This chapter is intended as an introduction to the choral medium, fulfilling a similar role to that of a guide to orchestration. It considers the forces involved, techniques, traditions, notation, and the avoidance of common errors. Works from all periods will be discussed and the techniques used therein will be considered as all equally valid to the contemporary composer.1 Let us begin with a short overview of some of the styles and textures that appear in the repertoire and are available to the contemporary composer. The choral medium is one deeply linked to history and tradition, both real and mythologised. It is also one continually open to imagination, and whilst the ‘instruments’ are the same as those used in the earliest choral pieces, languages and styles evolve continuously.
Historical Foundations
Monody
The simplest form of lyrical expression is the single line. This may be a solo voice or a unison line involving many singers. The soloist historically had a great freedom of improvisation, and many of history’s greatest melodies were passed down through the oral tradition as part of a great body of folk songs. Many of these were collected and transcribed by figures such as Cecil Sharp, Béla Bartók, and Ralph Vaughan Williams. Vaughan Williams wrote extensively about these melodies and the need to save this resource which had been passed down through the generations before it became lost in the rapidly modernising world of the twentieth century. He found homes for a number of these in the English Hymnal (1906), and these have become the established tunes for a number of hymns and carols. Plainsong also comes into this category, and the familiar forms known today still show some regional variations and provoke considerable discussion on interpretation. What is notable in both folksong and chant is the considerable rhythmic flexibility which is difficult to replicate accurately and clearly in modern notation; this becomes a greater consideration when writing monody for an ensemble, as individual decisions on rhythmic and melodic variation will prevent proper ensemble.
Homophony
Homophony in its simplest form consists of the harmonization of each note at a fixed interval – this, usually at the fifth, has its roots in organum. In a sense a ‘unison’ line with the voices an octave apart is a form of organum at the octave. This form of fixed harmony is still an immensely effective device to evoke the image of the Dark Ages. It has also been used extensively in modern repertoire, sometimes as a form of resonance with this older period and sometimes as a device in its own right. Arvo Pärt’s ‘tintinnabuli’ is in some senses a more sophisticated modern manifestation of this method, involving the use of fixed categories of intervals (e.g. the next note of the triad above the melody) to achieve ethereal simplicity. The result of this approach is varying harmony, and this is a far more common use of homophony; the most common example of this is the hymn tune or choral harmonization where essentially the melody is given one chord for each note. This style can provide clear structural markers and great clarity of textural expression, and the addition of harmony can provide a far greater emotional and expressive impact than that of monody alone.
Polyphony
The ‘grandest’ of styles, which reached its great flowering in the ‘Golden Age’ of the sixteenth century across Europe, is polyphony and is seen in the works of composers such as Giovanni Pierluigi da Palestrina, Tomás Luis de Victoria, Thomas Tallis, and William Byrd. All lines here are independent and of equal importance and can give the impression of an unceasing flow of sound. Elements of this arise from the use of passing notes as mentioned above; however, more sophisticated examples extend to precise canons on the largest scales. Imitation may be strict between the parts but can also be highly flexible, and this freer use is seen in the much later works of composers such as Kenneth Leighton and Herbert Howells, the latter’s style having been described as ‘impressionistic polyphony’ by Frank Howes.2 A striking feature of these polyphonic textures is their provision of breathing space for the voices; not all the parts will be in use at all times. The strategic use of long-term layering and adjusting the density of the parts, both in terms of their individual complexity and how they relate to each other, can help to achieve climaxes of immense power. If there is one general trend in much modern choral music it is verticalization, and many singers, conductors, and publishers have commented positively on the arrival of new music which shows some polyphonic writing. Without advocating a return to fugal writing, it is worth bearing in mind that ‘to deprive choirs of polyphony is to deprive of them of their lifeblood’,3 and works which show some form of counterpoint and individual vocal lines are likely to be particularly welcomed by performers.
