Printer Friendly

Breathing for singers: a comparative analysis of body types and breathing tendencies.


The fat opera singer remains a stereotype so pervasive that it has spawned a cliche applied to everything from sports to politics. Ask any person who is not knowledgeable of the genre and he or she may describe opera as fat ladies with voices that can shatter glass. But like all singers, classically trained singers appear in all shapes and sizes, and while some opera singers may carry a few extra pounds, others often possess slight or even athletic frames. Does this physical diversity add to the challenges of teaching breathing for singing? Should a voice teacher vary his or her breathing technique based on a singer's body type? The relationship between body type and breathing for singing remains an underinvestigated area of research. As a partial remedy to that circumstance, this study seeks to clarify the relationship between a person's body type and his or her breathing tendencies while singing.

Generally, voice pedagogues support the notion that breathing, or breath management, constitutes a crucial technical element in classical singing, fundamental to efficient singing and healthy vocal technique. Miller, for example, writes, "Breath management is the essential foundation for all skillful vocalism." (1) Alderson argues that "breath is the foundation on which singing is established and good breathing is the basis for all good singing." (2) Vennard writes, "If his breathing can be improved his singing can also." (3) Still, although voice teachers tend to agree that breathing is of extreme importance while singing, those same teachers are likely to disagree about which breathing technique is the most effective and efficient. Indeed, costal (or intercostal), pancostal, and appoggio breathing techniques all appear to be supported and utilized to some extent.

In costal breathing technique, the singer's rib cage area remains expanded while singing. Alderson defines costal technique as rib breathing, where "the intercostal muscles expand the rib cage and hold it except during long rests, at which time the ribs draw in to a more normal position." (4) In appoggio technique, the singer's epigastric region remains expanded while singing. Miller writes, "With regard to breath management, appoggio maintains for a remarkable period of time a posture near that which pertained at the beginning of the inspiratory phase of the breath cycle." (5) He also suggests, "Both the epigastric and the umbilical regions should be stabilized so that a feeling of internal-external muscular balance is present. This sensation directly influences the diaphragm." (6) In contrast to this, pancostal breathing technique requires the singer to pull the abdominals inward tightly while singing. Vennard defines pancostal technique as "breathing for singing while holding in the abdominals strongly, on the theory that this would maintain a high central tendon, causing the diaphragm to raise the ribs; chest breathing." (7)

Although no empirical evidence currently exists, initial anecdotal observation on the part of the researcher suggests that all of the above breathing techniques might be influenced by a singer's specific body type. Specifically, I have observed that 1) overweight individuals tend to breathe lower in the abdominal region than those with less body fat; 2) lean individuals seem to breathe higher in the thoracic region; and 3) singers with athletic builds, appear to breathe in the rib cage area.

In 1940, William Sheldon developed a system that was later shown to be a reliable method for determining body type. His method, somatotyping, was a process of quantifying fat measurements at selected locations of a person's body. The resulting figures correlated with one of three categories: endomorphs (relative fatness), mesomorphs (relative musco-skeletal robustness), and ectomorphs (relative linearity). (8)

In 1980, J. E. Carter published a manual based on Sheldon's findings, further defining somatotyping as "the quantification of the present shape and composition of the human body." (9) According to Carter, endomorphs possess a higher percentage of body fat than of muscle mass or leanness and tend toward roundness in physique rather than toward muscularity or leanness. Mesomorphs evidence more muscle mass than body fat and possess larger bone structures. Ectomorphs, on the other hand, tend to appear lean with limited body fat and muscle mass. (10)

In 1986, Hoit and Hixon measured twelve subjects and placed them into somatotype groups using Carter's formula. For the purpose of their study they required participants that displayed prominence in one component of body type. (11) Therefore, they modified previous somatotyping classification protocols. Specifically, in accordance with the formula, prominence of ectomorphy or mesomorphy was assigned as "a rating of at least three points higher ... than the other two components rated." (12) Endomorph prominence, however, was given only half a point higher as a primary component than the other two somatotypes. According to Hoit and Hixon, "a more lenient criterion for endomorph prominence was required because of difficulty in finding individuals with high endomorphy rating accompanied by relatively low mesomorphy rating." (13)

The present study adds to previous research in the domain of voice pedagogy by adapting Hoit and Hixon's methodology in order to explore body types of young

female singers, and to investigate the extent that a correlation exists between those singers' body types and

their breathing tendencies. To that end, the following

research questions guide this investigation: 1) To what extent do athletic singers breathe differently than their slim or stout counterparts? 2) Of what consequence is this predisposition to various voice teachers? 3) Should voice teachers modify their teaching styles to take advantage of this predisposition?



