Boy Sopranos and the lost Bel Canto Tradition

Extracts from a paper given by Stephen Beet to the Guild of Musicians and Singers at All Hallows by the Tower, London, in November 2005.

When the first ‘Better Land’  CD was issued six years ago, little did I anticipate the interest and controversy that would result.  Letters were received and articles written, exciting at one extreme the most favourable, and at the other the most hostile comments from some of the highest musical authorities in the land.

For many years I had been conscious of the changes in voice production which had taken place since the war and, although I had never liked them, I had thought them merely changes in interpretation, style and fashion but had never accepted them as progress towards a more perfect result.  It is my opinion that in fact there has not only been a change but a serious decline in standards.  And we have evidence today to prove the point and I should like to suggest that a renaissance in ‘bel canto’ methods be effected before it is too late.

The accepted view is of the generally poor state of music in English cathedrals after the First World War, prompting the Archbishop’s commission and leading to the founding of the English School of Church Music.  Most people to whom I spoke claim that singing and choral music had been generally depressed after the war and until the latter half of the 20th Century, and they were quick to point out the many second-rate recordings made commercially by some of our leading cathedral choirs before and just after the war, and they reminded us of the improvement in the standards of choral singing over the last forty years.  But most false assumptions are based on half-truths and I remain unconvinced by these arguments, as they seem based on little evidence.

The choirmasters who had been responsible for church and cathedral music in the immediate post war years spoke of booming basses, whining altos and, most of all, ‘hooting boys’.  One retired cathedral organist said, “I’ve spent my entire career getting rid of that dreadful sound, and you want to bring it back!”

When one realises there were over 300 church choirs in London alone at the turn of the century one can imagine that a good proportion would be amateur to say the least.  Timothy Day of the National Sound Archives pointed out that it was not until the advent of recorded sound, when performers could actually hear the sound they were making, that things began to improve.  But Noel Malcolm seems to contradict this view in his review of Tim Day’s ‘A Century of Recorded Music’ in the Telegraph when he wrote:  “We may have lived through a golden age in the last century – when the old traditions interacted with the new technology, but were not yet entirely dominated by it.”

What were these Old Traditions, especially in respect of boy soloists?  I believed from listening to quite a large selection of gramophone records that the accepted view could be challenged.  And, indeed, after the release of Volume One of The Better Land, Francis Jackson wrote to me, “Before the issue of these recordings, your views could have been dismissed as nonsense.  Now the recorded evidence speaks for itself.”  Until recently, the only really good pre-war recordings available of a boy were those made by Ernest Lough and the choir of The Temple Church.  Robert Tear remarked that Lough’s voice was pure ‘bel canto’.

I set about researching and collecting recordings to issue on a CD of boy sopranos, as they were invariably called in those days.  In fact the term ‘boy soprano’ far from being a continental label can be traced in English use as far back as 1863 when Master Richard Cocker, at the age of 13, was billed as a concert soprano.  ‘Treble’, ‘chorister’ and ‘singing-boy’ were the usual terms for boy singers in the 17th to 19th centuries.  ‘Choir-boy’ we can only trace in print back to 1843.  The term ‘treble’, defining the part sung, appears not to have been restricted to boys but applied also to women and to counter-tenors: the diary of John Evelyn (1620 – 1705) describes ‘the famous treble, Mr Abe, newly returned from Italy’.  But certainly the term ‘Boy Soprano’ was favoured not only by recording companies but generally in the 20th Century and survives to this day in the popular terminology of the North of England.

Going back to the Middle Ages, apart from the choral foundations and song schools, undoubtedly boys were engaged in popular entertainment in taverns and market places as dancers, jugglers, acrobats and singers.  A directive of Notre Dame in Paris dated 1411 states, “We do not want the boys to go to any places or dwelling or church to sing unless by special license of the superiors.  And then the master should be present to see that they do not indulge in too much food or otherwise behave themselves immodestly.”

And, of course, the earliest named boy singer was the young David, whom we may infer was about 14 years of age when he played the lyre to calm King Saul (1 Samuel. 16:23).  He was old enough to travel alone but not to fight.  Curiously the Old Testament does not actually state that he sang, but the lyre was usually played as accompaniment for singing.

I had known that boys other than Ernest Lough had made records but the sheer quantity turned out from the advent of electrical recording in the late twenties was quite remarkable.  These seemed to have been as a direct result of Ernest Lough’s success.  Listening to these recordings I began to realise that the best vocal techniques employed in those days were, indeed, far superior to those employed after the war and up to the present day.  It was a fundamentally different technique of singing, both for church trained boys and those who trod the music halls of the day.  The more I read of the training methods, the more I came to realise what had been lost – in fact, deliberately destroyed!  The boys sang with such ease, without any hint of strain.