Modern Trends
Modern trends in choral music (as in all contemporary music) are hugely varied, and the ones mentioned here are by no means exhaustive but serve to give some indication of the range of approaches. A prominent and continuing trend has been that of the ‘ecstatic style’,4 which Gary Cole suggests has its origins with Randall Thompson’s Alleluia (1940), although elements of the style are suggested by earlier works such as Olivier Messiaen’s O sacrum convivium (1937). These works use traditional technical methods to achieve a sense of stasis and luminosity which is particularly successful with voices, including use of ‘soft dissonance’ and added notes which lends a glow to what are often triadic harmonies. Thompson uses considerable imitation and counterpoint throughout, and the use of the second inversion chord as ‘home’ to develop a structure which seems to unfold constantly whilst never developing in a traditionally linear fashion. Messiaen achieves the sense of magical stasis through his distinctive modal harmonic language, added note harmony, and extremely slow tempi. This is combined with a sophisticatedly constructed melody line to reach a carefully built climax in what is a largely homophonic texture.
The choral medium seems particularly suited to the evocation of mystery and the eternal, at least in part due to its historical and religious associations, and its regular performance within cathedrals and other large buildings imbued with ‘the immemorial sound of voices’5. Overlapping voices invoke this well; for example, including two or more choirs in canon as in John Tavener’s Hymn to the Mother of God (1985) or the creation of a sense of endless movement by voices moving between different notes of the same chord in Benjamin Britten’s War Requiem (1961–2). The use of repeated fragments or chanted words independent to the other singers is a highly effective texture within even a tonal context. The rapid chanting of a single note very quietly can create a strong sense of unease – for example James Macmillan’s Tenebrae Responsories (2006) – whilst the overlapping of phrases created by the voices moving off at different times can create flowerings of great ecstasy, as in Gabriel Jackson’s Requiem (2009). Jonathan Harvey’s Come Holy Ghost (1984) is a masterful combination of Gregorian chant, fragments, aleatoric writing, and complex harmonic textures to create a soundscape which is completely at the service of the text with its evocation of the tongues of Pentecost.
The term ‘Choral Orchestration’ has been applied to large-scale works such as Sergei Rachmaninoff’s All-night Vigil and is an apposite term for the evocation of chant, bells, echoes, and organ-like blocks of sound, combined with the frequent octave doubling of melodic lines. Something of this work and others continues to inspire many works from the Baltic, with composers such as Rihards Dubbra, Uģis Prauliņš, and Vytautas Miškinis. This might involve theatricality and spatialisation, maybe including two or more choirs singing together or antiphonally, or smaller units and separate groups. This spatial separation can be immensely evocative in larger buildings, with the semi-chorus placed in a gallery or even out of sight. In these cases, the composer must consider questions of ensemble and balance, and consider how well the two groups will be able to sing tightly together. The final movement of Macmillan’s Tenebrae Responsories avoids this issue by requiring the soprano soloist to gradually walk off stage during their final monody in a gesture that fits well within the theme of the piece.
A common feature of modern writing across all styles has been the occasional use of extended techniques. These might include use of wordless choruses, whether hummed or vocalised to prescribed sound, speech (rhythmic, free, or semi-pitched as sprechgesang), the occasional shout – as in William Walton’s Belshazzar’s Feast (1931) – non-pitched vocal sounds (e.g. clicks, whistling, whispering, shouting, and breathing) or body percussion (e.g. stamping, slapping, and clapping). The gamut of extended techniques can be as wide as the imagination and the human body’s capabilities. Perhaps more than any other type of music-making, choral singing throughout the world is largely an amateur pursuit: the composer who wants a successful performance, and ideally a second performance, is wise to try and ensure accessibility. It should be remembered that most such techniques may be unfamiliar to less experienced groups and require considerable clarification and co-ordination: an older group for example may struggle to stamp rhythmically and quickly in time, and clapping will require a choir to be singing without music. Nevertheless, choirs generally do enjoy and appreciate challenges.
The ‘Holy Minimalist’ school of composers such as Arvo Pärt, John Tavener, and Henryk Górecki – who each have highly distinctive styles – has provided routes for modern composers who have chosen not to embrace rigorous pitch ordering (e.g. serialism) and atonality exclusively; here, the purity and sometimes extreme simplicity of harmonic language, combined perhaps with orthodox chant and trance-like repetition, is deeply suited to the expression of religious ideas. Alongside this, composers such as John Rutter, Bob Chilcott, and Malcolm Archer have forged a path writing highly crafted music in a style that is accessible and rewarding for choirs, often being performable by groups of different abilities and from different musical backgrounds.
Voice Types
The choir should be considered very much as a unit, and one of the key aims of any group is to achieve a convincing blend between individuals and voice types. Whereas with solo singing chief considerations will include the distinctive characterization and colouring of lines and individual moments and notes, in a choral context this is a group effort and individual voices should not emerge prominently. If there is more than one singer to a part (and this is usual in all but the smallest groups) these voices may occasionally divide for chords or longer passages. Choral singers will often consider themselves to be ‘1’ or ‘2’ within the standard voice types – soprano, alto, tenor, and bass – in the case of any one part dividing, the seconds will take the lower note and the firsts the upper in much the same way that a desk of string or woodwind players will treat divisi passages. The use of occasional divisi is a standard characteristic of much choral music, and immensely thrilling effects can be achieved when a divisi moment allows a chord to blossom into something fuller and richer. Conversely, the deployment of unison (or far more commonly unison at the octave) is hugely effective and sonorous when applied judiciously within the ranges of the voices. The spreading out of the voices subsequently is a highly idiomatic use of the medium.
As a general rule, voices are at their warmest and most lyrical in the mid-part of the range, becoming progressively more penetrating in the upper reaches but projecting less strongly lower down. This is an important consideration when it comes to scoring chords, particularly when these are widely spaced. Chords which would work well on orchestral instruments can be difficult to balance if the parts are widely separated unless there is a considerable ‘filling-in’ of the harmonies in the middle. More perhaps than with any other medium it should be considered that the extremes of range really are that, requiring stamina and technique at the upper end. It is also entirely possible that the extremes of range will not be achievable by all groups and composers should consider providing alternatives for use if needed; even within a section not all singers may have the extreme notes and so it is worth providing options for those voices which do not extend to the limits.
There can be considerable physical and technical effort involved in achieving the notes at the upper part of the range, and this adds to the visceral excitement of powerful climactic points; however, it is important to remember this to ensure that stamina is not exhausted and that these really are ‘stand-out’ moments rather than the regular tessitura. It is also possible for singers to ‘float’ notes in the higher part of the range, and in a choral context the soprano voice is particularly adept at this on suitable vowel sounds. It should be noted, however, that sustained very quiet singing is very taxing (as is loud singing). The same is also true to some extent for extended writing in the lower part of the range, particularly for higher voices.
Text
The question of what makes a suitable text is one which is highly personal, and virtually any text can be set to music: Harrison Birtwistle’s Moth Requiem (2012) for example includes the Latin names of various moth species as a key component in the text. Usually, the composer will feel some attachment to a text (or at least part of it), which leads them to compose the music. Gerald Finzi rather beautifully described this connection with texts as ‘[shaking] hands with a friend across the centuries’,6 and was scathing of the view that some texts were complete in themselves, saying that the composer ‘is driven to composition by the impact of the words’.7
One of the key challenges is reconciling musical form with textural form; the narrative arc of a poem may result in a final destination far removed from the opening, whereas much musical form is reliant to some extent on return and repetition. Some composers take the ‘Wagnerian solution’ of the leitmotif and its infinite processes of permutation, whilst for others, musical material completely follows the text with the mood alone dictating the form. Herbert Howells’ Stabat Mater (1965) is an example ‘that achieve[s] homogeneity through concentration of a single emotional mood’8 to create an immensely powerful and dramatic work governed by its prevailing mood. Gabriel Jackson’s Truro Service, on the other hand, is a deliberate attempt to write ‘an “objective” setting [of the mass], the chord movement in the homophonic verses being generated by the number of syllables in each word and their speech rhythm stresses’, although the composer allows himself ‘a whiff of word-painting at [the words] “to be a light”’.9
Key to the success of these and any other settings are considerations of moments which might prove suitable for musical colouring, word-painting, textual repetition, and those moments which are important structurally and rhetorically. A text as read or spoken aloud is likely to be considerably shorter than a satisfactory musical outpouring, and so a composer will often need to consider points for extension and tension building. This may include the return of material, direct repeats, mantra-like repetitions, or a single word or phrase extended polyphonically. Another decision to make is around how much of a text to use, and whether the ‘purity’ of the original is observed. For some composers, fidelity is a matter of principle, whereas for others the text is purely at the service of the composer and can be changed and cut to fit their aesthetic vision.
A text read in silence is inevitably shaped by the reader alone, whereas the composer is placing a mood, structure, and view on the words. Textural clarity is therefore another element which must be considered, although again the view on this will be a very personal one. Texts which are very well-known or in another language10 (e.g. in Latin) may be suited to settings in which individual verbal clarity is subsumed by structure, general mood, and effect. A little-known text or one with many moments which catch the ear is likely to require a great clarity of presentation, as the listener may become frustrated in trying to follow small fragments which then become buried. An approach like this will usually require musical material inspired directly by the meaning and shape of the words. Other approaches may require a consideration of what the music will be required ‘to do’ in the work, and in this instance may almost be created in isolation to the words. In a work like György Ligeti’s Lux Aeterna (1966) a singular sonic effect is created by the use of the idea of light (‘lux’) and its illustration through separated vowel sounds. The ultimate consideration for the composer should always be a sure and clear understanding of what they intend to ‘say’ about the text, and what they consider it to mean.
One mundane but important practical consideration is the copyright status of any proposed text. Laws vary, but a composer should be aware that it can be expensive and difficult to set any text which is in copyright, and it is always best to investigate and clarify this status before beginning a setting.
Word-Setting, Melisma, and Vowels
When setting words more ‘naturistically’, try to maintain the text’s natural speech-rhythm as much as possible. It should be noted that mis-stressed words are likely to stand out, or even become part of the composer’s aesthetic – for example like Igor Stravinsky’s angular and playful text-setting in his Symphony of Psalms (1930) – so it is important to make such things deliberate rather than accidental. Melodic shapes can be dictated by the images of the words (word-painting), the mood of the line and its rhetorical content, and by the desire to create a satisfactory musical line. Composers should also consider the particular sounds of the words and their vowels and consonants; the placing of these will have a strong impact on how the music is heard, and also how easy and effective it is to perform. Vowel sounds will affect the entire colour of a note, and these sounds modify in many languages with context; in other languages they are always sung to the same sound. Open sounds will tend towards brightness (ah, e (as in ‘air’)), where the more closed (oo, u) will tend towards darkness. The placing of much emphasis on longer and darker vowel sounds will create a more sombre mood, with the opposite for brighter vowels.
Unique to voices is the melisma: a placing of more than one note to a single syllable. These can be greatly extended, and some of their great outpourings are in the ecstatic works of the Renaissance, and even the earliest settings of four-part music such as Pérotin’s Sederunt principes (c.1199). For other composers, the syllabic setting has been a hallmark of style with its strong emphasis on textural clarity. Melismatic placing should be considered carefully along with word stress and vowel sounds to enable singers to negotiate lines in the most musical manner. The placing of these will usually feel natural in a well-written setting but will nevertheless require considerable thought. The word ‘alleluia’ for example has four syllables, all of which are equally suited to melismatic treatment, but which will each create a different sound and thus timbre. Consonant placing also needs consideration; particularly the hard, spoken consonants of ‘s’ and ‘t’ as sibilants can be very prominent in a musical texture. This is likely to occur when several parts have an ‘s’ in close succession, or where breathing is essential.
Breathing
All voices will require the opportunity to breathe, both on a small and a large scale: voices should be regarded as more akin to brass or woodwind instruments than strings. Large groups can ‘stagger’ to create seamless lines, and this can be indicated or left to discretion. This is more usually for special effect rather than for entire pieces. Composers should note that if they do not indicate breaths in the score (with rests or the placing of clear commas above the stave, as in instrumental music) the director is likely to make decisions themselves based on musical and verbal considerations. Dovetailing between the parts can also be considered as an alternative, with lines passed between different sections or divisions of sections, to achieve seamless results much as in orchestral writing.
Dynamics and Balance
Balance should never be simply left to dynamic instruction, although the clear use of markings is important. Balancing widely spaced chords is difficult and unsuccessful unless there is considerable reinforcement between the extremes; a bottom C is almost impossible to project with anything like the penetrating power of a top C. In a well-scored example of a full, rich chord, the lowest notes will provide an audible bass which will add to and be reinforced by the higher harmonic levels above. A solo line in an inner register (i.e. not on the top of the texture) will benefit from ‘textural space’ around it, and remembering that other voices may compete if they are placed in a stronger part of their range than that of the soloist. A singer is likely to project more prominently when given a solo line.
Difficulties and Limitations
There is a great proliferation of choral groups of widely varying abilities, and more perhaps than with other mediums the composer is likely to write for amateurs and semi-professionals. One of the most important first steps should be careful consultation with the director about the strengths and weaknesses of the group; take these considerations seriously during the writing process. Many choral composers have been able to adapt their demands to fit a group’s ability without losing their integrity, and James Macmillan’s Strathclyde Motets (2005–10) are wonderful examples of highly characteristic works written with simplicity and performability in mind. A number of works continue to suffer neglect, as they were too difficult for the choirs which first performed them and so have fallen into disuse.
Because singing is such an intensely personal means of production requiring great stamina and attention, it is best as far as possible to limit the difficulties in composition to as few areas as necessary at any one time; for example, a highly rhythmically difficult passage will be eased if the melodic and harmonic content is easier. Chromatic scales over long stretches and at speed are particularly disliked and difficult to negotiate with accurate tuning and are best avoided; such a melismatic run is considerably easier if it uses pitches familiar tonally or modally. Glissandi can be very effective but become difficult over large stretches due to the vocal passaggio; start and end points should therefore be considered carefully. Large and angular leaps are challenging for a similar reason, particularly if many of the leaps are in the same direction; whilst the rules of ‘Bach chorale’ style writing are not essential by any means, they are worth recalling in this regard – a large leap up is easier to negotiate if followed or introduced by a move down, particularly if this is by steps. Octaves are easy to negotiate for tuning but again likely to cross between different parts and strengths of the voice. Large leaps become particularly difficult to tune when they involve chords which are unrelated in familiar harmonic terms; as each note must be placed by the singer alone, the methods of pitching differ to other instruments.
Tuning is a prerequisite of good choral singing, but even experienced groups can find extended highly chromatic passages difficult in this regard, as each note will require meticulous tuning by the individual singer. Most singers pitch notes relatively, by interval or approximate musical feel rather than absolutely (i.e. they will not simply sing an E♭ for example, but will find it relative to the other notes and musical contexts, and many will sing even atonal music through some sense of tonal centres and the reading of intervals). Uncertainty is the enemy of choral pitching, and so it is essential that composers consider how singers are to find their notes. In general, close dissonances are better reached, if possible, by preparation and step – it is particularly challenging for, say, two soprano parts to jump to notes which require them to land a semitone apart; the natural temptation will be to reach a matching interval.
Practical solutions include the artfully placed ‘gathering’ moment where a choir can regroup – a recurring chord, or an easily found unison where things can be recalibrated if necessary. A difficult passage followed by an entry together on an unrelated chord is one which may well prove immensely difficult and create terror in performance. Tuning may slip over longer periods, and this becomes an issue when introducing other instruments. In addition, a piece will nearly always require a starting note to be given to the choir as a whole; if a piece begins with an extended section which is non-pitched, then the composer must consider how a choir is to enter on a chord which may be some time after the opening when there have been no other cues. A solution to this can be to ensure a soloist enters first who will have a tuning fork and thus ensure the pitch is established. A golden rule for more choral composers is to sing through each line of the score before handing it over to a choir or vocal group: it is often quite an eye-opening experience to find that what works well on the piano or on instruments presents some great difficulties when sung.
Notation
The key element of choral notation must always be clarity and readability; even more so than the instrumentalist. The singer will need to feel safe and confident in a score and able to provide what it requires. Elaine Gould’s Behind Bars11 is a superb ‘go-to’ guide for choral notation, but here are some broad issues and ideas. Firstly, singers are unique in having to read lyrics as well as musical notation, and to avoid confusion, lyrics are placed directly under the notes to which they refer, and all other markings such as dynamics are placed above the stave. Hairpins are preferable to words such as ‘cresc.’ as they seem to have a greater psychological impact on the singers and help to keep extraneous words to a minimum. Syllables should be divided logically to prevent any ambiguity in the vowel sound which might arise before the full word has been read. If a final syllable is sung on more than one note, then the underscore extender is used. Groups of notes which are sung to more than one syllable should also be slurred to remove any ambiguity. Note that slurs are not commonly used to indicate breaths in choral music unless the music is wordless; some composers do use them to indicate specific phrasing or breath groupings, but in general these are suggested by the words.
The clearest notation is to provide one stave for each vocal part. However, very simple passages can be notated on two staves in ‘closed score’ (N.B. the tenor voice is combined with the bass voice in this notation, in bass clef). In two-stave notation, words should be placed between the staves, and dynamics are placed on either side: occasional differences to underlay may require lyrics to be placed over or under notes in the short passage to which they refer; but if these passages are at all frequent, it will be clearer to use open score. Tail directions are crucially important in this layout, particularly if there is part-crossing at any point. If a voice part divides, then the complexity of the passage will dictate whether the two parts coexist on the same stave or whether a separate one is needed. As with orchestral scoring, simple chords may be written as just that as the lower note will be taken by the seconds and the upper by the firsts. There is no need to indicate ‘unis.’ when a single line returns in this instance. If the division is more complex but still on one stave, then tail direction is essential, and if only one part is singing, rests should be provided or an instruction such as ‘Sop. 1’. ‘Unis.’ should then be used when the music returns to a single line. If the divided voices will occupy two staves, these should be sub-bracketed together with ‘div.’ placed between the staves and the division arrows appearing at the end of the previous line. A return to a single line should be preceded by the opposite arrows and the instruction ‘unis.’ to remove doubt.
The notation of accidentals and intervals needs special consideration, as singers will generally read linearly and intervallically rather than by single specific pitches. It is very important to note that singers usually sing from the full vocal score and will use the other parts as cues for their notes and entries, so it is important that each part be notated in a way which makes sense for the other voices. This will often involve making decisions on enharmonic notation: as many singers will have a strong sense of ‘home’ when singing, so key signatures are useful if a tonal centre is clearly defined and will guide enharmonic writing. If this requires frequent chromatic alteration, it may prove better to write without one.
In music which is intended to be performed unaccompanied, it is standard practice to provide a keyboard reduction unless the music appears on two staves and therefore can be read simply, or too complicated for a reduction to be practically useful. The composer should make a decision as to whether this keyboard part is optional and for use if required, or purely for use in rehearsal. These reductions should never contain lyrics, and if the latter is the case, it is often best to include only the notes and no dynamics. If the reduction produces some passages which are unplayable due to large reaches, then the composer can decide whether to bracket these notes, add vertical rolls, or simply leave them to indicate what is there rather than what can be played. If this reduction is to be playable if required, however, the composer must make decisions on moments like this and also include dynamics.
Some works include spoken moments. If these are to be entirely unpitched and spoken as a narrator, then they can be simply written out. If a spoken moment appears in a vocal part, it should be clarified that this is spoken, as well as whether it is by a soloist or the section. If the section is to speak, it then must be made clear whether this is to be with a sense of ensemble or with every performer doing it in their own time. It is probably still worth marking such moments as ‘normal speaking voice’ just to avoid confusion. Notated speech which is purely rhythmic should use accurately notated rhythms with cross-head notes. More unusual sounds require clear instructions on both delivery and extended effect as well as clear indications on when to return to a normal singing voice. The presence of noteheads which appear to show specific pitches is a very powerful pull, and if these pitches are not intended, then it is best to notate them with crosses or headless notes as mentioned earlier in this chapter to remove this ambiguity. The same applies to aleatoric sections and their notation – the composer must be as clear as possible in their instructions as to what is required and the desired outcomes, and precise instructions on how and when the performers return to the written score.
Conclusion
As one of the most ancient of mediums, choral composition continues to prove itself capable of infinite rejuvenation. The voice possesses the unique ability to express and describe concrete thoughts and ideas and bring a depth of feeling through music which can create a fusion of unrivalled visceral and emotional power. Each voice is unique, and even through the blend of choral sounds each choir maintains its own identity, and this also helps to refresh the medium and inspire the composer with fresh ideas for different groups. Each performance of any piece varies, but choral performances are likely to be more divergent than other mediums due to this individuality. Closely tied to this and worthy of a final iteration is that singers will do their utmost to take care of their voices and be unwilling to subject it to damage. In addition, the personal and entirely self-contained form of production of sound requires singers of all abilities to be confident about what is required, and confident of their abilities to deliver it. This, and the provision of clarity should be the highest aim, regardless of style, difficulty, or scale.