Thirty volunteer participants underwent the somatotyping process. For the purposes of this preliminary study, only participants who clearly fit into one of Hoit and Hixon's three categories were used. Therefore, of the original thirty volunteers, only twelve matched the criteria for extreme endomorph, mesomorph, and ectomorph, and completed the full research project. The participants were all Caucasian women, who ranged in age from 18 to 22 years, and were within their first two years of vocal study. All were in good health with no known vocal, speech, or lung ailments, or history of tobacco use.


I utilized the anthropometric somatotyping method, which requires the measurement often dimensions, including stretch stature or height, body mass or weight, four skinfolds (triceps, subscapular, supraspinale, medial calf) to determine body fat percentages, two bone breadths (biepicondylar humerus and femur) to determine muscle mass percentages, and two limb girths (arm flexed and tensed, calf). (14) Endomorphy was calculated using the following formula: (-0.7182) + (0.1451 x subscapular measurement) - (0.00068 x supraspinale measurement) + (0.0000014 x medial calf measurement). Mesomorphy was calculated using the formula (0.858 x breadth of humerus) + (0.601 x breadth of femur) + (0.188 x measurement of biceps) + (0.161 x measurement of flexed calf) - (0.131 x weight in kilograms) + 4.5. Ectomorphy was calculated using the formula (0.732 x height/weight ratio) - 28.58.

These formulas resulted in a somatotype rating for each participant on each of the somatotype scales. Prominence toward a specific body type was determined by the ratios between the figures. For ectomorphs the figure was at least three points higher than the mesomorphic and endomorphic figures. For mesomorphs a figure of three points higher in mesomorphy was required. For endomorphs, only one half-point over the other two somatotypes was required. (15) We placed each subject into one of the three anthropometric categories according to the final results (Table 1).

Video Recording

Each participant was video recorded while singing the first four measures of "Caro mio ben" using a Panasonic VHSC model PV-L550D video recorder positioned 3.0 meters from the singer. Positions were marked on the floor for the tripod of the video recorder and the feet of the singer. A white cloth was hung on the wall behind the participants to increase the contrast between the participants and background. Participants stood with their left side toward the camera and wore white tank tops and slacks. They were in front of a contrasting backdrop so that the movement of the breath would be more visible. To facilitate measurement, strips of colored tape were positioned on the participants at the level of the chest, the base of the rib cage, and the umbilicus.

Each participant breathed at identical places in the phrase, which were the initial breath and a breath after each of the next three measures. The key of E-flat major and a preestablished tempo of a quarter note equals 72 in common time were the same for each singer. Only the researcher and the individual participant were in the room at any given time. Participants were given the initial pitch of E-flat and sang a cappella. They received several opportunities to rehearse before the recording process began. To determine the movement patterns of the torso by each subject, a side view of the torso was videotaped. As described earlier, each subject took breaths at identical places in the song. The extent of lateral movement during exhalation was measured at the lateral chest, base of rib cage, and umbilicus. In addition, the extent of vertical movement was measured at the shoulder. All three somatotype groups exhibited movement at all four measurement locations. The movements were measured by use of a millimeter scale.

Breathing Assessment

The videotape was uploaded to a computer and converted from analog video to digital video files using the computer program Adaptec Video Capture 1.1. Each participants video was saved as a file on a computer. The video file was viewed on a 17-inch monitor, and the movement at the chest, base of rib cage, and umbilicus was measured by placing a millimeter scale on the monitor and measuring the width of the image of the trunk at the measurement points in millimeters.

The maximum expansion at the end of each inhalation and the minimum position at the end of each exhalation, or sung breath, were measured, frame by frame, with a millimeter ruler. The differences between the two measurements were recorded at each of the measurement points. These differences represented the extent of lateral chest wall movement during the singing on each breath. Finally, the vertical movement of the thorax was measured. This measure was called vertical chest movement. All of the measurements reflected relative change as opposed to the actual magnitude of the change. The participants used four breaths each in singing the test song segment. The mean of each set of measurements for each somatotype grouping was then determined. These data were statistically analyzed using analysis of variance (ANOVA) techniques. The ANOVA used was a 3x3x3 design using the three somatotypes, three breathing styles, and three kinds of measurement as the levels of the factors. In addition, the breathing measurements were correlated to determine any relation with one of the three stated breathing techniques, that is, appoggio, pancostal, or costal.