I decided to trace as many of the lost boys as possible and to find out, before it was too late, the secret of their success.  In the meantime, Volume One of The Better Land was issued.   More than one or two choirmasters and critics said ‘Those boys were much older than boys singing today’.  ‘You can’t expect to cart a boy off to choir school and concentrate on nothing but singing all day long, as was the case in the past,’ was another remark, ‘I cannot train a thirteen year old boy to sing like that!’  Well, I know now that, although some of our boys were old than the average boy of today, these were false assumptions.  And these raised the question:  Why were they singing longer?  Was it a matter of their voices breaking later?  Or was it, as I suspected, something far more complex?

As we traced more and more of the lost boys, ranging in age now from sixty to eighty-eight, the same story was repeated time after time.  ‘It was the way we were taught; it preserved the voice beyond the natural break.’  ‘We were taught to bring our voices forward and throw it up into the head; it was simply a matter of correct breathing’.

‘What is this Head Tone?’ I was asked.  ‘What do you mean by it?’  These questions are generally asked by those of a post war generation.  Those brought up before the war know exactly what we mean by head tone.  It isn’t falsetto as some claimed – meaning an unnatural voice.  The debate surrounding this area of voice production was considerable and I discovered a veritable ‘viper’s nest’ of controversy raging in the early twentieth century.

The Victorian and Edwardian choir manuals are definite that the thick ‘chest register’ is not the natural singing voice of a boy, but rather the pure head register, which is capable of the greatest development.  Derek Barsham claimed that his voice was completely natural: he was discovered singing with other children in the Bible Class in Enfield.  All he needed to do was to practise his breathing.

The criticism that these boys were not singing naturally is unfounded.  All the boys interviewed were of one mind that they were simply singing in the best tradition of the boy soprano, or boy treble, taught by men of the highest reputation.  It was Derek Barsham who reminded me that the great Isabel Baillie’s voice was likened on many occasions to that of a boy.  And when Barsham understudied for Baillie in Messiah in 1946, it was precisely because she possessed that purity and natural quality that one expected from the best boy singers.

Victorian choirmasters believed in strengthening the lower register, sometimes by blending in the chest register, often, as in the case of Barsham and Bonner, by carrying down the head register.  James Bates of the London Chorister School in 1899 observed that all his boys were capable of singing the entire alto and soprano range in a pure head voice.

Much has been made of the supposed fact of earlier voice breaking, or ‘changing’ as we are supposed to refer to it these days.  Reliable historical data is scarce but a passage from ‘The Problems of Aristotle’ (published in London, 1597) states that “boyes [are] apt to change their voice about fourteene yeeres of age”.   Evidence from Notre Dame in Paris before 1550 shows that choristers advanced to lay clerks at 14 to 16 years.  Roger Bowers wrote (JRMA 1987) that “It is entirely clear that in the late fifteenth and early sixteenth centuries the trained singing-boy’s voice broke not at 18 [as some claim] but at about 14 of 15.”  He goes on:  “Neither physically nor vocally …. can the early Tudor chorister have differed very much from his present day successor.”

A summary of the 19th century training manuals indicates choristers started singing treble at 7 or 8 years of age and went on as soprano until their voices changed, usually at 14 or 15, sometimes moving to alto at 16 until their late ‘teens and sometimes beyond.  A survey of North Surrey boy trebles (in C. Cleall “Voice production in choral technique” [1969]) showed no “broken” voices at 11; few by 13; half at 14; most before 17 and all by 18.  These figures are in accord with those of the scientific survey by Kinsey in 1949, which took sample of 2279 boys.

3.5% had broken by the age of 11
17.5% by age 12
43.9% by age 13
69.9% by age 14
84.2% by age 15
93.3% by age 16
and 96.6% by age 17

David Lewer (Archivist of The Temple Church) found a remarkable consistency in the leaving ages of Temple Choristers from the time records began in 1842 right up to the retirement of George Talben-Ball in 1981, when several of the boys were sixteen and one seventeen.  Although boy sopranos of this age are not unknown today, the major change in recent decades seems to be the educational system that makes boys leave their choir/choir-school at 13½.  Even if they continue some soprano singing at their secondary school, there is neither rarely the same motivation nor the necessary daily practice.

The question remains, however, if boys were taught the traditional techniques, would they retain their voices longer?  The celebrated cathedral Alto, mentor of James Bowman, Frederick Hodgson, was in no doubt that they would.  In ‘Choirs & Cloisters’, published in 1990, Freddy says; “The full head resonance, a sound capable of filling a large building, does not fade at the onset of puberty.”  He recalls boys at Lichfield, well over six feet tall.  Choirmasters used to send boys to him to be given breathing exercises in order to restore the voice.

