Singing style in Handel opera and oratorio

The following seeks to examine the issues surrounding singing style and voice production in the performance of Handel opera and oratorio.  Whilst this essay predominantly deals with practical questions, some understanding of the philosophical background to the historical performance movement is a necessary prerequisite for any further discussion.  The following is an extremely brief overview of an incredibly dense seam of musicological debate.  Further reading however can be located in the bibliography.

The idea of an ‘authentic’ performance is highly problematic because of the binary associations of ‘right’ and ‘wrong’ that it carries with it.  The term implies that if one practice is ‘authentic’ then another, by definition, is ‘inauthentic’.  Richard Taruskin[1] raises these issues, further suggesting that even the now commonly accepted term ‘Historically Informed’ carries some unwelcome associations: it contains the tacit implication that some performances are therefore ‘historically uninformed’, and that their decision to perform the music a certain way arises from being ignorant of historical information, rather than being a conscious interpretative decision.

This has its roots in the notion of an ‘original’ performance: a performance ‘as the composer intended’.  This concept rests in the belief that the musical work and its performance is somehow historically and socially transcendent, constituting a notional ‘thing’ at the end of a musicological rainbow.

In Editing Early Music, John Caldwell describes the editor’s laudable aim “to reproduce the intentions of the composer” as being somewhat chimerical.[2]  If this is so for the rendering of a musical text, for which there is considerably more discrete, attestable evidence available, how much more is this the case for practical performance?

The problem of a texturally absolutist approach is that the information that remains today is by definition only partial.  As Kenneth Cooper observed: if we restrict ourselves to representing only what we ‘know’ then we can never hope to give anything more than a partial rendering of a piece, which creates a distorted rather than authentic image.  We should therefore “try and fill in a little of the creative energy – even if it’s not exactly the same creative energy (because we’ll never know what that is)”[3] as this helps us create a fuller picture.  Richard Taruskin takes this further, likening strict dependence on historical fact to a lottery “for the performer can exercise no control over the state of the evidence.”[4]  Otherwise this leads to what Lydia Goehr[5] refers to as a ‘museum culture’ in music performance rather than an ongoing creative process.

It is here that Historical musicological research, and practical historical performance must draw some division.  Research, though occasionally speculative, is fundamentally bound up with what it can prove.  It also has the luxury of not having to comment on what it can’t explain, resorting to the ever popular phrase ‘future research may shed light on…’  The performer does not have this luxury; he has to make practical ‘field’ decisions about how to perform the music in front of him.

One of the most hotly debated areas in the performance of early music is the issue of vocal vibrato.  This is an area in which historical documentation is particularly sparse and, unlike with instruments, there are no physical objects to shed light on singing style.  Questionable attempts to extrapolate backwards and invent a singing style culminated in an almost zealous abjuration of all vibrato from historical performance.

However, Haskell and Butt, amongst others have argued that this preoccupation with tone colour superimposes nineteenth- and twentieth-century values onto the music of an earlier period.[6] Neumann writes: ‘What matters to Bach is line, not colour.  Had he been concerned about colour, he would have prescribed the registration for his organ works, yet he never did; and he would not have transcribed so casually from any medium to just about any other.” [7] I dare say Handel expressed a greater concern for tone colour than Bach, but the general point remains. 

Neumann argues that the phenomena of ‘vibrato’ and the related ‘sonance’, which adds depth and tone colour, have “been known throughout the ages, and vibrato has quite certainly been used on instruments in imitation of voice”[8].  Two examples are the use of vibrato in strings and the tremmulent stop on the organ.[9] The latter is especially valuable since we can often date when the stop was installed.

“Vocal vibrato must be ageless because it develops spontaneously in most mature and in all artistically trained voices.  For such a voice to sing non-vibrato involves a special effort and means fighting a nature.”[10] Many aging singers find it impossible to remove vibrato from their voices.

The opposition to vocal vibrato seems to stem from two arguments:

1)                  the relatively pure sound of boys voices

2)                  The apparent silence of old treatises on the technical production of vocal vibrato.

The second of these points is far from conclusive; in fact it may be entirely circumstantial.  Neumann reasons that the seeming silence in treatises is because vibrato was a natural element of the voice, and not, like string vibrato, something that needed to be learned.[11]  

The first point is by no means decisive either.  In 1619, Praetorius devoted an entire chapter of his Syntagma musicum to the instruction of boys in the Italian manner of singing. In it he wrote: “These are the requisites: that a singer possesses a beautiful, lovely, trembling and wavering voice”[12]  These comments would apply even more to adult singers.

Returning to the eighteenth century, we know that Handel’s singers used vibrato because he has been documented on occasion complaining that one or another has used it too much.  This not only proves that it was used, but, more importantly, that there was a level of artistry and discretion that was considered tasteful.  Hence, the question becomes what is ‘tasteful’ when applied to Handel?

Doubts have often been expressed over the ‘Anglican purity’ of much early music singing.  Partly this is an objection to the absence of vibrato expressed above.  But, moreover, it feels particularly at odds with the style of the Italian opera singers Handel employed throughout his life. 

When asked about this, Nicholas McGegan, the British conductor and Handel expert, felt the greatest danger when experimenting with period singing style is making singers too uniform.  The evidence suggests that they certainly were not in the eighteenth century.[13]

Ultimately the most successful singers were a ‘total package.’  In addition to good voices and technique they had their own distinctive personalities and styles; in short they “sounded like themselves”[14] “It seems that they had a lot of colour in their voices…and weren’t too Anglican.’[15]

Some basic deductions can be made with respect to singing style.  There is evidence which suggests sopranos and tenors frequently used the head voice in their upper registers, rather than belting notes out as singers of later opera tend to do.  So we can assume there was a tendency towards a more lyric style of singer.

Furthermore, it should be noted that vibrato is an unavoidable necessity in romantic opera as it aids projection; it is hard for a voice to sing loud without vibrato.  But, in Handel, where the orchestra is smaller, the singer has much greater freedom to use vibrato as an expressive device, varying tone colour and creating light and shade in phrasing.  Again, this supports the assumption that a more lyric singing style would be preferable.

When we examine those who have typically sung this repertoire since the mid ‘90s it is precisely this style of rich lyric voice that is most prevalent.  There are still numerous lighter voices, particular amongst sopranos (Dawn Upshaw and Annick Massis spring to mind), but the ‘early music singers’ so frequently complained of during the ‘70’s and ‘80s have become a dwindling minority. Singers such as Rosemary Joshua, Loraine Hunt, Richard Croft and Kenneth Tarver now form the mainstay of performances and recordings.

One of the main reasons for this change has been renewed interest in Handel opera, brought about by the growing trend of staging his oratorios.  These have been described as ‘the best operas he never wrote.’  Many were written during lent when the staging of opera was forbidden.  As a result, works such as Theodora and Belshazzar contain an extraordinary level of drama; they are essentially unstaged operas.  Events such as the Göttingen Festival have been influential in promoting these staged versions.  In 2000, the Théâtre des Champs-Elysées began a complete cycle, over several years, of Handel’s oratorios and operas. 

The move towards staging oratorios has attracted a number of more adventurous directors, amongst them David McVicar, Christof Nell and especially the maverick: Peter Sellars.  His 1996 production of Theodora for Glyndebourne broke the mould, juxtaposing a modern setting with baroque gesture.  In response to this new attention and frequently cutting edge staging, the sound of such early music has ‘beefed up.’ 

Meanwhile, during the 1980s, historical orchestral playing advanced into later and later repertories. Roger Norrington’s performances of Berlioz on authentic instruments harnessed a particularly visceral sound.  This fuller style has fed back into the performance of eighteenth-century music, especially opera under the batons of figures like Norrington, René Jacobs, William Christie, Nicholas McGegan and John Elliott-Gardener.  As the orchestral playing has become richer, so has the singing. 

As Handel opera has become more high-profile there has been greater crossover with singers of later repertories.  Cecilia Bartoli is a notable example, as is Jessye Norman who has sung everything from Handel and Purcell through to Bizet, Verdi and Strauss.

McGegan observes that in the 1980s, if an opera company performed Handel, it would often be their ‘cheap show’.[16]  For practical reasons they employed smaller orchestras and casts, often justified by means of spurious scholarship (indeed, this is one argument for the spread of the so called ‘early music voice’ in the ‘70s and ‘80s).  Now Handel operas are more high profile there is a tendency to cast more grandly, meaning both that more established singers are willing to take the parts and that opera companies are willing to pay their fees.

A particularly interesting development has been the proliferation of counter-tenors.  Thanks to pioneers like Alfred Deller, the countertenor voice is now so ubiquitous as to be considered ‘the very emblem of Early Music.  No Baroque opera revival can get by without it.’[17]  However, there are practically no roles written for the voice during this period.  One exception is Handel’s Faramondo, where the countertenor William Savage sang a combination of ‘Childerico’ and the tenor part[18].  It’s is only recently with roles such as ‘Oberon’ in Britten’s A Midsummer Night’s Dream that this voice has broken the confines of cathedral choirs and ventured onto the opera stage. 

Yet, increasingly, the counter-tenor voice is surfacing in repertories that would have been unthinkable twenty years ago.  Paul Esswood has recorded Dichterliebe and David Daniels, Poulenc.  The emblematic status of the counter-tenor as a mark of “good musical housekeeping”[19] is symptomatic of what Taruskin describes as “the modern sound of early music.”[20] 

However, what is truly revealing about their employment in Handel opera is that the style of counter-tenor we see most often is once again this rich, lyric tone quality.  Singers such as David Daniels, Michael Chance and Andrea Scholl all have a warm, fluid vibrato, comparable to that of the singers already discussed.  Even smaller counter-tenor voices, such as Bejun Mehta are a far cry from the sound of Alfred Deller.  McGegan, when asked about the historical efficacy of using countertenors responded: ‘In Handel opera, I’m so grateful to find singers…who can hold the stage as well as sing the notes that I don’t hesitate to use them if they’re countertenors.’[21]  Thus, it seems that the most successful singers are those with interesting and distinctive lyric voices who are all-round performers. 

Closely linked to the issue of tone colour and vibrato is the use of ornamentation.  This has become something of a hot potato amongst musicologists and performers alike.  Particular vitriol has been channelled into discussion of the da capo aria.  Winton Dean is a particular critic of the over-use of embellishments in early music.  He notes that Handel singing has “moved from a period when all da capos were literal repeats…to a fashion where decoration is allowed to sprout anywhere, even in A sections and da capos release salvos of rockets in the style of Rossini or Bellini.”[22]

However, an absence of reliable historical documentation makes it very difficult to ascertain what a suitable balance between these extremes might be. Most singing treatises we have at our disposal date from a much later period.  A commonly cited text is by Domenico Corri.  However, this dates from approximately fifty years after Handel’s death.  Its insights into vocal technique may be valuable, yet, written, as it was, in the age of Haydn, the ornamentation it contains cannot be relied upon to denote the style of Handel’s day.

There are a few examples of written out ornaments by Handel himself.  At first glance these seem to be quite valuable, however, the complaint against them is that the evidence they provide is too specific.  They were written specifically for individual singers, often because the singer concerned was too incompetent to improvise their own cadenzas or da capo decorations. Hence these examples provide us with evidence of an individual singer’s style, rather than an archetype for improvisation in general.

There is a further problem with using these in that one of ornamentation’s main purposes was to express the singer’s abilities and, moreover, personality.  This problem is one that pervades the study of ornamentation in general, namely that if its practice becomes too dogmatic, the results will be over-codified and generic rather than expressions of the singer’s individuality. 

The most important concern with regards to ornamentation is to make the additions appear spontaneous, rather than written into the text. Frequently, written out ornamentations sound too rhythmically deliberate, lacking a feeling of unprompted rubato.  “A very good performer can make you think they’re improvising, even when they’re not, just as a really good actor can make you think he’s scratching his ear because it itches, even though the scratch was rehearsed to the last detail.”[23] 

McGegan encourages his singers to experiment freely, partly because this is a good way of finding out what does and does not work, but more importantly because the more comfortable singers become with the practice of experimentation, the easier it becomes for them to create the illusion of spontaneity.[24] 

Ultimately, the choice of ornamentation is a matter of taste.  Neither of Dean’s extremes fulfils the requirements of the period.  A literal repeat fails to show off the singer’s ability and personality, yet at the other end of the spectrum the original piece of music is entirely obliterated.  The aim of ornamentation is, as it says, to ornament the music, rather than to re-write it.  Whatever is decided, the most important aspect remains the impression of improvisation, without which even the most meticulously researched decorations cannot hope to create the desired effect.

A debate which has long concerned the production of straight theatre, but is increasingly an issue in period opera is that of pronunciation.  It is undeniable that Handel was writing for a specific English pronunciation that no longer exists.  Much has been written on the changes that have taken place in English pronunciation since the Renaissance and there are persuasive arguments for the reconstruction of a form of English that would have been spoken in Handel’s time. 

However, there has not been a standardised English pronunciation until the twentieth-century ‘If you had a regional accent you kept it’,[25] and still differences persist between English, American, Australian etc.: when Philharmonia Baroque performed Judas Macabbeas, one American Critic complained that a singer did not make ‘hands’ and ‘commands’ a full rhyme:[26] not only would they not have rhymed in eighteenth-century England, but they do not rhyme in twenty-first century English either.

Moreover, many of Handel’s singers were foreigners and often sang in their own languages.  At least six of the arias in L’Allegro were first sung in Italian.  When they sang in English the results were often unrecognisable.  In one performance of Esther the Italian singer “made ‘I come, my queen, to chaste delights’ sound like ‘I comb my queen to chase the lice.’”[27]

Therefore, whilst it may be possible to recreate a form of pronunciation that is authentic, the idea that there is a single manner of pronunciation is demonstrably false.

Changes in syllable stress have the potential to disrupt musical phrases.  However, Handel’s melodic lines generally denote the accented syllables and phrasing for the singer.  Whilst vowel sounds may not be consistent with eighteenth-century practice, Handel’s use of singers from a variety of regional and national backgrounds suggests details of pronunciation were less important to him than style, tone colour and the characters of his performers. 

Furthermore, if the pronunciation ‘feels’ unnatural to the singer, the overall ‘effect’ of communication is likely to be impeded.  It should also be remembered that Handel wrote in English because it was the vernacular; the language spoken and understood by his listeners.  Laying too much stress on period pronunciation runs the risk of English coming across as a foreign language, even to English listeners.  

In 1970 Saddlers Wells Opera staged The Valkyrie in Andrew Porter’s English translation.  This translation was so good that it was subsequently re-translated back into German.  Arthur Jacobs, writing in Opera Magazine, remarked that it was the first time that ordinary Germans had understood what their Wagner was about!  The point is; it is surely a more authentic experience for listeners to hear a dialect they understand, rather than a ‘historically accurate’ one which they do not.

Ultimately, as always with historical performance, there is insufficient information to pass any hard-and-fast rules regarding the ‘correct’ way of doing something.  Even if sufficient information could be found, what would it say about us as interpreters to slavishly depend upon such evidence? The final arbiter of performance style is always going to be taste.

Bibliography

Butt, John: Playing With History (Cambridge: 2002)

Caldwell, John: Editing Early Music (Oxford: Clarendon Press, 1985)

Dean, Winton ‘Scholarship under Handel Revival, 1935-85’, in Tercentenary Collection, ed. Stanley Sadie and Anthony Hicks (London: Macmillan 1987)

Dean, Winton: From Handel and the Opera Seria (Oxford University Press, 1970)

Goehr, Lydia: The imaginary museum of musical works: an essay in the philosophy of music (Oxford : Clarendon Press, 1992)

Haskell, Harry: The Early Music Revival: A History (New York: Dover Publications Inc. 1996)

Kenyon, Nicholas (Ed.), Authenticity and Early Music: A Symposium (Oxford: University Press, 1988)

Neuman, Frederick: ‘Authenticity and the Vocal Vibrato’ in Essays in Performance Practice (UMI Research Press, 1982)

Sherman, Bernard D.: Inside early Music: Conversations with Performers (Oxford University Press, 1997)

Taruskin, Richard: Text and Act (Oxford University Press, 1995)