Lung Function Assessment

Spirometry was used to ensure that each subject had normal lung function at the time of the measuring. Spirometry is defined as a "versatile test of pulmonary physiology." (16) For this study, emphasis was placed on spirometric measures related to vital capacity. Each singer formed an airtight lock around the mouthpiece of the spirometer with her lips and blew as hard and as long as possible into the spirometer so that forced inhalation and exhalation after a deep breath was measured. The spirometer recorded the amount of air and rate of air breathed in and out over a specified time. The spirometer then graphed the expired air and measured the volume of the air and the rate of the airflow. The FEV (forced expiratory volume), FVC (forced vital capacity), and the FEV/FVC were measured to analyze lung function. The FEV measured the volume of air expelled in the first second of maximal forced expiration. (17) The FVC measured the maximum amount of air that can be forcefully expired from a position of full inspiration. (18) The FEV/FVC measured the ratio of the forced expiratory volume in the first second to the forced vital capacity of the lungs. The normal value for this ratio is 70. The subjects repeated the spirometer test three times and the average of those three times was used in calculating breath lung function.


Lung Function Assessment Results

The measurements of lung function revealed the following results for FVC. The ectomorphs exhibited a mean FVC of 3.46 liters with a standard deviation of 0.26 liters; the mesomorphs exhibited a mean FVC of 3.51 liters with a standard deviation of 0.30 liters; and the endomorphs exhibited a mean FVC of 3.71 liters with a standard deviation of 0.40 liters. As shown in Figure 1, there was very little difference among and within the somatotype groups. The analysis of variance (ANOVA) confirmed this observation, as no significant differences were found among the somatotype groups [F (2, 9) = 0.692, p =.525, [eta][p.sup.2] =.133].


Lung function measurements revealed the following FEV results. The ectomorphs exhibited a mean FEV of 2.88 liters with a standard deviation of 0.14 liters; the mesomorphs exhibited a mean FEV of 3.00 liters with a standard deviation of0.30 liters; and the endomorphs exhibited a mean FEV of 3.14 liters with a standard deviation of 0.25 liters. An ANOVA revealed no significant differences among the somatotype groups [F (2,9) = 1.193, p = .347, [[eta].sup.2] = .210].

Using the lung function measurements, the following results for FEV/FVC-% could be determined. The ectomorphs exhibited a mean FEV/FVC-% of .84 with a standard deviation of .10; the mesomorphs exhibited a mean FEV/FVC-% of .86 with a standard deviation of .09; and the endomorphs exhibited a mean FEV/FVC% of .85 with a standard deviation of .04. An ANOVA revealed that no significant differences were among the somatotype groups. [F (2,9) = 0.057, p = .944, [[eta].sup.2] = .013] (Figure 2).

The spirometer revealed the following results for peak flow. The ectomorphs exhibited a mean peak flow of 5.80 liters per second with a standard deviation of 0.70 liters per second; the mesomorphs exhibited a mean peak flow of 5.49 liters per second with a standard deviation of 0.69 liters per second; and the endomorphs exhibited a mean peak flow of 6.48 liters per second with a standard deviation of 0.48 liters per second. An ANoVA revealed no significant differences were among the somatotype groups [F (2, 9) = 2.547, p = .131, [[eta].sup.2]= .364] (Figure 3).



Breathing Results

The differences between the somatotypes were compared using a repeated measure ANOVA. The analysis revealed that there were significant differences among the somatotypes in terms of the area of movement and place of breath. [F (2,9) = 6.206, p = .020, [[eta].sup.2] = .580] (see Figure 2). The partial eta squared effect size of .58 indicates a large effect size. (19) This indicates differences in movement related to the somatotype. The four places of breath movement presented in Figure 4 show the pattern of breathing for each of the three somatotype groups (endomorphs, mesomorphs, and ectomorphs).


Because the breath movement data had different degrees of variability across the four places, lateral chest, rib cage, umbilicus, and vertical chest, a correction factor needed to be used for the degrees of freedom. An analysis of the similarity of variation among the measured levels of the variables indicated a need to lower the degrees of freedom in order to achieve appropriate ANOVA results. A Greenhouse-Geiser correction factor was used to complete the ANOVA. Analysis revealed significant differences among the somatotypes in where the breath movement occurred [F (1.5, 13.6) = 5.047, p = .030, [[eta].sup.2] = .359]. The [[eta].sup.2] indicates a large effect for this statistic, meaning that the somatotype was a major factor in the breathing pattern differences.