‘Those boys are singing falsetto’ more than one choirmaster has said to me.  ‘He sounds like women; (Humphrey Carpenter remarked of Derek Barsham) ‘and must have been trained by women.’  In those early stages of research I had no way of answering, but I suspected they were completely wrong.

It is here that we encounter severe difficulties in terminology especially with regard to ‘falsetto’ and ‘head tone’.  Terminology used in the 18th century is confusing.  St. Jerome of Moravia writes of men having both head voice and chest voice, but it seems the term ‘head voice’ was then interchangeable with ‘falsetto’.  Mengozzi in 1804 explains that a man has two registers ‘one from the head and the other from the chest’.  And there are random comments here and there which confirm the belief that English boys ‘sang falsetto’.  One Oxford Don added ‘how else do you explain the difference in the sound of King’s and Westminster Cathedral.  Both Alfred Deller and Freddy Hodgson sang continuously across the period of voice breaking and both were aware of no change in voice production between singing treble and falsetto.

Evetts & Worthinton (1928) claimed that sixteenth century England led the world in vocal music and singing.  This was the true ‘bel canto’ described simply as beautiful singing, the result of perfect technique, for which the Old Italian School was noted.

In the 16th century learning of singing was by imitation, the tradition being handed on from one singer to another.  The treble voice of a boy can and should be produced throughout its compass of at least two octaves by the thin mechanism.  In this way evenness of tone is ensured and strain prevented.  The use of the thick mechanism gives raucous tone on the lower notes, followed above the point of change by ‘hooty’ so-called ‘head’ notes, and clear diction is difficult.

Several authors make reference to the habit of forcing the thick mechanism up to C with disastrous results.  In 1906 the eminent voice specialist, Sir Morell Mackenzie wrote: ‘The high falsetto of men and the head voice of women are produced by a similar mechanism.’  The falsetto of the man, answering to the head register of women.  The singing teacher, Madame Marchesi of Paris who was taught by Garcia (1837–1925) and in direct line from the Italian masters, claims that these head tone techniques produced the ‘bel canto’ taught by the Italian masters:   ‘It is safe to carry a high register down, but it is always risky, and may be injurious to the throat, to carry a lower up beyond a certain point,’ she said. 

George Stubbs, writing in 1894, attacks the ‘false and pernicious methods based on the development and extension of the thick register.  The growth of the voice – the production of power – is a slow process.  That the head register is capable of great development is not fully grasped by the majority of choirmasters.  That the use of the head tone throughout the entire vocal compass ... embodies the old Italian method, practised for ages by choirmasters of Italy and later introduced into England is a fact either unknown or ignored.’

Most writers of this period make no distinction between the training of boys and women (the training of young girls seems little discussed).  In fact, Stubbs states:  ‘What is the boy’s voice from the voice-trainer’s standpoint?  It is THE WOMAN’S VOICE.  It would be a blessing if the term ‘boy voice’ could be abolished entirely.  It insensibly tends to foster the idea that Nature intended the boy to have a singing voice perfectly unique.  Small wonder that choristers are accused of singing ‘falsetto’.  The larynges of boys and girls shew no differences.  That surprise should be so generally expressed when boys sing like women is most unfortunate.  The boy’s voice is plastic; it is capable of marvellous development; it shews in course of time singular purity, and it exceeds in range the adult female voice.’

But it was John Spencer Curwen who was perhaps the most influential writer in relation to the specific training of boys’ voices.  He held that boys and women had several registers, which could be blended at will to produce an ‘ideal chorister sound’ but held personally that the ‘pure head tone’ should be used.  In the late 1890s he toured England noting the training methods of many cathedral and church choirmasters.

The famous Dr Buck of Norwich, a leading exponent of the bel canto techniques, had been long dead by the time Curwen was writing, but old boys of Buck were more than willing to describe his methods.  He would have no chest voice above F or G and would take a piece of tissue paper, the size of a postage stamp, hang it by a fine thread in front of the mouth, and make the boy sing to it without blowing it away.

‘Dr Buck did not often play for services but liked to stand at the spot where only the boys could see him.  The raising of the left hand would indicate ‘bad tone’; of the right, ‘very bad tone’.  The head register was taught by singing and humming with the mouth shut, and opening it when the right sound had been obtained.  Dr Buck would then insert his fingers in the boy’s mouth.  “Shew me six teeth” he would say.  Boys who were hoarse had cloves and gum Arabic given them, and a delicate boy was supplied with a ‘pocket pistol’ containing port wine, which he swallowed just before his solos.  To illustrate a penitential psalm, he would describe the captivity of the Jews till the boys dropped tears; and to get a boy to realise the words ‘without Thee all is dark’ Dr Buck shut him in a cupboard.  "Is it not dark without me?", he asked; to which the boy replied, "No Sir, there's a crack in the door!" '

Curwen describes the London Training School for Choristers, which was established in 1894 in Porchester Terrace by James Bates.  ‘His boys are all trebles.  He started in the Church room in the evenings with seventy boys.  All the boys can sing alto, having a good compass.  Mr Bates begins with the middle of the voice, aiming to get F and G round and forward.  He does not regard the alto as a separate voice.’

Curwen stresses that when probationers learn from older boys that no voice building exercises will be needed much will be learned from the older boys.  This seems to have been the approach adopted by Walford Davies at The Temple Church.  Walford boasted that he had never given a boy a singing lesson in his life.  George Dixon commented that at Temple younger boys learnt by copying the older boys.  ‘Temple tone was learnt by osmosis,’ commented Ernest Lough.

In the 1940s fashions changed and the new techniques were introduced, largely as a result of George Malcolm’s influence at Westminster Cathedral and Michael Howard’s work at Ely.  Many choirmasters had ‘decided to do something different’, (to quote more than one) but generally until relatively recently they blended the chest voice and head voice so that at least the top register was in the head.  An ‘halfway house’ seems to have been achieved by Stanley Vann at Peterborough, and the tone achieved there was very much in vogue until the 1980s; when Peter Phillips surveyed many cathedrals in 1978 he remarked that only then were choirmasters moving away from that (which he describes as a head voice) towards the so-called ‘Continental tone’, a move then strongly resisted by older choirmasters.  Whether today’s choir trainers are now moving back to a somewhat ‘centre ground’ is a matter of debate, but I fear that today’s cathedral sound would not have been considered acceptable by the masters of the past.

Sadly, today, we rarely hear this true boy’s voice, but this is where Harry Sever comes into the story.  Already an experienced and highly talented singer, Harry came across The Better Land CDs and wished to take up the torch laid down by Derek Barsham in 1947 and Billy Neely in 1951.  When he heard that Derek and Billy were to appear at the launch of my book in The Savoy Chapel, he immediately contacted me asking if he could come along to meet them.  Derek invited Harry to take part in the concert given by Derek, now a baritone, and they sang a moving duet together.


My task, as I see it, is to alert people to the ‘former things which have passed away’.  Harry Sever recently wrote in ‘The Singer’: ‘The next compilation album available is called ‘Once were angels’ ... there was a mere fifteen years between the last of The Better Land Boys and the first ‘Almost Angels’, but during this time the voices had changed …  Gone was the vibrato, gone the maturity and the expressive and emotive range had also largely disappeared.  Perhaps there are a similar number of boys today with high quality soprano voices who will never be heard because peer pressure dictates that singing high is somehow uncool or girlish.'  I would suggest that, almost without exception, the present training of boys builds on the foundation of the ‘chest’ or ‘thick’ voice.  The true voice is neither found nor developed.  Our Victorian forebears would not have recognised this voice as that of a boy soprano.

I fear that modern choirmasters are reluctant to return to the traditional training methods for a number of reasons:

1.                 Several admit that they would not know how to train boys in the traditional techniques even if they could convince them it would be a ‘cool’ thing to do.  Others are reluctant to change anything because they don’t really understand anything about the voice.  Those who do understand belong to the earlier generation that consciously made that break with tradition.

2.                 Some fear that the sound produced would not be acceptable to modern ears.  In the same way, several choirmasters now have a ‘cut off’ age of fourteen, as they believe it is not socially acceptable for boys to sing treble after that age.

3.                 Finally, although many do recognise that the old ways were kinder to a boy’s voice, they deny that the head tone is suitable for a certain kind of music and they also deny the evidence that the new techniques (i.e. the forcing up of the chest voice) are actually injurious to the voice and responsible for the premature loss of boys’ (and adults’) voices.

This is all, I fear, a sad reflection of our society.  Much can be learnt by today’s choir trainers from a study of past techniques.  This would lead to better training, correct placing, less damage and to the considerable prolonging of a boy’s treble voice.  This is in no way to belittle the excellent work done by choirmasters in these difficult days, many of whom have expressed a desire to hear the fruits of my research.

Stephen Beet
November 2005

Copyright © 2005 Stephen R. Beet


Copyright © 2006
This page was last modified on 10 March 2007