The American vocal pedagogue, Kenneth Phillips, has this to say about what he terms the modern “English choirboy”(1). Although he acknowledges that “the English model . . . widely admired . . . does produce a beautiful quality in the upper register from C5 to C6” he does not consider it desirable of emulation, for
English choirboys sing only one vocal part – treble . . .the head voice sound is extended below pitch C5 as low as possible. In order to keep the sound pure, breath support must be diminished, and the singing must be very light; any sound below pitch E4 becomes lifeless and does not project . . . (pp92, 93)
A beautiful quality in the upper register serves the Edwardian choral repertoire of the Stanford idiom well, but its appropriateness for other genres may be rather more questionable. The present author has had for some time a particular interest in the sacred choral music of the sixteenth century, when boys sang at lower pitches than they do today(2). Prior to the Reformation, there were two voice parts for boys. The meane ranged from Ab3 to D5, and the treble from C4 to G5 at today’s pitches. After the Reformation, the treble voice fell out of use and meane parts tended to be written slightly higher, most commonly from C4 to D5. Clearly in either case, if the singing of modern English choirboys is “very light . . . lifeless. . . and fails to project below pitch E4”, there are going to be issues with Renaissance choral music when sung by boys.
There are two possible reasons for there hitherto being little attention paid to this difficulty. The first is that many choirs perform the music transposed up a minor third, which renders it easier for the boys to sing parts that would be designated mean, and for falsettist counter-tenors to sing parts that would have been designated contratenor. The second is that when more scholarly, historically informed performances are attempted, these are almost invariably by adult choirs specialising in the music of the period. As Peter Phillips wrote:
” . . . the problems set by Tallis’s anthems are greatly expanded in his Latin music. To the AATB base was added at least one and often two higher parts, the nature of which has caused a lot of argument centred on whether they were both boys’ parts. If they were, upward transposition from original pitch seems inevitable, at least if modern boys are to sing them.” (3)
An inability to sing effectively below E4 is not an inevitable consequence of being a “modern boy”. It is more the case that it is a difficulty confronting the “English choirboy”, as described by Phillips. The author has twice worked with boys in Germany and more recently in the United States and on all three occasions has been impressed by the significantly more robust tone achieved at lower pitches. Phillips does acknowledge that some English choir directors have been influenced by a “Continental sound that introduces some chest mixture into the pure upper register”. He quotes Richard Miller’s 1977 observation that “some English choirmasters are beginning to question its (i.e. “English choirboy”) appropriateness to all choral situations”(4). This may well be so. References are indeed made from time to time to something called “continental tone”. Nevertheless, the present author’s investigations of English choristers suggest that movement in this direction has been haphazard. The majority of boys seen in research (5) still produced a characteristic single register acoustic profile of spectral lines undisplaced by register change, exemplified in Figure 1.
Weakness in that part of the range essential for effective performance of Renaissance meane parts at original pitch is commonly the result. This paper reports a preliminary investigation of the phenomenon undertaken with a view to historically informed performances by boys.
Registration in children’s singing
The question of registration in children’s voices has excited much debate and disagreement. Phillips has a clear view. He talks of a “CT register” in which cricothyroid action is dominant and a “TA register” in which thyroarytenoid action is dominant. (6) Between the two, there is a middle or mixed register in which the CT and TA are in balance. There is, he asserts, a 50-50 balance of registers at approximately F#4. This, at least, is what he urges his readers to aim for in training children’s voices. Critically, he locates the middle register as being from C4-C5. In other words, it is the principal register of the sixteenth century mean, and a register not possessed by “modern (English) boys”.
Jenevora Williams provides an account that differs somewhat in detail. She recognizes high pitches made with long, thin vocal folds, as would be produced by a dominance of cricothyroid action (M2). Similarly, lower pitches are made with shorter, thicker vocal folds as would be the case in thyroarytenoid dominance (M1). However, she does not describe a mixed, middle register. Instead she suggests that the transition from thick to thin fold can be made over a wide range of pitches. Almost the entire range can be sung either thick or thin fold. The choice of voice quality in the mid-range can be purely stylistic, though with the rider that both thin fold taken too low and thick fold taken too high are “hopelessly inefficient”(7).
Neither of these accounts says much about resonance and the ability of the singer to modify tone over and above what is attributable purely to M1 or M2 larynx mechanism. Castellengo et al concluded from an empirical investigation of five adult singers that the mixed register can equally be produced by M1 or M2. The singers made intuitive adjustments to both mechanism and resonance such that “the aesthetic finality of voix mixte is to simulate the sound quality of another mechanism (M2 when in M1, M1 when in M2).”(8) Whether or not the same conditions apply to children’s voices that have yet to achieve maturity in features such as the vocal ligament is an important consideration.(9) Also important are misconceptions about what “continental tone” might be. Some writers have simply equated it with so-called “chest voice” and describe the ugly tone that results from “chesting up” (i.e. carrying the modal speech voice as high as it will go in singing).
As far as the author has been able to ascertain, this is not the practice at Westminster Cathedral, the English establishment most associated with “continental tone” where boys also sing alto. The vocal coach at Westminster has described how the boys achieve a middle register by “mixing down”. Phillips agrees that the technique of blending registers is “best learned from the top down,” stating further that “Early attempts to blend these registers with vocalises from lower to upper may prove difficult; students tend not to shift out of the chest voice as they ascend” (10). So, the term “chesting up” may not be without foundation and the fact that the author has hardly ever witnessed downward vocalisation in the warm-ups of English cathedral choristers is relevant. Almost invariably, English conductors were seen to warm boys’ voices by ascending exercises rising to ever higher pitches.
Complications arising from puberty
Phillips also writes about what he calls the “infamous English voice break”, which he defines as “ignoring the voice-change problem among adolescent males” and keeping boys “singing in the upper voice as long as possible until the voice ‘breaks’”(13). To attribute this practice to the Royal School of Church Music, as Phillips does, is a perhaps a little dated as the RSCM has been at the forefront of promoting good practice with changing voices since introducing its Voice for Life scheme in 1999.(14) Nevertheless, the practice of keeping boys singing in the upper voice as long as possible remains widespread in English cathedrals, with the result that there are boys singing “treble” right through to the final stages of puberty when their speaking voices have reached adult baritone pitch. (15)
English vocal coaches remain divided as to whether such practice is harmful, but there is little doubt that the use of falsetto to prolong the “treble” career plays havoc with the neat order of events charted by researchers such as John Cooksey. Cooksey’s research has, over the last fifteen or so years, become widely accepted by English vocal coaches and is undoubtedly relevant to the question of how boys might render historically informed performance. In 1992, Cooksey wrote that “unchanged voices reach [their] climax of beauty and fulness” near the end of Grade 5 (UK Y6), lasting sometimes to early grade 7 (UK Y8)”(17). He wrote that for part assignment, the pre-mutational boy “usually sings soprano, but [is] also capable of singing soprano II or alto”(18). As has already been discussed, this may be so in the United States when boys are taught the full vocal range but is rather less commonly the case in England where relatively little attention is paid to the range below E4. Cooksey identified in 1992 the “initial pubertal period” as beginning in grade 6 (UK Y7), with the majority in grade 7 (UK Y8). For this stage he wrote “usually sings alto in SATB music”. It is at this point that current English practice departs somewhat from most of the rest of the Western world. An enquiry into historical performance practice, however, needs to ask whether this has always been the case.
Before this question can be answered the complication regarding the ages of sixteenth century choristers and the time during their career that they would have reached the “initial pubertal period” needs to be addressed. Although records of the ages of sixteenth century choristers are hard to come by, ongoing interrogation of the available literature has led the present author to the conclusion that the age range of sixteenth century choristers did not differ radically from that of boys today. Careers seem to have begun a little earlier at around six or seven years of ages and lasted perhaps a little longer until fourteen or fifteen years of age. However, the age at which a chorister reached the peak of his powers seems to have been “about eleven or twelve” – much the same, in other words as it is today.(19)
If Cooksey specified 10-11 years of age for the majority of boys reaching the “climax of beauty and fullness” in their voices, difference in pubertal timing between 1992 and 1575 do not seem to be that great. Puberty looks to have been perhaps six to twelve months later in 1575 than in 1992. The present author and a colleague from Leipzig conducted a comprehensive meta-analysis of the relevant literature and concluded that for the highly disparate range of methods used to identify voice change (not the same as actual puberty) a small regression from just over 14 in 1860 to 13 in 2000 could be plotted.(20)
Plotting back earlier than 1860 becomes progressively harder and ever more reliant upon anecdotal or single case evidence. Concerning the period between 1992 and the present, two more recent studies by the author lend some credence to press reports of choir directors’ dismay at losing boys earlier than the traditionally expected age of thirteen and a half.(21) There is, though, no medically credible, statistically valid evidence to support the oft-made assertion that puberty was a great deal later in past centuries, improbable ages such as eighteen being quoted by some authors.(22) What matters is that if, as is more likely, it was no more than eighteen months later than today and “about twelve” was the age the sixteenth century boy voice was at its best, we can be confident that almost all singers of meane parts would have had unchanged voices. For those boys remaining in choirs until fourteen or fifteen, unchanged voices were still possible, but some would have reached pubertal onset – the stage that Cooksey describes as “usually sing alto”.
It seems unlikely that sixteenth century boys would have used any form of falsetto to prolong their voices. For one thing, there were no trebles after 1549. For another, to sing meane (C4-D5) with a voice that is beginning to approach its climax of change is considerably harder than to sing in falsetto at pitches above E4. The difficult passagio that develops between new and old voices in the C4-E4 region would be crossed frequently. It is quite probable that in sixteenth century England there were no “English choirboys” (as defined by Phillips). Those boys that did sing as choristers were aged between about seven and fourteen, the majority sang as meanes with unchanged voices, and were at their best at around the age of twelve.
Effectiveness on the mean part
To be effective, a meane part must not only sound “sweete”, it must also balance the other parts in the polyphony and be heard evenly throughout its range. In particular, at points of imitation, its entry needs to be heard distinctly against the other voice parts. This is a requirement both obvious and deceptively simple. In practice there are some quite significant acoustic difficulties. The ear does not perceive loudness equally throughout the pitch range. The lower the frequency, the more the energy that is required for the same perception of loudness. This was shown originally in 1933 by the Fletcher-Munson curves of equal loudness.(23) Unfortunately, the energy generated by boys’ voices does not increase as pitch decreases – the reverse appears to happen. Figure 3 shows a sine wave tone sweep through the full range of a boy’s voice. Sound pressure level (yellow line) remains constant. The constant SPL of this sweep can be compared with a tone glide through what would be the meane range by an English chorister. The boy shown in figure 4 is aged twelve years and five months and at the time of the recording had shown little sign of voice change. An acoustic spectrogram showed the typical single register profile of the English boy – no displacement of the spectral lines at any point between highest and lowest notes phonated.
The portion of the glide shown starts at C4, which has been standardized to an 80dB reference level on the note A4. A fall of 6dB represents a halving of sound pressure level (SPL) and this occurs around the note F#4. By the time C4 is reached, SPL is down to 64dB. The lowest clear note in the glide is A3, where the SPL has fallen to 56dB. As already detailed, measured SPL is not the same as perceived volume, which is a psycho-acoustic property. 10dB is regarded as the psycho-acoustic point at which volume appears to have halved. This being the case, the note C4 would be perceived as half the volume of the note F#4. If F#4 (the centre of Phillips’ mixed voice) be taken as the middle of the mean range, would the part be equally clear in the polyphony above and below this note, when sung by boys with similar vocal profiles?
A digital version of the same tone glide was then prepared with a solo boy’s voice from the East West Symphonic Choir software sample library. This is shown in Figure 5. It will be seen that the SPL is constant throughout the meane range, more like Figure 4 than Figure 3. There is no falling off in tone or presence from E4 downwards. It is not known whether any artificial enhancement to the lower tones was applied during the recoding process for the sample library, but the voices used were those of American boys, presumably taught through whole voice, three register methods similar to those advocated by Philips.
Pilot Empirical Study Method
Experiment One: Recorded and digital voices
Before this question was investigated with living boys, a preliminary experiment was performed using a point from the Alleluia from the Gloria of Ludford’s Missa Dominica Lady Mass for treble, meane and tenor. It will be seen that the three parts enter successively tenor, meane, treble. Two commercial recordings of the work were available, one by an English boys’ choir and the other, transposed down a fifth, by an adult group employing counter-tenors on the treble part.
It was noted that, for the adult group, all three entries could be heard distinctively, and the parts were well balanced.
In the boys’ version, the meane entry was hard to hear. It was almost as though the order of entry was tenor, treble. However, by the time the meane pitch had ascended to A4, balance was restored. The light tone and lack of projection of “English choirboys” on notes below E4 would be one possible explanation.
A digitally created recording of this extract was made with the East West Symphonic Choir software. In the third sample, the meane entry is perfectly clear and distinct.
It was not possible to disaggregate the commercial recording to individual parts. However, a second digital version was produced with the sound level of the meane part reduced to match that of the English boys’ version on the entry note of C4. The reduction in level was 10dB, i.e. a reduction to half the perceived volume. At this level, the balance was poor for the whole four bars.
Finally, a manipulated version of the digital meane part was created to replicate as far as possible the “English choirboy” profile of Figure 3. First is heard the three-part mix and finally just the trebles and meanes.
The question that then arises is that of whether similar results might be obtained with living boys. Do any classes of living boy replicate the digital profile, and can any patterns linking either training regime or physical maturity be identified?
Experiment Two: Living Boys
Two choirs kindly agreed to participate in an experiment involving English boys, normally trained as “trebles”, singing as means. Clearance for work with living boys on the project was provided by the Edge Hill University Ethics Committee after rigorous scrutiny with reference to the requirements of the Helsinki Declaration. Participants gave fully informed consent to anonymised use of data in academic publications and were offered the right to withdraw. Choir 1 had recently completed a commercial recording of Tudor music at historically informed pitch with the boys singing notionally as means. The track of Byrd’s Verse (Second) service (a new edition at historic pitch specially prepared for the project by Andrew Johnstone) was used. The director of Choir 2, a cathedral choir in the north of England, was agreeable to performing from the same edition during a “quire pitch evensong” at some time in the future.
The director of Choir 1 wrote that:
our boys are no different to other boys in other church choirs. They do not have physically lower voices and they are not on average older than similar choirs. But the post-1549 meane part, as epitomized by Byrd’s 2nd Service (in G minor edition by Andrew Johstone) is perfectly singable. On our disk we recorded this piece with 14 boys, 8 lower voices and the St Teilo Organ.(24) It balanced well. It is my contention that range was sung by normal ‘treble’ boys but who did not sing outside of the meane range; not a counter-tenor, and not a changing boy’s voice. Some would describe this as not singing outside of ‘chest’ register.
Assessment of these boys confirmed the director’s statement that they were not older on average than boys in other choirs. Indeed, access to data on their ages revealed that they were probably younger on average. The average age of all the boys present on the day of the research visit was 10:09. On the basis of speaking voice fundamental frequency (SF0), 50% of these boys were reckoned to have unchanged voices, 12% to be on the cusp of change, 31% to have changing voices and one boy to have passed the critical stage of change. This is shown in Table I.
This is a young choir, dominated by unchanged voices. Were the same boys alive in the sixteenth century, it is probable that all but perhaps the fourteen-year-old would have had unchanged voices. Their conductor describes them as “normal trebles” whilst suggesting that some commentators would consider the singing to be “chest voice” on account of it being in the meane range. One such commentator would very likely have been David Wulstan.(25) I have discussed Wulstan’s views on the meane voice elsewhere.(26) Whatever the merits or otherwise of Wulstan’s analysis, it was clear that the boys of Choir 1 were not singing in “chest voice” – at least, not if “chest voice” is equated with speech register. Most of the boys appeared to be carrying over their training as “trebles”.
The boy who sang the verse part “as he promised to our forefathers, Abraham and his seed for ever” was selected for particular attention. He was aged eleven years and ten months, 142cm (5’ 0”) in height and a capable young musician. His SF0 was 261Hz, clearly an unchanged, pre-pubertal voice. The acoustic profile was similar to the “English choirboy” profile given in Figure 3. It had been noticed in the recording made by the record company one week previously that there was an increasing tendency for the organ to mask the voice as pitch fell below E4. The boy was asked to re-record the verse part under controlled conditions so that the voice could be examined both with and without an organ accompaniment, and with and without the large abbey acoustic. All recordings were made under controlled conditions. Distance between singer and microphone was held constant and all voices were recorded during the same session at identical levels of gain. For the purposes of analysis, each recording was then equalised against a white noise reference of 80dB for the passage “as he promised”, which is sung on the note A4 (mean measured boys’ pitch 470 Hz with the St Teilo organ). This allows meaningful comparison of the SPL profile of each boy as he sings down towards the bottom of the meane range. The results for the soloist from choir 1 are shown in Figure 7. The blue (top) arrow represents the 80dB reference where the organ and voice are well in balance. The orange (middle) arrow represents the point at which SPL has fallen by 6dB, and the red (bottom,) arrow represents the point at which the voice becomes all but inaudible through masking by the organ. This occurs on the final syllable of “ever” and can be heard in the sound sample.
Studio reconstruction shown in Fig 7 (controlled conditions)
This can be compared with Figure 8, which shows the same passage sung by a digitally generated voice in which the sound pressure level does not drop off. The key question to be answered here is whether any living boys are able to produce as similar profile. An older boy, aged 14:04, was asked to perform the same test as the original soloist. The result is shown in Figure 9 and can be heard in the audio sample.
The fourteen-year-old voice was a falsetto one. The boy’s ability to maintain even falsetto tone right down to C4 was notable, though it is unlikely that this reflects the technique of a sixteenth century mean since such boys would almost certainly have had unchanged voices unless they were aged fifteen or more. This seems unlikely given the available documentary evidence on ages.
Boy aged 14:04, studio reconstruction (falsetto voice)
Work with choir 2 allowed further exploration of this issue through replication of the experiment with five boys put forwards by the choir director as potential candidates for the Byrd verse in a “quire pitch evensong”. The boys differed quite widely in age and physical maturity, as shown in Table II. Only numbers 1, 3 and 5 had unchanged voices and thus conformed to the profile of the sixteenth century meane considered likely. No. 2 had reached the first or second stage of change and was able to phonate the alto F3 as a result, though paradoxically produced less power in the C4-E4 region (see below). Number 4 had reached a later stage of change.
Recordings were made under the same controlled conditions as for Choir 1. The five performances of the Byrd verse are illustrated in Figure 10. The gain was adjusted at the post-production stage such that the note Cb5 was 80dB in all cases.(29) The quantity shown in dB for each boy is the SPL reached at “ever” on Cb4. The closer this figure is to 80, the more consistent is the boy’s tone toward the bottom of the range. This is readily seen by the gap between the red and blue arrows. If the criterion for choosing the soloist were to be the ability to produce even tone throughout the meane range with least loss of acoustic energy at lower pitches, numbers 4, 3 and 1 might be candidates. Numbers 2 and 5 might be discounted and the masking effect of the organ can indeed be heard sound samples 11 and 14. The singer with least power difference between top and bottom of the range is number 4 (sample 13). As previously, this is the falsetto voice and might be discounted on the grounds that the sixteenth century means were unlikely to have sung in this way. The remaining choice between singer 1 and singer 3 is largely a subjective one (samples 10 and 12), though it might be noted that singer 1 was the youngest boy (09:08) and singer 3 the oldest (14:04).
This preliminary investigation has been concerned with the possibilities of rediscovering a voice that is lost to choral singing – the boy meane of sixteenth century England. One justification for the effort expended is the potential for enrichment of historically informed performance. Almost all historically informed performances at original pitch are by specialist adult choirs. The sound of the boys’ voices that would originally have performed the music can only at present be the subject of conjecture. Historically informed performance by boy meanes might be considered a somewhat esoteric aspiration within a niche market. However, another perhaps stronger justification for the line of enquiry arises through the assertion of Kenneth Phillips that most English choristers sing in a single register that fails to develop the boy voice to its full potential across its whole range. Phillips’ verdict on the practice is that the English approach is not a healthy pedagogy. It is the failure to produce a balanced sound across the middle register that is his concern. Indeed, he is entirely positive about the English upper voice, unlike the other form of single register singing – upward forcing of the speech tones with the resultant ugly tone and potential physical harm if sustained for too long.
“English choirboys” might, therefore, carry on in their time-honoured way, provided they are not asked to sing as means. The desire for a “ringing, robust quality” that is not “harsh” or “ugly” in the mean range, however, surely changes everything. English singing teachers such as Williams would probably agree that inefficient singing resulting from incorrect choice of vocal mechanism is not desirable.(30) Boys in choirs such as Westminster are already being taught to “mix down” rather than “chest up”. Moreover, there seems no good reason why the singing of all English boys should be constrained by conservative practices that can still be found in other cathedrals. Those concerned with the importance of tradition might consider that the more widespread use of “mixing down” into a middle register represents less a departure from tradition than a rediscovery of more ancient and venerable aspects of it.
What, then, has been learned from this small, research-based start? First, the problem of light, lifeless sound that does not project at lower pitches seems genuine. It may not be a significant issue for as long as boys sing only at soprano pitches but using the same voices as means resulted in points of imitation that could not be clearly heard and voices disappearing behind other parts or instruments. This was not the case for all the boys tested, however. Notable exceptions were the two falsetto voices. This represents first a challenge to any who might regard falsetto production by a boy at pitches in the region of C4 as inefficient. Inefficient it may be, but it appears to produce sufficient volume and be free from danger of a “crack”. Second, it is a challenge to the view that the sixteenth century boys singing mean would nearly all have had unchanged voices. Whatever the challenge, the evidence reviewed here and elsewhere suggests that a young, falsettist mean is a chance phenomenon of the twenty-first century with no counterpart in the sixteenth.
What, then, of boys in early change, the ones Cooksey might have expected to see on alto parts? There were two such boys in the admittedly small sample, but they were the least effective of all on singing mean parts. This surely raises an important question for further research. It is possible that these boys’ relative lack of suitability at lower pitches is explained by the non-use of the new voice that is on the cusp of emerging whilst the old voice was less able to descend to the lowest pitches than the voices of the unchanged boys in the same choir. This might only be the case if blended register or “mixing down” has never been taught so there is surely a question of some significance to investigate here. Of the unchanged voices, the two in Choir 2 were the most successful of all voices in the study on the mean part, which at least confirms the likelihood that the sixteenth century means were all pre-pubertal. What cannot be known here is whether, had these boys received more training in efficient singing at these pitches, they would have been even more successful. There is also the consideration that the unchanged voice from Choir 1 found it harder to be heard at the lowest pitches in his verse part. Is this all a question of random chance? The sample size is too small to say.
Singing by boys in the range C4 – D5 is unlikely to be as satisfactory as in the range E4 – G5 if it is attempted in single, upper voice register. If there is a vocal colour in English choral singing that was known in the sixteenth century but subsequently forgotten, progress in its rediscovery may well be made by a renewed focus on the middle register, following the advice of Phillips that the blend must be accomplished from the top downward; that is vocalises begin in the upper voice and move downward into the middle register (C5 down to C4). (29) The present author has used this approach for some time and has not encountered difficulty in obtaining robust, ringing tone from boys in the C4 region. Boys elsewhere “mix down” and have no difficulty with alto parts. We have no means of knowing whether downward vocalisation was regularly practised in sixteenth century England. Sadly, Richard Mulcaster’s text does not answer this particular question for us.(30) The present study, however, gives grounds to push for more downward vocalisation in English boys’ choirs. It equally gives grounds to push for choirs in those parts of the world where downward vocalisation and the mixed middle register are common to pay more attention to the old English mean voice. Until these things are happening on sufficient scale, we will not be able to produce a statistically valid answer to the questions tentatively asked in this paper.
Notes and References
(1) Phillips, K. (2013) Teaching Kids to Sing. New York: Schirmer.
(2) LeHuray, P. Music and the Reformation in England 1549-1660. London: Herbert Jenkins. p122.
(3) Phillips, P. (2005) Treble or soprano? Performing Tallis, Early Music, 33(3), 497.
(4) Miller, R. (1977) English, French, German and Italian techniques of singing: A study in national tonal preferences and how they relate to functional efficiency. Methchen, NJ: Scarecrow Press. p150
(5) Ashley, M. (2018) Beautiful Swansongs of English Cathedral Music, NATS Journal of Singing, November/December. Also, Ashley, M. (2013) The English Choral Tradition and the secular Trend in pubertal timing, International Journal of Choral Singing, , 4 (2), 4-27.
(6) Phillips, K. op. cit. pp90-92.
(7) Williams, J. (2012) Teaching Singing to Children and Young Adults. Abingdon: Compton. pp99-100.
(8) Castellengo M., Chuberre B. and Henrich N. (2004) Is Voix Mixte, the Vocal Technique Used to Smooth the Transition across the two Main Laryngeal Mechanisms, an Independent Mechanism? Proceedings of the International Symposium on Musical Acoustics, March 31st to April 3rd (ISMA2004), NARA, Japan
(9) Hirano, M., Kurita, S. and Nakashima, T. (1981) The structure of the vocal folds. In Vocal Fold Physiology, University of Tokyo Press, pp. 33-39. Also, Nicole, Y., Herris, H. and Mongeau, L. (2013) Current Understanding and Future Directions for Vocal Fold Mechanobiology, Journal of Cytology and Molecular Biology, 1(1): 1- 15.
(10) Morrison, A., quoted in Ashley, M. (2014) Contemporary Choral Work with Boys. Abingdon: Compton. Pp181-182.
(11) Phillips, K. op. cit. p93.
(14) Royal School of Church Music (2009) Voice for Life Chorister Training Scheme (ed. L. Perona-White) https://www.rscm.com/voiceforlife/
(15) Ashley, M. (2018) Beautiful Swansongs of English Cathedral Music, NATS Journal of Singing, November/December. Also, Williams, J. (2012) Teaching Singing to Children and Young Adults. Abingdon: Compton. P70.
(16) Baldy, C. (2010) The student voice: an introduction to developing the singing voice. Edinburgh: Dunedin Press.
(17) Cooksey, J. (1992) Working with the adolescent voice. St Louis, MO: Concordia, p55.
(19) Lin, S. (1991) How old were the children of Paul’s? Theatre Notebook, 121 – 131.
(20) Ashley, M. and Mecke, A-C (2013) Boyes are apt to change their voices at about fourteene yeares of age: an historical background to the debate about longevity in boy treble singers. Reviews of Research in Human Learning and Music. 1, 1 – 19.
(21) New York Times Novemeber 8th 2013 Where have all the sopranos gone? https://www.nytimes.com/2013/11/10/magazine/where-have-all-the-sopranos-gone.html
Daily Telegraph 9th October, 2010. Choirs in deep trouble over voices breaking early https://www.telegraph.co.uk/news/religion/8052873/Choirs-in-deep-trouble-over-voices-breaking-early.html
(22) Daw, S. (1970) Age of boys’ puberty in Leipzig, 1729-49, as indicated by voice breaking in J.S. Bach’s choir members. Human Biology 42:87-89.
(23) Fletcher, H. and Munson, W. (1933) “Loudness, its definition, measurement and calculation”, Journal of the Acoustical Society of America 5, 82-108.
(24) One of the three historic 5’ transposing organs, recreated by the Early English Organ Project. The St Teilo organ is normally housed in the St Fagan’s National Museum of History in Cardiff, Wales, where it forms part of the experience of worship project in the rebuilt St. Teilos’ church.
(25) Wulstan, D. (1979) Vocal colour in English sixteenth-century polyphony. Journal of the Plainsong & Medieval Music Society, 2, 19-60.
(26) See Fearest Voyces of England
(27) The slightly odd C flat is given to reflect the pitch that sounds when the note C in the Johnstone edition is sung at the pitch of the St Teilo organ.(28) Williams, J. (2011) Teaching Young Voices Safely. Paper presented at Foundations for Excellence, Dartington Hall. p5.
(30) Mulcaster, R. (1581) Positions Wherin Those Primitive Circumstances Be Examined, Which Are Necessarie for the Training up of Children, Either for Skill in Their Booke, or Health in Their Bodie? Edn. William Barker, Toronto: University of Toronto Press