The purpose of the present investigation was to determine whether or not beginning level, young female voice students of different body types differed in their breathing tendencies while singing, and whether these tendencies correlated to the stated breathing techniques. The similar spirometric data for all of the singers indicated that any differences that occurred were not because of different vital capacities or rate of air flow. The endomorphs exhibited breath movements that were lower in the thoracic region than the other two somatotype groups. The breath movements were most prominent in the umbilicus area.

Initially, anecdotal observations were made that singers with more body fat had a tendency to breathe lower than those who were lean or muscular. The results of the research demonstrate that people in this study with more fat tended to breathe lower in the thoracic region than those who were more lean or muscular. This result correlated with one of the stated breathing techniques, the appoggio technique; however, the data also showed that along with marked breath movements in the umbilicus area there was a large amount of vertical chest movement as well. The appoggio technique does not advocate vertical chest movement. On the contrary, it maintains that the chest should find a comfortably elevated position and remain stabilized throughout the breath cycle.

Mesomorphs displayed large amounts of rib cage and vertical chest movement while exhibiting little movement in the umbilicus area. Initially, anecdotal observations were made that singers who were muscular would tend to breathe primarily in the rib cage area. This would correlate with the costal technique. The results of the study, however, showed the greatest movement in the vertical chest and the second greatest movement in the rib cage area. Therefore, although rib cage movement was high among the mesomorphs in this study, vertical chest movement was greater. As in the case of endomorphs, mesomorphs demonstrated high vertical chest movement. High vertical chest movement is not advocated in the costal technique; instead a stabilized chest is used while breathing for singing.

Ectomorphs followed a similar pattern of movement as the mesomorphs; however, they displayed the highest degree of movement in the lateral and vertical chest wall. Initially, anecdotal observations were made stating that singers who are thin and lean tend to breathe higher in the thoracic region, which correlates with the pancostal technique. although their lateral chest movement was significant, there was a higher degree of vertical chest movement among ectomorphs. Again, the pancostal technique does not advocate vertical chest movement.

All three somatotype groups exhibited a preponderance of vertical chest wall movement. These movements probably were indicators of a lack of training among these singers. These data indicated that all of these singers needed to adapt their breathing technique to reduce the vertical chest wall movement as part of their singing training. in any breathing for singing technique it is crucial that the chest wall remain stabilized in order to facilitate a healthy breathing technique. Vertical breathing in singing is discouraged in any healthy breathing for singing technique.


The small sample size of the study limits the ability to generalize its results. Future investigations should include a larger sample size with the same criteria. The larger sample size would be helpful in determining if the vertical chest wall movements and separate movement patterns of the three somatotypes observed among these singers are due simply to inexperience. It will also be necessary to complete a further study using male singers as participants. For example, Hoit and Hixon's results for the speech breathing of men were similar to the results of the present investigation. Although the measurements were focused on speech breathing, their study showed marked similarities with the present investigation. Endomorphs demonstrated abdomen movements in the Hoit and Hixon study just as they showed movement in the abdomen area (specifically the umbilicus) in the present study. Ectomorphs in their study breathed more in the rib cage area than chest area; in other words, they breathed higher than the endomorphs just as in the present study. Finally, mesomorphs in Hoit and Hixon's study exhibited a combination of breath movement characteristics as did mesomorphs in the present study.

Further research could be done to compare men and women in terms of breath movement while singing. Also comparisons could be made between breathing for singing and breathing for speech.


The data showed marked correlations between body type and breathing tendencies that corresponded with the stated anecdotal observations. For example, endomorphs showed movement in the umbilicus area (appoggio), mesomorphs showed movement in the rib cage area (costal), and ectomorphs showed movement in the lateral chest area (pancostal). Because there was a large prominence of vertical chest movement in all three somatotype groups, it is difficult to relate the subject's breath tendencies to the stated breathing techniques (appoggio, costal, and pancostal). None of the breathing techniques mentioned condones vertical chest movement; in fact, they require the opposite, that is, a stabilized chest while breathing for singing. Although all singers displayed elements of the various stated techniques, they also displayed aspects of breathing that did not correlate with those techniques. some had a combination of several places of movement, and all three body types showed considerable vertical chest movement, which is forbidden in any solid breathing for singing technique.

