers skillfully mask a laryngeal or resona-
found in vocal fry is responsible for the low fundamen- tory register transition. That latter situation leads to
tal frequency fo. The perception of vocal fry is closely the question of whether there are clearly defined and
linked to fo,21 with the upper threshold being at fo ≈ 70 measurable minimal changes of (psycho-) acoustic
Hz22 or about C#2. In other words, vocal sounds with a parameters that reliably allow one to identify register
fundamental frequency of 70 Hertz or below are likely transitions. In other words, is there a “just noticeable
perceived as vocal fry. difference” (JND) that is required for two vocal sounds
to be perceived as different registers?29 If so, how would
Chest vs. Falsetto Register it be defined?
When considering the perceptual differences between As indicated above, voice registers may be percep-
what is provisionally called chest and falsetto register in tually differentiated on three levels: pitch, loudness,
this text, one can—at least as regards older literature— and timbre. The JNDs for pitch and loudness have
again refer to an excellent review paper by Sundberg.23 been rigorously researched,30 and there are serious
According to this review, several studies suggest that attempts to investigate and describe the JND of
especially trained listeners can discriminate between the timbre.31 However, establishing JNDs for the differ-
chest and falsetto registers relatively well,24 particularly entiation of vocal registers may be more challenging,
when these two voice qualities are presented as distinct due to potential perceptual interactions between the
antagonists. This has been at least partially corroborated aforementioned parameters pitch, timbre, and loud-

178 Journal of Singing

 Registers—The Snake Pit of Voice Pedagogy, Part 1: Proprioception, Perception, and Laryngeal Mechanisms

ness, and possibly also additional influences from the put and disregarding its inner workings. Such a modus
temporal dimension. (When does a gradual transition operandi reduces the pedagogue’s available didactic
turn into a “register break”?) To the best knowledge strategies to only imitation learning through trial and
of this author, this very complex topic is still under- error, likely introducing a certain degree of inefficiency
researched, requiring further scientific investigation. when student and teacher are not of the same voice type
or Fach, or in the presence of a fundamental functional
(Perceptually) Unifying the Registers
voice production issue on the part of the student. In
Particularly within the aesthetic context of classical sing- particular, if the physiological reasons and principles
ing, it is required to unifiy the registers in a way that no of physics (i.e., the inner workings of the “black box”)
transition between them is audible. While this makes for any (un)wanted vocal phenomenon are unknown,
sense on a perceptual level, some authors go one step it may be difficult to find the most appropriate teaching
further and altogether deny the existence of registers. For strategies.
instance, Lilli Lehmann stated in an essay, which still
This may be especially true for the issue of vocal
seems to have some influence in the German-speaking
registers. As will be shown in the following sections of
community of singing teachers, “When teaching a voice,
this article, register transitions can be predominantly
registers should not exist nor should they be created”;
of laryngeal nature, or they can be caused by vocal
“Do registers naturally exist? No.”; and “As long as the
tract influences. Naturally, since both these potential
term ‘register’ is retained, registers will not vanish”
causes are attributed to different subsystems of the voice
[translations by CTH].32
In the opinion of this author, even these relatively apparatus, they would require fundamentally different
extreme postulates have some merit, but only when pedagogic interventions, because different motor control
being assessed and interpreted exclusively on a psycho- aspects of voice production are targeted.
acoustic level. They might then serve as an aesthetically The perceptual ambivalence linked to register transi-
motivated definition of the perceptual end result of voice tions is best illustrated with the example from Donald
production. However, they hardly bear any noteworthy Gray Miller’s book Resonance in Singing, shown in
relevance to the physiological and physical reality of Figure 1.33 The G major scale sung by a mezzo soprano
singing voice production. on vowel [ɑ] has two audible transitions: one at t≈3 s
(from B3 to C4) and the other at t≈5.5 s (from E4 to
The Purely Perceptual Approach F#4). (It will be shown in the following sections that the
in Voice Pedagogy transition at t≈3 s is laryngeal and the other at t≈5.5 s
It should be evident that a voice pedagogue always is resonatory in nature.) Repeated ad hoc listening tests
should consider the final “product” of teaching, that is, performed by this author at various symposia with a
the sound of the singer’s voice, and how it is perceived larger audience revealed that, singing teachers and voice
within the chosen aesthetic context. In classical singing, therapists about equally chose either transition when
for instance, abrupt timbral and pitch changes expose asked to “identify one register transition in the scale,”
inexpertly executed register transitions (see remarks with very few declared abstentions. The ambivalence of
about blending the registers in Part 2 of this article), opinion found even among experts demonstrates the
while such phenomena might actually be crucial features limitations of the purely perceptual approach.
in other singing styles, such as CCM (contemporary
commercial music) or some forms of world music. For 3. LARYNGEAL MECHANISMS
these reasons, a teacher’s assessment of the singing voice
should always have a perceptual component. In the first two sections of this article, vocal registers
However, a purely perceptual approach in voice peda- were discussed from the propriceptive and perceputal
gogy—neglecting the physiology and physics of voice point of view. A fundamentally different approach was
production—does have a clear limitation: It treats the proposed in the mid nineteenth century by Manuel
voice as a “black box,” targeting only the system’s out- Garcia. He suggested that a vocal register is

November/December 2020 179

Christian T. Herbst

Figure 1. Example for a perceptually ambivalent register transition. (A) Fundamental frequency; (B) spectrogram of the
acoustic signal; (C) relative amplitudes of harmonics 1 through 5; (D) dEGG wavegram; (E) Electroglottographic (EGG)
waveforms. Left panel: shortly before (t = 2.5 s, solid line) and shortly after (t = 3.5 s, dashed line) the laryngeal register
transition. Right panel: shortly before (t = 5 s, solid line) and shortly after (t = 6 s, dashed line) the resonatory transition. The
dashed vertical lines in (B), (C), and (D) indicate the time offsets at which the EGG waveforms shown in (E) have been extracted.
(Listen to this recording at nats.org/JOSmedia.) (Example 5.7 from Miller, 2008,33 with permission from the author.)

180 Journal of Singing

 Registers—The Snake Pit of Voice Pedagogy, Part 1: Proprioception, Perception, and Laryngeal Mechanisms

a series of consecutive and homogeneous tones going vocal effect. Perceptual aspects of vocal fry have been
from low to high, produced by the same mechanical discussed above. While some of the physiological and
principle, and whose nature differs essentially from physical aspects of vocal fry as a distinct register have
another series of tones equally consecutive and homo-
been addressed previously in research, further comple-
geneous produced by another mechanical principle. All
mentary investigation may be necessary.
the tones belonging to the same register are consequently
Past research centered on speech rather than singing
of the same nature, whatever may be the modifications
of timbre or of the force to which one subjects them.34 has suggested that vocal fry might be a distinct vocal
register, differing from the pitch-wise superjacent chest
Garcia’s description of laryngeal registers, then a radi- (modal) register in several ways. In contrast to chest
cally novel concept, is the centrally relevant concept even (modal) register, no correlation between the funda-
today. A comparable, but slightly different definition was mental frequency (fo) and vocal fold length was found
proposed by Hollien: “A vocal register is a series or range in vocal fry (overall, vocal fold length was lower in vocal
fry), suggesting a different mechanism for fo control as
of consecutive voice frequencies which can be produced
compared to chest/modal register.36 Furthermore, no
with nearly identical voice quality [emphasis by CTH].”35
correlation between fo and the thickness of the vocal
While Garcia’s definition considers only laryngeal phe-
folds and/or ventricular folds could be documented,
nomena of vocal fold vibration and sound production,
and the ventricular space was smaller than in chest
Hollien’s conceptualization (at least implicitly) also
(modal) register, indicating ventricular fold impinge-
encompasses resonatory phenomena introduced by the
ment. 37 Vocal fry was shown to have lower airflow,
vocal tract. Hollien further suggests that registers should
lower cricothyroid and interarytenoid muscle activity,
be operationally defined on perceptual, acoustic, physi-
and increased thyroarytenoid activity in comparison to
ologic, and aerodynamic levels.
modal register.38 Comparisons of subglottal pressure in
As far as the laryngeal voice production mechanism is
vocal fry vs. chest (modal) voice resulted in ambivalent
concerned, often four “main” registers, or, rather, laryn-
data.39 One type of vocal fry was found to be constituted
geal mechanisms, are considered. A decision was made by subharmonic vocal fold oscillation,40 where each
here provisionally to adhere to this classification, even complete vibratory period consisted of two or three
if there is some disagreement among scholarly sources. amplitude-modulated vibrations of the vocal folds.41
(In this context please recall the initial disclaimer: The
purpose of this text is not to propose a “correct” clas- Whistle Register (M3)
sification and terminology for voice registers, but rather A certain degree of disagreement about the physical
to review the different ways of defining and discuss- nature of the whistle register exists, particularly in older
ing registers, in order to provide deeper information literature. Some authors suggested an aerodynamic whis-
about the underlying principles.) These four laryngeal tle mechanism where the sound is created by vibrations
mechanisms are typically termed as: vocal fry (M0, pulse of air in the absence of vocal fold vibration, or where
register); chest voice (M1, modal register); falsetto (M2, vocal fold vibration at least would not causally contrib-
head voice?); and whistle register (M3). ute to sound generation.42 An alternative hypothesis
In the following sections, these mechanisms are dis- suggests that the whistle register is produced in analogy
cussed, beginning with the less common vocal fry and to the “default” mode of voice production via passive
whistle registers. Then, a distinction is made between self-sustaining vocal fold vibration,43 as described by the
the two central singing voice registers, chest and falsetto. myoelastic-aerodynamic theory of voice production.44
Possibilities of blending these latter two registers are In that case, laryngeal dynamics are hypothesized to be
considered in part two of this study. characterized either by a shortened glottis (“damping”)
to produce a “flageolet” tone,45 or by vibration along the
Vocal Fry (M0) full length of the vocal folds.46
While vocal fry is basically unused in classical singing, That latter hypothesis is corroborated by empirical
it is relevant in some substyles of CCM, at least as a data from laryngeal high speed videoendoscopy. In a

November/December 2020 181

Christian T. Herbst

single-subject pilot study, Echternach et al. documented This most fundamental vibratory component allows for
full glottal closure along the entire antero-posterior glot- movement of the vocal folds along the medio-lateral
tal width in a professional classical singer phonating C6 dimension, thus opening and closing the glottis and
to G6 (1047 Hz to 1568 Hz). This suggests that—at least gating the airflow. Interestingly, if the system would be
in classical singing—there is little reason to believe that modeled with only this one degree of freedom, the vocal
the whistle register is fundamentally different from the folds would likely just be blown apart by the tracheal air
falsetto register as far as fundamental laryngeal mechan- stream and then stay medialized/separated. Therefore,
ics are concerned. Rather, the main difference probably a second degree of freedom is required, which can be
may be found in resonatory adjustments.47 constituted in either of two ways (note that in reality, the
A different concept termed “glottal whistle” or M4 following two phenomena are likely both contributing
was proposed by Edgerton in the context of the “extra- to vocal fold vibration at various degrees, depending on
normal voice.”48 This glottal whistle, produced with both laryngeal and vocal tract configuration, subglottal pres-
ingressive and egressive phonation, was speculated to sure, and individual vocal fold morphology).
be “the result of a vortex produced at the upper edges of 1. In the presence of a coupled supraglottal vocal tract,
the vocal folds.”49 Fundamental frequencies of the glottal the delayed response of the supraglottal air column,
whistle were found in the range of 1000 Hz to as much caused by the mass (inertia) of the air molecules,
as 6503 Hz, or pitch G#8.50 sets up an asymmetric aerodynamic driving force
that allows the intraglottal air pressure to become
Chest (M1) and Falsetto Registers (M2) negative at the end of the open phase, thus facilitat-
The distinction between chest and falsetto registers ing vocal fold closure.53 This may result in nonlinear
at the laryngeal level is not as trivial as it may seem at interactions between vocal tract and voice source.54
first glance. Here, an attempt is made to describe the Major influence factors to this phenomenon are the
relevant physical and physiological underpinnings in coupling strength between source and vocal tract and
more depth than what is typically found in texts aimed the resonance structure of the vocal tract (see section
at an audience of singing teachers. For that reason, a few “Nonlinear source-filter interaction effects” in Part 2
concepts and insights from voice research conducted in of this article for details).
the past five decades are reviewed first. Please note that 2. The second degree of freedom can also be provided
in order to keep the focus on providing intuition—rather by a rotational vibratory component, constituted by
than detail—about the underlying principles, this is in out of phase movement of the inferior and superior
some instants done in a somewhat simplified/idealized vocal fold margins.55 This creates a time varying glot-
way that may not withstand the most rigorous scientific tal profile (convergent in the closing phase, divergent
scrutiny. Nevertheless, despite the author’s best attempt in the opening phase), also facilitating an asymmetric
to provide a clear and accessible account of the relevant driving force in analogy to what has been indicated
material without sacrificing too much fundamental infor- above. In the larynx, this phenomenon typically
mation, the following sections may be a challenging read. manifests itself through a so called mucosal wave at
For a summary of the following sections, please refer to various levels of magnitude. In the presence of such
Figure 2 and Table 1. a mucosal wave, vocal fold vibration is characterized
by a vertical phase delay:56 the inferior vocal fold edge
VIBRATORY MODES AND leads the vibration, with the superior edge trailing
MUCOSAL WAVES behind. The magnitude of the phase delay is depen-
dent on the vertical speed of the mucosal wave and
Early computational modeling 51 has shown that the
typically ranges from 60 to 90 degrees, i.e., 1/6 to 1/4
laryngeal sound generator needs two oscillatory degrees
of the vibratory cycle.57
of freedom in order to lapse into and remain in self-sus-
taining oscillation.52 One degree of freedom is constituted As a very simplified first approximation, the vocal
by a translational vibratory component of the vocal folds. folds can be modeled as two parallel vibrating masses or

182 Journal of Singing

 Registers—The Snake Pit of Voice Pedagogy, Part 1: Proprioception, Perception, and Laryngeal Mechanisms

Figure 2. Simplified schematic illustration of vocal fold vibratory modes, with vocal folds modeled as thin plates capable of
translational (i.e., away from the glottal mid-line and toward the glottal mid-line) and rotational (i.e., turning) motion. (A)
Symbolic illustration of vibratory components in the right vocal fold: translational (signifying medio-lateral vocal fold motion,
enabling glottal opening and closure) and rotational (representing vibratory phase delay along the inferior-superior dimension,
i.e., the vertical aspect of mucosal waves).* (B) Simplified schematic illustration of two characteristic eigenmodes of vocal fold
vibration: x-10 (a, b—representing the translational vibratory component) and x-11 (c, d—representing the rotational vibratory
component). The vocal folds are shown at two opposite phases of the vibratory cycle.**
* Taken from B. Story, “An overview of the physiology, physics and modeling of the sound source for vowels,” Acoustical Science
& Technology 23, no. 4 (July 2002); used with permission.
** Taken from J. G. Svec, “On vibration properties of human vocal folds: Voice registers, bifurcations, resonance characteristics,
development and application of videokymography” (Doctoral dissertation, University of Groningen, the Netherlands, 2000);
used with permission.

two vibrating plates (Figure 2). When considering the the so-called x-10 (or x-n0) mode, and the rotational
possibilities of motion of these idealized masses or plates, vibratory component is likened to the x-11 (or x-n1)
the two mechanical vibratory components described mode. More such modes exist, and vocal fold vibration
earlier come into play: the translational vibratory com- can be explained as a superposition of a certain number
ponent—fundamental to vocal fold vibration—is always of eigenmodes at various amplitudes. In periodic (i.e.,
present as soon as the vocal folds oscillate, facilitating regular) vocal fold vibration, the observed vibratory pat-
glottal opening and closure. Disregarding the aforemen- terns of the vocal folds are made up almost exclusively of
tioned option of a coupled supraglottal vocal tract for low-order modes such as x-n0 and x-n1,59 while higher
now (this concept is discussed in more detail in Part 2), order modes have a stronger influence in irregular vocal
the second feature is the rotational vibratory component, fold vibration, as found in screams in some singing
introduced by the vertical phase delay of vocal fold styles, or in pathologic voice production.
motion in the presence of a mucosal wave.
These two vibratory components (translational and ANATOMY AND PHYSIOLOGIC CONTROL
rotational) are schematically illustrated in Figure 2A. The anatomic boundary conditions for the distinction
When considering the vocal folds as a mechanical between chest and falsetto registers are constituted by
dynamic system consisting of two simple strings, the layered structure of the vocal folds, described by
these vibratory components would constitute so called the body-cover theory.60 While the cover consists of
vibratory modes or eigenmodes.58 In particular, the epithelium and the superficial (outer) and intermediate
translational vibratory component is the equivalent of (middle) layer of the lamina propria, the body is made

November/December 2020 183

Christian T. Herbst

up of the deep layer of the lamina propria and the thy- of the rotational vibratory component. The singer’s
roarytenoid muscle (TA; note that this muscle is often huge potential to diverge from the aforementioned
described as having two portions, i.e., the more lateral m. stereotypical scenarios gives rise to two possible—and
thyrovocalis and the more medial m. thyromuscularis).61 not mutually exclusive—interpretations: There may be
The choice of chest vs. falsetto register is mainly con- a great potential for producing what might be consid-
trolled by contraction/relaxation of the TA. In chest, the ered “mixed voice” at the laryngeal level; contrasting
TA is typically more active/contracted than in falsetto, modeling and textbook examples, identification of vocal
leading to a thickening, shortening, and medial bulging— registers at the laryngeal level is in reality much less
and thus at least partial adduction—of the membraneous trivial than hoped for. Finally, vocal tract interactions
portion of the vocal folds. For this reason, this maneuver (discussed in Part 2) may have an influence that is not
can also be called membranous medialization.62 to be neglected.
The contraction of the TA typically leads to an A highly simplified summary of the relation between
increased tension of the vocal fold body, which, all other vibratory components, eigenmodes, and the result-
things being equal, facilitates a relaxation of the vocal ing registers (most certainly not covering the entire
fold cover. Particularly at lower degrees of activity in the complexity of phenomena seen in reality) is provided
cricothyroid muscle (CT), considered to be an antagonist in Table 2.
to the TA,63 this results in different degrees of stiffness
in the vocal fold body and the cover, therefore allowing INFLUENCE OF VOCAL FOLD ADDUCTION
a certain degree of vibratory independence of these two
A somewhat surprising alternative physiological explana-
portions of the vocal folds.64 In such a case, the vocal fold
tion for the chest (modal) vs. falsetto dichotomy is given
cover typically assumes an x-11 (or x-n1) vibratory mode,
by Titze, suggesting a “spectral slope transition” (see
introducing a phase delay between the inferior and the
below for acoustic effects) that “can occur as a result of
superior vocal fold edge during vocal fold vibration, thus
either an adductory change or a loudness change” (text
exhibiting a mucosal wave. The resulting vibratory char-
emphasis by CTH). 66 Apart from the unmentioned
acteristics are typically an identifying hallmark of chest
implicit contribution of the adductors (i.e., the lateral
(or modal) register, sometimes also termed mechanism
cricoarytenoid [LCA] and the interarytenoid [IA]
M1. (In this context it is worth noticing that individual
muscles), Titze suggests that the bottom of the vocal fold
vocal fold morphology can play a huge role. Some per-
may be more adducted in modal (chest) register than in
sons may as a baseline have a more “chest-like” voice
falsetto via contraction of the TA. This conceptualization
than others, simply because their vocal fold mucosa is
is in line with later empirical findings in both humans
thicker, more pliable, and thus more prone to exhibiting
and dogs, showing that cartilaginous adduction (through
tranlational or x-11 vibratory modes.)
the LCA and IA muscles) and membranous medializa-
In contrast, relaxation of the TA and contraction of
tion (via the TA) can be controlled independently.67 That
the CT typically are associated with a greatly reduced or
independence of motor control facilitates the produc-
even absent mucosal wave, thus considerably decreasing
tion of a wide variety of different sound qualities at the
the rotational vibratory component (and thus more or
laryngeal level.
less eliminating the x-11 or x-n1 mode, or at least limit-
ing it to the epithelium instead of the entire vocal fold
ACOUSTIC MANIFESTATION OF THE
cover).65 Such a vibratory pattern would then quintes-
CHEST AND FALSETTO REGISTERS
sentially constitute the falsetto register, sometimes also
termed mechanism M2. In chest register, both the membranous adduction (via
Both these scenarios (TA and CT at absolute antago- the vertical bulging of the vocal fold) and the vertical
nistic levels of engagement) describe stereotypical cases. vibratory phase delay between the inferior and supe-
In reality, the activity of the TA vs. CT, particularly in rior vocal fold margins prolong the closed phase, or the
trained singers, can also be varied gradually, likely lead- time interval of a glottal cycle during which the glottis
ing to intermediate scenarios with various contribution is closed (either fully, or—in the case of breathy voice

184 Journal of Singing

 Registers—The Snake Pit of Voice Pedagogy, Part 1: Proprioception, Perception, and Laryngeal Mechanisms

TABLE 2. Simplified summary of the relation between vocal fold vibratory components, vibratory eigenmodes, observable
oscillatory phenomena, and resulting vocal registers.
vibratory component: translational rotational
eigenmode: x-10 x-11
vibration mechanics: medio-lateral vocal fold vibration (glottal phase-inverted vibration of inferior and
opening and closing) superior vocal fold margins; constitutes inferior-
superior aspect of mucosal wave
stereotypical cause: always present during vocal fold vibration TA contraction, CT relaxation; anatomical pre-
disposition (thick mucosa)
resulting vocal register: either falsetto register (in the case of a reduced needed for chest register
rotational component) or chest register (in the
presence of a rotational component)

production—partially, typically in the more anterior of complexity into the concept of sound quality control
part). The relative duration of the closed phase, often at the laryngeal level.
expressed through the closed quotient (i.e., the dura-
tion of the closed phase divided by the respective glottal LARYNGEAL EVIDENCE OF A
cycle’s period) typically corresponds with the strength REGISTER TRANSITION
of noteworthy harmonics (i.e., overtones) found in the
Any laryngeal sound production phenomenon in speech
voice source.68 Within limits, reached in “pressed phona-
and singing can be assessed on two levels: on a purely
tion,” a longer closed phase results in stronger overtones output-related level (thus treating the voice as a “black
and thus a more “carrying” voice. box”), only considering the radiated sound as docu-
A simplified explanation for this phenomenon—for mented by the acoustic signal; or also considering the
instance, disregarding nonlinear vocal tract influences— laryngeal dynamics of the voice production phenom-
is provided as follows: During each vibratory cycle, enon. A practical, noninvasive method for the latter
the glottis is open for a certain duration. During this approach is constituted by electroglottgraphy (EGG).
open phase, higher-frequency voice source harmonics The EGG signal is a physiological correlate of vocal fold
(overtones) are dampened out in the trachea. Therefore, contact during phonation, measuring the relative vocal
a shorter open phase, and thus a longer closed phase, fold contact area (VFCA). Within reason, it can provide
typically results in a “brighter” voice sound with stron- in many cases an approximate indication about the
ger overtones.69 Furthermore, as a first approximation, relative duration of vocal fold contact per glottal cycle,
the acoustic pressure (i.e., the sound) created in the thus serving as an inexact “proxy” for glottal closure.72
larynx is proportional to the rate of change of air flow, In section 2 of this article, the example presented in
with the main acoustic event typically occurring dur- Figure 1 was briefly discussed on perceptual grounds,
ing air flow deceleration at the end of the open phase. indicating that it contained two candidates for registra-
A quicker deceleration of the glottal air flow leads to tion events: one at t ≈ 3 s and the other at t ≈ 5.5 s. It was
more abrupt pressure changes (described in the concept claimed that the first transition at t ≈ 3 s was a laryngeal
of the maximum flow declination rate [MFDR]) in the register transition. This assumption is based on the EGG
larynx,70 producing increased acoustic output.71 Because data presented in Figure 1D and E. The dEGG wavegram
chest register typically has a longer closed phase than shown in Figure 1D provides an intuitive visualization
falsetto register, it is, as a rule of thumb, characterized of the development of relative duration of vocal fold
by stronger overtones. Note that in this context, the contact per glottal cycle.73 In a dEGG wavegram, overall
degree of (cartilaginous) adduction has an impact on time is mapped onto the x-axis, going from left to right.
the closed phase duration, introducing a further layer The relative distance between the dark horizontal line

November/December 2020 185

Christian T. Herbst

(representing the contacting event within each glottal setto, head, M2, or “thin.” Both females and males have
cycle) and the light horizontal line (representing the comparable laryngeal anatomy and morphology, and
de-contacting event) documents the temporal develop- they have access to the same laryngeal configurations
ment of the relative duration of vocal fold contact per through activity in the same intrinsic laryngeal muscles.75
cycle. Around t ≈ 3 s there is a clearly observable and It is speculated here that different vocal tract settings
relatively abrupt reduction of that relative contact dura- and resonatory strategies in various singing styles may
tion, as is typically seen when changing from chest to have an influence on whether that register is perceived
falsetto register. Because the EGG signal in many cases as “supported” or not, and this might then perhaps have
is an imprecise proxy of glottal closure, it can thus be an influence on the chosen terminology. Furthermore,
hypothesized that at t ≈ 3 s a reduction of the closed the confusion of falsetto register and head voice may
phase also occurred. That presumed reduction of the perhaps come from the somewhat surprising notion that
closed phase was likely brought about by relaxation of in temporary classical singing, males likely extend the
the TA (reducing the vertical bulging of the vocal fold chest register beyond the zona di passaggio (D4), mainly
and thus reducing membranous medialization of the applying resonatory adaptations, while females probably
vocal folds) and shortening of the glottal closure dura- seek a laryngeal register transition, switching from chest
tion by reducing the vertical phase delay between the to falsetto. Rigorous empirical research targeting both
inferior and superior vocal fold margins (via decrement voice production, perception, and potentially also pro-
of the rotational vibratory component, i.e., the x-11 prioception is required to shed more light on this matter.
mode—recall the previous discussion).
An alternative portrayal of that laryngeal register tran- PEDAGOGIC RELEVANCE
sition is provided in the left panel of Figure 1E. The two
It has been shown here that, broadly speaking, vocal
EGG waveforms, representing one glottal cycle (normal-
sounds produced in chest register typically have stron-
ized in time) immediately before and after the register
ger overtones than those produced in falsetto register,
transition clearly document the reduction of the relative
resulting in a “brighter” and thus more “carrying”
duration of vocal fold contact that is characteristic for
voice. This may give rise to the somewhat inconsider-
a chest-falsetto transition. In contrast, the laryngeal
ate assumption that it might be advantageous to always
evidence for the event occurring at t≈5 s (right panel
sing in chest register, rather than falsetto register. On
of Figure 1E) reveals a stable vibratory regimen, thus
physiological grounds, that notion might be supported
suggesting that that transition at t≈5 s is most likely not
by maximizing the closed phase (recall the discussion
of laryngeal nature, but is rather caused by resonance
above, highlighting the general relation between closed
effects (see section 3 for a discussion).
phase and occurrence of overtones), through extreme
TA dominance (relative to CT activity) and/or maximiz-
FALSETTO VS. HEAD VOICE REGISTER
ing vocal fold adduction. In the opinion of this author,
Particularly in the German speaking community, some however, such an extreme strategy would be certainly
authors seem to make a clear distinction between the unfavorable, to say the least, for the following reasons:
terms “falsetto” and “head” register. For instance, 1. Most probably, each voice has its own closed quo-
Seidner and Wendler suggest that the term falsetto tient limit, likely influenced by individual laryngeal
should be exclusively used in the context of male voice, anatomy and morphology. Such a limit should
while in females the register above the primo passaggio be determined and approached carefully in vocal
should be termed “head voice” (Kopfstimme).74 training. Surpassing this limit, typically through
However, when considering vocal registers at the excessive vocal fold adduction and helped by TA
laryngeal level via vibratory mechanisms, in the opin- dominance, will result in “pressed voice,” giving rise
ion of this author there is no reason to assume that the to an unwanted reduction of the overall amplitude
register above the chest is functionally any different in of the voice source spectrum and a reduction of the
females and males, regardless of whether it is called fal- MFDR and flow pulse amplitude.76 Furthermore,

186 Journal of Singing

 Registers—The Snake Pit of Voice Pedagogy, Part 1: Proprioception, Perception, and Laryngeal Mechanisms

such a phonation scenario even bears the potential Science-Based Theory of Register Phenomena,” in Second
risk of inducing vocal injury via increased vocal fold International Conference The Physiology and Acoustics of
collision force.77 Singing (PAS2) (Denver: National Center for Voice and
Speech, 2004), 64.
2. TA contraction has a more or less complex influence
on fundamental frequency.78 Higher frequency ranges   2. N. Henrich, “Mirroring the Voice from Garcia to the Present
Day: Some Insights into Singing Voice Registers,” Logopedics
can only be reached with a relaxed TA, facilitating
Phoniatics Vocology 31, no. 1 (2006): 3–14; J. A. Stark, Bel
falsetto register. In such a glottal configuration, the Canto: A History of Vocal Pedagogy (Toronto: University of
better part of the longitudinal tension is borne by Toronto Press, 1999).
the vocal fold cover, which is more suited for high   3. L. Zacconi, Prattica di musica utile et necessaria si al composi-
frequency oscillation, due to its biomechanical prop- tore per comporre i canti suoi regolatamente, si anco al cantore
erties.79 Failure to relax the TA may thus considerably per assicurarsi in tutte le cose cantabili, 2nd edition (Venice:
limit the achievable vocal range. (In plain terms, Carampello, 1596); M. Garcia, Mémoire sur la voix humaine
in falsetto register, higher pitches can typically be présenté a l’Académie des Sciences en 1840, vol. 2 (Paris:
reached, as compared to chest register.) Duverger, 1847); E. Behnke, “The Registers of the Voice,”
Proceedings of the Musical Association, 13th session (1886),
3. An excessive TA dominance likely impedes the abil-
1–16; J. van den Berg, “Vocal Ligaments versus Registers,”
ity to blend or mix the chest and falsetto registers The NATS Bulletin 20, no. 2 (November/December 1963):
(Part 2), potentially introducing audible register 16–31; W. Vennard, Singing: The Mechanism and the
breaks into the voice range. While such features Technique (New York: Carl Fischer, 1967); H. Hollien, “On
may be part of some singing style aesthetics, they are Vocal Registers,” Journal of Phonetics 2 (1974): 125–143; D.
certainly undesirable in classical singing. (In this con- G. Miller, “Registers in Singing: Empirical and Systematic
text, one may argue that some singing styles actually Studies in the Theory of the Singing Voice” (Doctoral
dissertation, University of Gronigen, 2000); B. Roubeau,
would never make use of falsetto register. However,
N. Henrich, and M. Castellengo, “Laryngeal Vibratory
even in such cases a certain degree of freedom in
Mechanisms: The Notion of Vocal Register Revisited,”
TA contraction [and vocal fold adduction] may be Journal of Voice 23, no. 4 (July 2008): 425–438.
required, allowing the singer to adeptly vary the cre-
  4. M. Mörner, G. Fransson, and G. Fant, “Voice Register
ated vocal timbre through voice source control on Terminology and Standard Pitch,” Speech Transmission
the laryngeal level.) Laboratory Quarterly Progress and Status Reports [STL-
QPSR] 4, no. 4 (1963): 17–23.
The discussion presented here suggests that, greatly
simplified, the singer’s choice of register is a com-   5. J. Sundberg, The Science of the Singing Voice (DeKalb, IL:
Northern Illinois University Press, 1987).
promise, influenced by the following considerations:
chest register produces stronger overtones; falsetto   6. van den Berg.
register produces higher pitches. Likely, the choice is  7. Roubeau.
not a (binary) dichotomy, but has to be made along a  8. Zacconi.
continuum. This is where mixing or blending the reg-   9. R. Miller, The Structure of Singing. System and Art in Vocal
isters comes into play, a notion that will be discussed Technique (New York: Schirmer Books, 1986).
in Part two. 10. K. Verdolini, D. Druker, P. Palmer, and H. Samawi,
“Laryngeal Adduction in Resonant Voice,” Journal of Voice
ACKNOWLEDGEMENTS 12, no. 3 (September 1998): 315–327.
11. J. Sundberg, “Phonatory Vibration in Singers: A Critical
I sincerely thank Kristen Murdaugh and Jan Svec for Review,” Music Perception 9, no. 3 (Spring 1992): 361–382;
their feedback to the manuscript. Any oversights and J. Sundberg, “Phonatory Vibration in Singers: A Critical
errors that remain are mine alone. Review,” STL-QPSR 32, no. 1 (1991): 37–51.
12. J. Sundberg, “Chest Wall Vibrations in Singers,” Journal of
NOTES Speech and Hearing Research 26, no. 3 (September 1983):
  1. L. Thurman, G. F. Welch, A. Theimer, and C. Klitzke, 329–340.
“Addressing Vocal Register Discrepancies: An Alternative, 13. Sundberg, Music Perception; Sundberg, STL-QPSR.

November/December 2020 187

Christian T. Herbst

14. T. Kitamura and K. Ohtani, “Non-contact Measurement 30. See, e.g., A. J. M. Houtsma and J. Smurzynski, “Pitch
of Facial Surface Vibration Patterns during Singing by Identification and Discrimination for Complex Tones
Scanning Laser Doppler Vibrometer,” Frontiers in Psychology with Many Harmonics,” Journal of the Acoustical Society of
6 (November 2015): 1682. America 87, no. 1 (1990): 3034–310; T. Rossing, The Science
15. H. Hollien, “Report on Vocal Registers,” in A. Askenfelt, of Sound (Boston: Addison-Wesley Publishing Co., 1990); J.
S. Felicetti, E. Jansson, and J. Sundberg, eds., Proceedings H. Johnson et al., “Just Noticeable Differences for Intensity
of the Stockholm Music Acoustic Conference 1983, Vol. 1 and their Relation to Loudness,” Journal of the Acoustical
(Stockholm: Royal Swedish Academy of Music, 1983), 27–35. Society of America 93, no. 2 (1993): 983–991; A. J. Simpson
and J. D. Reiss, “The Dynamic Range Paradox: A Central
16. Sundberg, “Phonatory Vibrations” (1991, 1992).
Auditory Model of Intensity Change Direction,” PLOS ONE
17. Ibid.
8, no. 2 February 2013): e57497.
18. L. Lehmann, Meine Gesangkunst (Wiesbaden: Bote & Bock,
31. See, e.g., K. Jensen, “The Timbre Model,” Journal of the
1922).
Acoustical Society of America 112, no. 5 (November 2001):
19. I. R. Titze, Principles of Voice Production, second printing 2238; T. Rahne, L. Böhme, and G. Götze, “Timbre dis-
(Salt Lake City: National Center for Voice and Speech, 2000). crimination in cochlear implant users and normal hear-
20. H. Hollien and J. F. Michel, “Vocal Fry as a Phonational ing subjects using cross-faded synthetic tones,” Journal of
Register,” Journal of Speech and Hearing Research 11, no. 3 Neuroscience Methods 199, no. 2 (August 2011): 290–295;
(September 1968): 600–604. D. M. Howard and A. M. Tyrell, “Psychoacoustically
21. H. Hollien and R. W. Wendahl, “Perceptional Study of Informed Spectrography and Timbre,” Organised Sound 2,
Vocal Fry,” Journal of the Acoustical Society of America 43, no. 2 (August 1997): S1355771897009011—and especially
no. 2 (March 1968): 506–509; M. Blomgren and Y. Chen, R. Carlson, B. Granström, and G. Fant, “Some Studies
“Acoustic, Aerodynamic, Physiologic and Perceptual Concerning Perception of Isolated Vowels,” STL-QPSR
Properties of Modal and Vocal Fry Registers,” Journal of the 11, no. 2–3 (1970); K. Stephens, “On the Quantal Nature of
Acoustical Society of America 103, no. 5, part 1 (June 1998): Speech,” Journal of Phonetics 17, 1–2 (January-April 1989):
2649–2658. 3–45.
22. A. Keidar, “An Acoustic-Perceptual Study of Vocal Fry 32. Lehmann, 27, 31, and 32.
Using Synthetic Stimuli,” Journal of the Acoustical Society 33. D. G. Miller, Resonance in Singing (Princeton, NJ: Inside View
of America 73, Supp. 1:s.3 (1983): 183–194. Press, 2008). For further details, see C. T. Herbst, W. T. Fitch,
23. J. Sundberg, “The Perception of Singing,” in D. Deutsch, ed., and J. G. Svec, “Electroglottographic Wavegrams: A Technique
The Psychology of Music, 2nd ed. (London: Academic Press, for Visualizing Vocal Fold Dynamics Noninvasively,” Journal
1999), 171–214. of the Acoustical Society of America 128, no. 5 (November
2010): 3070–3078.
24. A. Keidar, R. R. Hurtig, and I. R. Titze, “The Perceptual
Nature of Vocal Register Change,” Journal of Voice 1, no. 3 34. Garcia.
(1987): 223–233. 35. H. Hollien, “On Vocal Registers,” Journal of Phonetics 2, no.
25. G. L. Salamão and J. Sundberg, “Relation between Perceived 2 (March 1974): 125–143.
Voice Register and Flow Glottogram Parameters in Males,” 36. H. Hollien, H. Damsté, and T. Murry, “Vocal Fold Length
Journal of the Acoustical Society of America 124, no. 1 (July During Vocal Fry Phonation,” Folia Phoniatrica 21, no. 4
2008): 546–551. (1969): 257–265.
26. Sundberg, “The Perception of Singing.” 37. E. L. Allen and H. Hollien, “A Laminagraphic Study of Pulse
27. R. H. Colton and H. Hollien, “Perceptual Differentiation of (Vocal Fry) Register Phonation,” Folia Phoniatrica 25, no. 4
the Modal and Falsetto Registers,” Folia Phoniatrica 25, no. (December 1972): 2431–250.
4 (1973): 270–280. 38. R. E. McGlone and T. Shipp, “Some Physiologic Correlates
28. Ibid. of Vocal Fry Phonation,” Journal of Speech and Hearing
29. M. F. Davis, “The Psychoacoustics of Audio and Electro­ Research 14, no. 4 (December 1971): 769–775.
acoustics,” in T. D. Rossing, ed., Springer Handbook of 39. T. Murry and W. S. Brown, “Subglottal Air Pressure
Acoustics (New York: Springer, 2007), 747–757; D. M. During Two Types of Vocal Activity: Vocal Fry and Modal
Howard and J. A. S. Angus, Acoustics and Psychoacoustics, Phonation,” Folia Phoniatrica 23, no. 6 (1971): 440–449;
4th ed. (Oxford: Oxford University Press, 2009). McGlone and Shipp.

188 Journal of Singing

 Registers—The Snake Pit of Voice Pedagogy, Part 1: Proprioception, Perception, and Laryngeal Mechanisms

40. H. Herzel, d. Berry, I. R. Titze, and M. Saleh, “Analysis of 49. M. E. Edgerton, The 21st Century Voice: Contemporary and
Vocal Disorders with Methods from Nonlinear Dynamics,” Traditional Extra-Normal Voice, 2nd ed. (Lanham, MD:
Journal of Speech and Hearing Research 37, no. 5 (November Rowman & Littlefield, 2015).
1994): 1008–1019. 50. Edgerton, “Pitch Profile.”
41. R. Whitehead, D. Metz, and B. Whitehead, “Vibratory Patterns 51. J. Flanagan and L. L. Landgraf, “Self Oscillating Source for
of the Vocal Folds During Pulse Register Phonation,” Journal Vocal Tract Synthesizers,” IEEETransactions on Audio and
of the Acoustical Society of America 75, no. 4 (April 1984): Electroacoustics 16, no. 1 (March 1968): 57–64; K. Ishizaka
1293–1296. and J. L. Flanagan, “Synthesis of Voiced Sounds from a Two-
42. J. van den Berg, “Vocal Ligaments versus Registers,” The NATS Mass Model of the Vocal Cords,” The Bell System Technical
Bulletin 20, no. 2 (November/December 1963): 16–31; R. Journal 51, no. 6 (July-August 1972): 1233–1268.
Luchsinger and G. E. Arnold, Voice-Speech-Language: Clinical 52. For more detailed summaries, see Elemans et al., 2; C. T.
Communicology: Its Physiology and Pathology (Belmont, Herbst, “Biophysics of Vocal Production in Mammals,” in
CA: Wadsworth Publishing Co., 1965); F. Klingholz, Die W. T. Fitch, A. N. Popper, and R. A. Suthers, eds., Vertebrate
Akustik der gestörten Stimme (Stuttgart: Thieme, 1986); HK. Sound Production and Acoustic Communication (New York:
Herzel and R. Reuter, “Whistle Register and Biphonation in Springer, 2016), 328.
a Child’s Voice,” Folia Phoniatrica et Logopaedica 49, no. 5 53. I. R. Titze, “The Physics of Small-Amplitude Oscillation of
(February 1997): 216–224; W. R. Zemlin, Speech and Hearing the Vocal Folds,” Journal of the Acoustical Society of America
Science—Anatomy and Physiology, 4th ed. (Boston: Allyn and 83, no. 4 (April 1988): 1536–1552.
Bacon, 1998).
54. J. Flanagan, “Source-System Interaction in the Vocal
43. J. G. Svec, J. Sundberg, and S. Hertegard, “Three Registers in Tract,” Annals of the New York Academy of Sciences 155,
an Untrained Female Singer Analyzed by Videokymography, no. 1 (November 1968): 9–17; M. Rothenberg, “Acoustic
Strobolaryngoscopy and Sound Spectrography,” Journal of Interaction Between the Glottal Source and the Vocal Tract,”
the Acoustical Society of America 123, no. 1 (January 2008): in K. N. Stevens and M. Hirano, eds., Vocal Fold Physiology
347–353. (Tokyo: University of Tokyo Press, 1981), 305–328; I. R.
44. J. van den Berg, “Myoelastic-Aerodynamic Theory of Voice Titze, “Nonlinear Source-Filter Coupling in Phonation:
Production,” Journal of Speech and Hearing Research 1, Theory,” Journal of the Acoustical Society of America 123,
no. 3 (September 1958): 227–244; C. P. H. Elemans et al., no. 5 (May 2008): 2733–2749.
“Universal Mechanisms of Sound Production in Birds and 55. Ishizaka and Flanagan; J. Liljencrants, “A Translating and
Mammals,” Nature Communications 6 (November 2015): Rotating Mass Model of the Vocal Folds,” STL-QPSR 32, no.
8978. 1 (1991): 1–18.
45. F. Martienssen-Lohmann, Der wissende Sänger: Gesangslexikon 56. For a more detailed definition and discussion, see S. P.
in Skizzen, vol. 3 (Zürich: Atlantis Musikbuch-Verlag, 1981); Kumar, K. V. Phadke, J. Vydrová, A. Novozámaky, A. Zita,
Thurman; I. R. Titze and E. J. Hunter, “Normal Vibration Bkj. Zitová, and J. G. Svec, “Visual and Automatic Evaluation
Frequencies of the Vocal Ligament,” Journal of the Acoustical of Vocal Fold Mucosal Waves through Sharpness of Lateral
Society of America 115, no. 5, part 1 (May 2004): 2264–2269. Peaks in High-Speed Videokymographic Images,” Journal
46. M. Echternach, M. Döllinger, J. Sundberg, L. Traser, of Voice 34, no. 2 (March 2020): 170–178.
and B. Richter, “Vocal Fold Vibrations at High Soprano 57. T. Baer, “Investigation of Phonation Using Excised Larynxes”
Fundamental Frequencies, “ Journal of the Acoustical Society (Doctoral dissertation, MIT, 1975).
of America 133, no. 2 (February 2013): EL82-EL87; M. 58. D. A. Berry, H. Herzel, I. R. Titze, and K. Krischer,
Garnier et al., “Glottal Behavior in the High Soprano Range “Interpretation of Biomechanical Simulations of Normal
and the Transition to the Whistle Register,” Journal of the and Chaotic Vocal Fold Oscillations with Empirical
Acoustical Society of America 131, no. 1 (January 2012): Eigenfunctions,” Journal of the Acoustical Society of America
951–962. 95, no. 6 (June 1994): 3595–3604; J. G. Svec, “On Vibration
47. See, e.g., M. Garnier, N. Henrich, J. Smith, and J. Wolfe, Properties of Human Vocal Folds: Voice Registers,
“Vocal Tract Adjustments in the High Soprano Range,” Bifurcations, Resonance Characteristics, Development and
Journal of the Acoustical Society of America 127, no. 6 (June Application of Videokymography” (Doctoral dissertation,
2010): 3771–3780; and Part 2 of this article. University of Gronigen, 2000).
48. M. Edgerton, “Pitch Profile of the Glottal Whistle,” Malaysian 59. M. Döllinger, N. Tayama, and D. A. Berry, “Empirical
Journal of Science 32, no. 1 (June 2013): 78–85. Eigenfunctions and Medial Surface Dynamics of a Human

November/December 2020 189

Christian T. Herbst

Vocal Fold,” Methods of Information in Medicine 44, no. 3 Journal of the Acoustical Society of America 53, no. 6 (June
(January 2005): 384–391. 1973): 1632–1645.
60. M. Hirano, “Morphological Structure of the Vocal Cord as 70. E. B. Holmberg, R. E. Hillman, and J. S. Perkell, “Glottal
a Vibrator and its Variations,” Folia Phoniatrica 26, no. 2 Airflow and Transglottal Air Pressure Measurements for
(1974): 89–94; O. Fujimura, “Body-Cover Theory of the Male and Female Speakers in Soft, Normal, and Loud Voice,”
Vocal Fold and its Phonetic Implications,” in Stevens and Journal of the Acoustical Society of America 84, no. 2 (August
Hirano, 271–288. 1988): 511–529.
61. See, e.g., Titze, Principles of Voice Production, 11 and 71. G. Fant, “Glottal Source and Waveform Analysis,” STL-QPSR
Figure 1.10c; Zemlin, 128–129. 1 (1979): 85–107.
62. Cf. this author’s contribution about independent laryngeal 72. However, for some important caveats when interpreting
control of membranous medialization and cartilaginous EGG signals and derived parameters, see C. T. Herbst,
adduction, C. T. Herbst and J. G. Svec, “Adjustment of “Electroglottography—An Update,” Journal of Voice 34, no.
Glottal Configuration in Singing,” Journal of Singing 70, no. 4 (March 2019): 503–526; doi:10.10016/j.jvoice.2018.12.014.
3 (January/February 2014): 301–308.
73. For details, see Herbst et al., “Electroglottographic
63. But note that data from electromyographic studies suggest Wavegrams.”
that activity of the CT is more relevant to regulation of
74. W. Seidner and J. Wendler, Die Sängerstimme, 4th ed. (Berlin:
fundamental frequency than vocal registers: M. Hirano, W.
Henschel Verlag, 2004).
Vennard, and J. Ohala, “Regulation of Register, Pitch and
Intensity of Voice,” Folia Phoniatrica 22 (1970): 1–20; K. A. 75. Cf. Herbst, “Membranous and Cartilaginous Vocal Fold
Kochis-Jennings, E. M. Finnegan, H. T. Hoffman, and and Adduction in Singing.”
S. Jaiswal, “Laryngeal muscle activity and vocal fold adduc- 76. C. T. Herbst, M. Hess, F. Müller, J. G. Svec, and J. Sundberg,
tion during chest, chestmix, headmix, and head registers in “Glottal Adduction and Subglottal Pressure in Singing,”
females,” Journal of Voice 26, no. 2 (March 2012): 182–193. Journal of Voice 29, no. 4 (July 2015): 391–402.
64. J. Yin and Z. Zhang, “The Influence of Thyroarytenoid and 77. K. Verdolini, R. Chan, M. Hess, and W. Bierhals,
Cricothyroid Muscle Activation on Vocal Fold Stiffness “Correspondence of Electroglottographic Closed Quotient
and Eigenfrequencies,” Journal of the Acoustical Society of to Vocal Fold Impact Stress in Excised Canine Larynges,”
America 133, no. 5 (May 2103): 2972–2983. Journal of Voice 12, no. 4 (December 1998): 415–423.
65. Titze, Principles of Voice Production, 291. 78. Yin; D. K. Chhetri, J. Neubauer, E. Sofer, and D. A. Berry,
66. Ibid., 289–290. “Influence on Interactions of Laryngeal Adductors and
67. C. T. Herbst, S. Ternström, and J. G. Svec, “Investigation of Cricothyroid Muscles on Fundamental Frequency and
Four Distinct Glottal Configurations in Classical Singing—A Glottal Posture Control,” Journal of the Acoustical Society
Pilot Study,” Journal of the Acoustical Society of America 125, of America 135, no. 4 (April 2014): 2052–2064.
no. 3 (March 2009): EL104-EL109; C. T. Herbst, Q. Qiu, H. 79. Titze, Principles of Voice Production.
K. Schutte, and J. G. Svec, “Membranous and Cartilaginous
Vocal Fold Adduction in Singing,” Journal of the Acoustical Christian T. Herbst studied voice pedagogy at Mozarteum University,
Society of America 129, no. 4 (April 2011): 2253–2262; D. Salzburg, and worked as a voice pedagogue for two decades. Driven by
K. Chhetri, J. Neubauer, and D. A. Berry, “Neuromuscular his interest in the physics and the physiology of voice, he enrolled in a
Control of Fundamental Frequency and Glottal Posture PhD program in Biophysics at the University of Olomouc, Czech Republic.
at Phonation Onset,” Journal of the Acoustical Society of
Since his graduation in 2012, he has been working as a full time voice
America 131, no. 2 (February 2012): 1401–1412.
scientist. As of November 2019, he accepted a position at the University
68. J. Flanagan, “Some Properties of the Glottal Sound Source,” of Music and Performing Arts Vienna, Antonio Salieri Department of Vocal
Journal of Speech and Hearing Research 1, no. 2 (June 1958): Studies and Vocal Research in Music Education. The focus of Christian’s
99–116. work is on both basic voice science and the acoustics and physiology
69. M. Rothenburg, “A New Inverse-Filtering Technique for of the singing voice. He received several international scientific awards,
Deriving the Glottal Air Flow Waveform during Voicing,” and has published, among others, in the prestigious Science journal.

190 Journal of Singing