The prominence of vertical chest movement is significant because it shows that young singers, regardless of body type, tend to use vertical chest movement when inhaling for singing. This tendency among all body types points to a lack of understanding of proper breath function; it further suggests that understanding and facilitating a healthy breathing technique for singing is a learned experience and requires training.

These findings demonstrate that 1) singers with different body types do tend to breathe differently, and 2) that beginning level singers of every body type show a prominence of large vertical chest breaths. Knowing which body types have which breathing tendencies will be helpful for the teacher of singing in determining where to look for each students breath movement, and in guiding the student accordingly.


The successful completion of the treatise and the adaptation to article form is due to the support of Dr. Richard J. Morris, associate professor from the School of Communication Science and Disorders Department at Florida State University. Without his expertise this project would not have been possible. In addition, I would like to thank Larry Gerber, professor of voice at Florida State University (director of the treatise), Yvonne Ciannella, professor of voice emeritus at Florida State University, and Roy Delp, professor of voice at Florida State University for their support and guidance of this project.


(1.) Richard Miller, Training Soprano Voices (New York: Oxford University Press, 2000), 32.

(2.) Richard Alderson, Complete Handbook of Voice Training (New York: Parker Publishing Company, 1979), 28.

(3.) William Vennard, Singing; the Mechanism and the Technic (New York: Carl Fischer, 1967), 18.

(4.) Alderson, 36.

(5.) Richard Miller, The Structure of Singing: System and Art in Vocal Technique (New York: Schirmer Books, 1986), 24.

(6.) Ibid., 24.

(7.) Vennard, 243.

(8.) James and Tyra Arraj, Tracking the Elusive Human, Volume 1, Inner Explorations,

(9.) J. E. L. Carter, The Heath-Carter Anthropometric Somatype (San Diego: J. E. L. Carter, 1980), 2.

(10.) Ibid.

(11.) Jeannette Hoit and Thomas Hixon, "Body Type and Speech Breathing," Journal of Speech and Hearing Research 29, no. 3 (September 1986): 314.

(12.) Ibid.

(13.) Ibid.

(14.) Carter, 2.

(15.) Ibid.

(16.) Thomas Gross, "Internal Medicine," The Virtual Hospital, 01 June 2002, SpiromethryHome.html

(17.) Andrew Coker, ed., "Pulmonary Function Test" The Health Network, 03 December 2000, http://www.healthnetwork. com/au/search-display.php?cat=tests&name=PULMONARY %20FUNCTION%20TEST.

(18.) Ibid.

(19.) Ibid.

Soprano Jennifer Griffith Cowgill is currently an Assistant Professor of Voice at the University of Alabama. She has performed in recitals, operas, oratorios, and musicals throughout the nation, including solo performances with The Atlanta Choral Guild and the Atlanta Ballet Orchestra. Her stage credits include roles in The Magic Flute, The Coronation of Poppea, The Seagull, Pirates of Penzance, and Little Red Riding Hood. Her most recent music theater roles include Lady Larken in Once Upon a Mattress and Laurey in Oklahoma. Dr. Cowgill received the Doctor of Music in vocal performance from Florida State University and the Master of Music from Peabody Conservatory at Johns Hopkins University. She earned her bachelor's degree in vocal performance from Oberlin Conservatory where she studied with the internationally renowned voice pedagogue Richard Miller. In addition to her college studies she was awarded fellowships to participate in such summer music festivals as The Tanglewood Festival, The Chautauqua Summer Institute and the Dicapo Opera Young Artists Program in New York City.
TABLE 1. Somatotyping results.

Subject      Endo          Meso       Ecto             Final

EN1            6             5        0.1             6-5-0.1
EN2            7             5        0.1             7-5-0.1
EN3            5             3        0.1             5-3-0.1
EN4            6             5        0.1             6-5-0.1

ME1            3             6        1               3-6-1.0
ME2            2             5        2               2-5-2.0
ME3            3             6        0.1             3-6-0.1
ME4            3             6        0.1             3-6-0.1

EC1            1             1        4               1-1-4.0
EC2            1             1        4               1-1-4.0
EC3            1             1        4               1-1-4.0
EC4            1             1        4               1-1-4.0
COPYRIGHT 2009 National Association of Teachers of Singing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Cowgill, Jennifer Griffith
Publication:Journal of Singing
Article Type:Report
Geographic Code:1USA
Date:Nov 1, 2009
Previous Article:The end of breath for singing: exhalation and the control of breath at the end of the phrase.
Next Article:The teachings of Jean-Antoine Berard: content, context, and legacy.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |