will come out roundly’ (Prose Works, Eng. trans.,
1894, iii, 202).Singing is a fundamental mode of musical expression. It is especially suited to the expression of specific ideas, since it is almost always linked to a text; even without words, the voice is capable of personal and identifiale utterances. It is arguably the most subtle and flexible of musical instruments, and therein lies much of the fascination of the art of singing.
Because it imparts to words a heightened expression that they do not have when merely spoken, or even declaimed in a dramatic manner without musical pitch, singing (or incantation) played a vital role in many early forms of religious ritual, and in the early theatre. Even outside religion, singing has long been held to have moral and cultural value. Aristotle quoted the bard Musaeus, ‘Song is man's sweetest joy’, and went on to warn against using musical instruments, such as the aulos, which interfere with or prevent the act of singing. Athenaeus (Deipnosophistae, 2nd century) reported that ‘it is no disgrace to confess that one knows nothing, but it is deemed a disgrace among them to decline to sing’. In the history of Western civilization, and of other civilizations, an ability to sing well has repeatedly been viewed as a mark of culture and humanity.
OWEN JANDER/ELLEN T. HARRIS (1, 3–6), DAVID FALLOWS (2), JOHN POTTER (7)
The historical study of the voice is difficult and frustrating. As opposed to the study of instruments, there are no models to examine, and little information can be gleaned from visual depictions of singers. The development of recorded sound helps enormously from the end of the 19th century, but also reveals how little can be gauged from written descriptions. Nevertheless, it is possible to draw some conclusions about the ideals of vocal production from a study of theoretical treatises, vocal tutors and descriptions of singers throughout history.
Two qualities always required in a singer were good (some authors say perfect) intonation and clear enunciation. Tinctoris (De inventione et usu musicae, c1481) listed the qualifications of a good singer as accurate rhythm, a good sense of pitch, enunciation and a good voice (‘ars mensura, modus, pronunciatio, et vox bona’). Giulio Caccini's list (Le nuove musiche, 1601/2; see fig.1) does not markedly differ; he calls for ‘the tuning of the voice in all the notes’, ‘a command of breath’, enunciation (‘unless the words [are] understood’ the singer cannot ‘move the understanding’) and expression (‘to delight and move the affections of the mind’). In the 20th century, Sergius Kagen (On Studying Singing, 1950) still calls for ‘a keen musical ear’, ‘natural singing voice’ (Tinctoris's ‘vox bona’) and proper pronunciation and expression of the text. None of these qualities, however, explains how the voice is produced, and around 1800 a major change took place affecting both singers and composers.
Vocal production entails the use of the vocal registers, otherwise known as head voice (Voce di testa, often equated with Falsetto) and chest voice (Voce di petto). James Nares (A Treatise on Singing, c1780) clearly states the situation of the singer who has moved beyond the beginning stages:
I should have observed that, after the Scholar has gained a good Intonation and some Management of his voice, the Master should make him acquainted with the Compass of his Voice, shewing him where his Voce di petto ends and where to cultivate the falsetto, or Voce di testa, and instruct him how they should be joined, so as to be imperceptible, without which the pleasing variety will be lost.
As Nares implies, singers in the 18th century were taught to blend the registers so as to eliminate the break but also to maintain the ‘variety’ of the distinct sounds. Earlier tutors sometimes encouraged singers to choose to perform in one or the other. Caccini, for example, identifies two registers as the ‘natural’ and the ‘falsetto’ and counsels singers to avoid the latter by performing arias in keys suitable to their natural voice. Bénigne de Bacilly (Remarques curieuses sur l'art de bien chanter, 1668) defines the same two registers and allows that both have their adherents:
Some people are proud of their high voices, and others of their low tone, taking the view that a high voice is little more than a screech. Those who have natural voices scorn the falsetto as being artificial and shrill, while on the other hand falsetto singers are usually of the opinion that the beauty of a song is more evident when performed by the shimmering brilliance of their vocal type than when done by a natural tenor, which, although it ordinarily has better intonation, does not have the brilliance of the falsetto.
With the extended range of vocal music at the end of the 17th century and throughout the 18th, singers had to learn (as Nares states) to use both registers and to unite them. Pietro Francesco Tosi (Opinioni de’ cantori antichi e moderni, 1723), perhaps the first author to address this issue, clearly states the relation of this necessary technique to range:
A diligent Master, knowing that a [male] Soprano [castrato], without the Falsetto, is constrained to sing within the narrow Compass of a few Notes, ought not only to endeavour to help him to it, but also to leave no Means untried, so to unite the feigned and the natural Voice, that they may not be distinguished; for if they do not perfectly unite, the Voice will be of divers Registers, and must consequently lose its Beauty.
The end of the 18th century and the beginning of the 19th witnessed a marked change in production, when singers began to carry the full weight of the chest voice into the highest registers. One of the most significant effects of this change was to make the high voice much more powerful than was possible when using head voice or falsetto on the same notes. The Irish tenor Michael Kelly was among the first singers to produce this sound, as a contemporary description of his voice attests: ‘His compass was extraordinary. In vigorous passages he never cheated the ear with feeble wailings of falsetto, but sprung upon the ascending fifth [from d' to a'] with a sustained energy that electrified the audience’. This change in vocal production reversed the previously taught relationship of dynamic to range.
In 1739, for example, Johann Mattheson (Der voll-kommene Capellmeister) referred to a Latin ‘rule which has already served for two hundred years, that each singing voice, the higher it goes should be produced increasingly temperately and lightly: however in the low notes, according to the same rule, the voice should be strengthened, filled out, and invigorated’. He is probably quoting from Conrad von Zabern, whose Latin treatise on singing (De modo bene cantandi choralem cantum, 1474) had been published in Germany. Zabern is certainly clear on the principle of volume as it relates to range:
Another fault which is more obvious than the others is singing notes with an unstintingly full and powerful voice … When this shouting is done by individuals with resonant and trumpet-like voices it disturbs and confuses the singing of the entire choir, just as if the voices of cattle were heard among the singers … In order to recognize this error completely it must be realized that whoever wishes to sing well and clearly must employ his voice in three ways: resonantly and trumpet-like for low notes, moderately in the middle range and more delicately for the high notes – the more so the higher the chant ascends … Therefore, let him who wishes to sing flawlessly never again presume to sing with a full and strong voice in the upper register, for this disfigures the chant, pointlessly weighs down and fatigues the singer, makes him hoarse and consequently useless for singing … But on the other hand, when one sings with a delicate tone in the upper register the voice then corresponds to the high-pitched sound of the small pipes of the organ, as well as the upper range of the monochord.
Little more than 50 years after Mattheson had cited this principle as having existed for over 200 years, it was overthrown. In 1791, William Jackson (Observations on the Present State of Music in London) complained that ‘instead of developing their voices so as to be soft at the top and full at the bottom, singers were achieving the opposite effect’ (FiskeETM, 1973, p.270), and in 1810 Domenico Corri (The Singer's Preceptor) may have been the first author to instruct that the voice should increase in volume as it ascended and decrease when descending (p.52). With this change, the ground was laid for the development of the dramatic soprano, the Heldentenor and other weighty voices, and of a new repertory that privileged power over brilliance or flexibility. One cannot, however, simply define vocal production in one way before 1800 and in another after; issues of voice range, genre and nationality all contributed to a more complex picture.
Before the 17th century, two main considerations make the topic hard to study. First, the names given to voices in the surviving music normally denoted their function rather than their range or timbre. Thus, for example, ‘tenor’ was the voice-line that stood at the core of the polyphony, sometimes borrowed from chant but almost always the line in relation to which everything else happened. Similarly, ‘contratenor’ was before the 16th century simply a voice that functioned broadly in the same range as the tenor, hence its name.
Secondly, written or named pitches did not generally have any fixed pitch in the modern sense of a frequency. This can be seen most easily in the Gregorian chant repertory, where a piece in the 7th mode would characteristically have a written range from g to a' and a piece in the 2nd mode would have a written range nearly an octave lower, from A to b, but both would almost certainly have actually sounded in the same register. That is, written pitches were chosen not according to the frequency but to give the simplest possible notation of the modality: key signatures other than one flat do not exist before the mid-14th century; and even in the mid-16th century any further key signature was a rare gesture for a special purpose.
So any study of singing in those years must begin from the ranges and relative ranges of the written music. Broadly speaking, Gregorian chant has a range of about a 9th, and must be assumed to be at a pitch comfortable for a large body of singers – normally men, presumably with the choirboys (documented from the 11th century) singing an octave higher; women seem never to have sung alongside men in church, though in nunneries they plainly sang at a pitch that was suitable for them. Exactly what pitch was considered comfortable or suitable in these cases must have depended on techniques and ideals of vocal production, concerning which the available information is mostly anecdotal and hard to interpret with confidence. Chaucer's famous description of the Prioress, whose singing was ‘Entuned in hir nose ful semely’, was intended as light humour and can hardly be used for historical reconstruction.
With the rise of extended monophonic works, starting with the sequence of the late 9th century, there is a marked increase in vocal range. From the time of Notker, sequences often exceed a 12th in range. In the 14th century the monophonic lais of Guillaume de Machaut routinely cover two full octaves: they can last up to 20 minutes, and their very rare modern performances demand extremes of vocal flexibility and stamina. That is perhaps the right context for understanding the description by Hieronymus de Moravia (late 13th century) who mentions vox pectoris, vox gutturis and vox capitis – chest voice, throat voice and head voice.
The earliest two-voice polyphony most often had a vox principalis, often a Gregorian chant, and a vox organalis, which was more florid and had a wider range, but in essentially the same register. That remains the case even in the late 12th century, with the two-voice organa normally credited to Leoninus: here the vox organalis can be exceedingly florid, evidently intended for virtuoso display. Any evaluation of its vocal technique must consider that virtuoso element and a similar manifestation of vocal floridity coupled with intricate rhythms found in much Italian polyphonic song of the 14th century. It is hard to resist thinking that brilliance and lightness of touch characterized the best singing in these repertories. That is in fact spelt out in the Trecento song Oselletto salvaço of Jacopo da Bologna:
Per
gridar forte non si canta bene
Ma con soave et dolce melodia
Si fa bel canto et ciò vuol maestria.
(You do not sing well by shouting loudly, but with sweet and elegant melody fine song is made, and that needs skill.)
More or less the same was said at greater length by Conrad von Zabern in 1474 (see Dyer, 1978).
There is no reason to believe that any polyphony before the 15th century was sung with more than one voice to a part. The liturgical organa of the Notre Dame repertory all set exclusively the solo sections of the chants, simply using three or four soloists rather than just one; the remainder of the chant was sung monophonically by the schola, whose members continued to be the core of any church choir.
In the course of the 14th century the voices in polyphony begin to polarize into two different ranges: increasingly the ‘discantus’ (and occasionally also a ‘triplum’) stood in a range roughly a 5th higher than the tenor and contratenor. This remains broadly true until about 1450, when composers began to cultivate additionally a ‘bassus’ voice in a range roughly a 5th below the tenor.
Around 1440 there is the first clear indication of the pitch area implied by these relative ranges. Two works in the Trent codices, Battre's Gaude virgo and Bourgois’ Gloria, specifically denote sections to be performed by ‘pueri’ alongside other sections marked ‘mutate voces’ (changed voices: presumably adult men). The relative ranges of the Battre piece are, for the ‘mutate voces’, tenor d–d', contratenor d–e', discantus c'–c''; for the ‘pueri’, tenor a–b', contratenor c'–c'', discantus d'–d''. Here the discantus lines of the ‘mutate voces’ sections go as high as those of the ‘pueri’ sections; thus it seems clear that the discantus must have been sung in a high men's range that could also be sung by boys. If it is legitimate to project that information back to the 14th century, it would suggest that the tenor and contratenor lines were in a range of roughly a 10th from tenor c and the discantus a similar range from about g. Certainly there is documentation from the early 16th century that in Italy the master of the choirboys sang along with the boys in unison. That in its turn would mean that the ‘bassus’ lines introduced in the middle of the 15th century were approximately at the pitch of the modern bass.
This conclusion is obviously surprising and remains in dispute, because it implies that polyphony before about 1450 avoided the baritone and bass registers that now seem the most common ‘natural’ voices of grown men. But such arguments are hard to bring any further without firmer information about vocal production and ideals of sound.
A further hint about these matters comes from the chapel statutes of the court of Burgundy codified in 1469 (Fallows, 1981). These state that in performing four-voice polyphony there must be at least six men on the top line, three on the tenor, two on the contratenor (which was then still normally in the same range as the tenor) and three on the bassus. The surprise here is the six on the top line. By good luck the payment lists of the Burgundian court choir in that year contain enough information for it to be certain that there was nobody under 20 years old, so they were not choirboys. It therefore seems almost certain that these were grown men singing in a falsetto register but with an extremely light tone.
Already by the late 15th century there are clear statements of specialization in particular ranges: Tinctoris (De inventione et usu musicae, c1481) describes the different voices and names particularly distinguished exponents, including Ockeghem as a bass. In 1481 Siena Cathedral despaired at losing their tenor singer, despite having two contratenors evidently used to singing in the same range, because ‘senza tenore non si può cantare’. These were the years in which singers such as Jean Cordier and Giles Crepin travelled from court to court, receiving ever-increasing payment for their services.
The church polyphony of the years around 1500 is remarkable for its wide voice ranges. Josquin's masses, for example, seem to expect each voice to have a range of almost two octaves. Again, lightness and flexibility seem to be implied. By contrast, 80 years later, in the Palestrina generation, voice ranges appear to have diminished: only rarely does Palestrina expect a single voice to exceed a 10th; and Nicola Vicentino's L'antica musica (1555) firmly recommends those ranges. While the reasons for this change have not yet been explored, it is plausible to think that one element was a change in vocal ideals: a need for a more focussed sound that concentrated on the best notes in the voice.
Some hint of the change can be seen in the distinction between the quiet voce da camera and louder voce da chiesa, first found in a letter of 1491 (Fallows, 1985, p.64) and most clearly spelt out in a letter of 1568, in which the singer Carlo Durante is reported as saying that he cannot sing with voce da camera because he has recently been singing regularly in church but that when his voice is rested he hopes to be able to sing in the chamber (‘et come la voce sara riposata si crede gli servirà per camera’). It looks very much as though the techniques and ideals of singing in church changed substantially in the 16th century whereas chamber music retained the older style.
The history of singing in the 17th and 18th centuries is characterized by several trends: the rise of the professional opera star, inaugurating a continuous succession of nationally and internationally famous singers; the wide popularity of the castrato and the soprano; the formation and dissemination of the Italian style of singing, along with a concurrent tendency towards national differences; and the cultivation of vocal ornamentation to a peak of artifice. All these trends were supported by specialist teachers of singing, working either independently or in institutions such as the Neapolitan conservatories and the Venetian ospedali. Just as previously the authors of singing treatises tended to be tenors, they now tended to be Italian castratos, and the most important of these treatises, by Tosi and later Giambattista Mancini (Pensieri, e riflessioni pratiche sopra il canto figurato, 1774), must therefore be used with caution when applied to other voices and other countries.
For example, the joining of the head and chest registers over the break, so important to both Tosi and Mancini, was apparently not as valued in France or Germany. J.J. Quantz (Versuch einer Anweisung die Flöte traversiere zu spielen, 1752) writes, ‘Joining the chest voice to the falsetto is as unknown to [German singers] as it is to the French’. Apparently French and German singers continued the older tradition of singing in one register as much as possible, using transposition (as suggested by Caccini) to facilitate this where necessary. Where the compositional range demanded vocal expansion beyond one register, the natural break was probably accepted, as it was in many voices well into the 19th century.
Raguenet (Paralèle des italiens et des françois, en ce qui regarde la musique et les opéra, 1702; Eng. trans., 1709) states that one essential difference between French and Italian opera was the variety of ranges in the French; he especially praises the deep French bass as opposed to the ‘feign'd Basses among the Italians, which have neither Depth nor Strength’. He speaks of the resultant ‘agreeable Contrast’ in French music arising from the ‘Opposition’ of the bass with the treble parts, something that is lacking in Italian music – ‘the Voices of their Singers, who are, for the most part, Castrati, being perfectly like those of their Women’. The partiality of the French to the low bass, or Basse noble, continued past the 18th century as an identifying feature (see Bass (ii)).
Tosi mentions a distinction between the treble voices of castratos and women when he states, ‘Among the Women, one hears sometimes a Soprano entirely di Petto, but among the Male Sex it would be a great Rarity, should they preserve it after having past the Age of Puberty’. Handel wrote for a number of renowned sopranos, and a comparison of the surviving descriptions of their voices with the music they sang confirms that Handel was careful to place the highest notes in weak, unaccented positions, a practice that tends to confirm the use of the head voice and a lesser dynamic in the upper register. There is, however, at least one exception to this practice in the music Handel wrote for Anna Maria Strada (such as in the role of Alcina), where the high notes are frequently accented in both word and rhythm (Harris, 1988–9). Strada may have been one of these women who was able to sing completely di petto. Nevertheless, the definitive change in vocal production towards a strong and resonant upper register did not occur until after 1800.
The first half of the 19th century was a period of significant change in the history of Western singing, especially in opera. Newer categories of voice such as the tenore robusto, tenore di forza, Heldentenor, ‘Verdi baritone’, ‘Falcon soprano’, ‘dramatic soprano’ and lirico spinto reflect a taste for weightier timbres, more brilliant upper registers, more sonorous low notes and increased volume in general. Although the new taste for greater volume and more dramatic expression extended to all voices, its impact is most clearly apparent in the careers of several 19th-century tenors, including Adolphe Nourrit, Enrico Tamberlik, Jean de Reszke and most notably Gilbert Duprez, who became famous (and in some circles infamous) for his use of the Voix sombrée and for his clarion high c''.
The development of the high, powerful tenor voice spelt the end of the reign of the castrato, a tradition that had already waned at the beginning of the century with the substitution of the female Musico for the castrato in heroic male roles such as Rossini's Tancredi. The soprano voice was also extended upward in range and power, leading to the separate development of the dramatic mezzo-soprano, a range closely associated with the parallel development of the baritone. Both are particularly well served in Verdi's operas, to the extent that the dramatic baritone is generally referred to as the ‘Verdi baritone’ (see Mezzo-soprano and Baritone (i)).
Wagner sought dramatic tenor voices of unusual strength and endurance. Although he never used the term ‘Heldentenor’, now closely associated with the Wagnerian tenor type, he adamantly distinguished what he wanted from the French dramatic tenor of his day. The Heldentenor differs from the French and Italian tenor (Tenore robusto) in having a smaller range and a sound closer to that of a baritone. Not surprisingly, many of the most famous dramatic tenors, including in the 19th century Jean de Reszke and in the 20th Placido Domingo, began their careers as baritones.
One of the distinct changes resulting from the cultivation of the heavier voice was the increase in vibrato. At first considered an ornament in the expression of passion, vibrato was not generally considered acceptable as a constant part of vocal production before the end of the 19th century. A particularly clear early injunction against it was given by Christoph Bernhard (Von der Singe-Kunst, oder Maniera, c1649):
Fermo, or the maintenance of a steady voice, is required on all notes, except where a trillo or ardire is applied. It is regarded as a refinement mainly because the tremulo [sic] is a defect … Elderly singers feature the tremulo, but not as an artifice. Rather it creeps in by itself, as they no longer are able to hold their voices steady. If anyone would demand further evidence of the undesirability of the tremulo, let him listen to such an old man employing it while singing alone. Then he will be able to judge why the tremulo is not used by the most polished singers, except in ardire.
In the 18th century, too, Tosi warned singers to learn to hold notes without vocal ‘trembling’, for those who do not ‘will become subject to a Flutt'ring in the Manner of all those that sing in a very bad Taste’. In the 19th century ‘vibrato’ is written as a special instruction at certain points in scores of Donizetti, Halévy, Meyerbeer and others. Although it continued to be criticized, many Italian singers seem to have begun using an audible vibrato on every sustained note by the middle of the century.
The
significant changes in vocal production in the first half of the 19th century,
including the use of the chest voice and increased volume in the upper
register, together with the increasingly continual use of vibrato, were not
universally welcome. Rossini is said to have exclaimed, ‘Alas for us, we have
lost our bel canto’, and it is right around this time that the phrase,
‘beautiful singing’, took on a specific meaning. Associated with legato
production, light tone in the upper register, and agile and flexible delivery,
it was contrasted with the weightier, speech-inflected (Sprechgesang) style.
While for its adherents the term became both a nostalgic symbol for a declining
tradition and a battle cry for its revival, for its detractors it was simply
pejorative: Wagner, for example, derided the bel canto model that was concerned
only with ‘whether that G or A will come out roundly’ (Prose Works, Eng. trans.,
1894, iii, 202).
Another development of this period was the increasingly ‘scientific’ approach to singing and vocal production. The most important and influential publication was the Traité complet de l'art du chant (1840) by Manuel Garcia, baritone and singing teacher. Garcia's invention of the laryngoscope in 1855 furthered the increasing interest in the physiological properties of the voice. His teaching method was based upon a thorough understanding of the workings of the ‘instrument’ (larynx, throat, palate, tongue etc.), and his work became a model for numerous books on singing after 1850.
The trends in concert and operatic singing in the 19th century, towards a new sense of grandeur, in terms of the size of the halls and the size and volume of the orchestra as well as in the production of the voice (all of which developments were closely related), were offset by the largely contemporary rise of the solo song with piano accompaniment that called for an almost unprecedented intimacy. Although increased intimacy was also evident in operatic characterization, it was especially manifest in a new class of singer who specialized in recitals and oratorio, such as the baritone Julius Stockhausen for whom Brahms wrote his Magelone songs and the baritone part in the German Requiem.
The change in vocal production in the 19th century and the consideration of the voice as an instrument affected all repertories. Singers of songs could and sometimes did have smaller voices than their operatic counterparts, but this was not essential and, indeed, much of the song repertory called for strength and endurance in the upper register, such as Schubert's ‘Suleika’ songs. The primary effect was to draw a clear distinction between operatically trained and ‘popular’ singers, a distinction that had not existed in the 18th century, when the English tenor John Beard could move easily among Italian opera, English oratorio, popular ballad opera and English song.
The most important development for singing in the 20th century was the invention and expansion of electronically altered and amplified sound as well as the vast proliferation of recorded sound. Techniques of electronic amplification, including the microphone, have, especially in popular singing, altered vocal production, further dividing the ‘classical’ and ‘popular’ singer. Without the need to project the voice naturally over robust (and often amplified) accompaniments in vast halls, popular singers, such as Billie Holiday and Frank Sinatra, did not need to power their voices physically (through diaphragmatic support, use of chest voice and breath control), but could make previously inaudible intimate vocal nuances, such as whispers or murmurs, audible in live performance. Subsequently expanded vocal experimentation in various rock music genres led to the inclusion of screams, growls and the like and the manifestation of the male falsettist (Michael Jackson and many predecessors) who specializes in intense and distorted sound in the upper registers (see §7 below).
Experimentation with sound was not limited to popular singing. The technique of Sprechstimme, a highly stylized mode of vocal expression halfway between singing and speaking, was associated particularly with the so-called Second Viennese School in such works as Schoenberg's Pierrot lunaire (1912). Composers of both choral and solo music also explored a range of special vocal effects, including choral recitation, Bocca chiusa, glissando and controlled shouting. Since 1950, electronic amplification and alteration have also been increasingly used by classical composers. In Crumb's Ancient Voices of Children (1970), for example, a soprano sings (or shouts) into an open piano and the sympathetic vibrations of the undamped strings are then amplified through a contact microphone. Singers such as Bethany Beardslee, Jane Manning and Cathy Berberian have specialized in singing avant-garde music and in developing the new techniques that this entails.
Perhaps the most radical innovation in 20th-century singing is, paradoxically, allied to the movement towards historically accurate performances of early music. Although at first more focussed on performing practice (e.g. in matters of ornamentation) and on the use of period instruments and correct ways of playing them, the rediscovery of the pre-1800 singing style has transformed and reinvigorated contemporary understanding and appreciation of early music. The countertenor voice, which had remained in constant use in England since the Renaissance but had been forgotten elsewhere, was rediscovered by an international audience in the 1950s, chiefly through the career of Alfred Deller. Since then singers of all ranges, such as the soprano Emma Kirkby, the mezzo-soprano Lorraine Hunt Lieberson, the countertenor Andreas Scholl and the bass David Thomas, have become specialists in the performance of a wide repertory of early music. Interesting points of contact exist between the early music movement and the avant garde, including a similar singing style that emphasizes a pointed tone with little or no vibrato (see Bass (ii), Countertenor, Mezzo-soprano and Soprano).
Recording technology, although it does not necessarily alter the production of sound, has had a great effect on both popular and classical singing. It provides recorded documentation of specific singers, yielding more information than any verbal account has ever been able to convey, and further permits composers to supervise recorded documentation of their intentions, allowing a more precise transmission than is possible in a written score. Ironically, at least in the performance of classical music, the technology that made it possible to capture and disseminate the remarkable variety of singers' styles has tended to encourage stylistic norms and led to an increased internationalization and homogenization of sound production and performing practice. In popular music, individuality, sometimes reaching to the extremes, has been more welcome.
The various 20th-century singing styles – classical, popular, avant-garde, early music – manifest certain resemblances that help to define this period. First, experimentation with new sounds has been a dominant trend in all forms of new music, whether classical or popular; secondly, a retreat from the 19th-century sensation of pulling the chest voice into the upper register (with a subsequent reduction in vibrato) has been a factor not only in popular folk singing but also in both early and contemporary music performance; thirdly, a renewed preference for extremely high voices can be seen in the dominance of the soprano in avant-garde music, the falsettist in popular music and the countertenor in early music (and the virtual disappearance of the contralto); and, fourthly, the dramatic rise in so-called crossover projects has led to various kinds of transference, such as, on the one hand, the association of the tenor Pavarotti with the rock artists U2, Sting and others in his ‘Pavarotti and Friends’ projects, and, on the other, the rock singer Michael Bolton's forays into opera.
One of the most important aspects of singing is the way singers have interpreted the notes on the page. Portamento, for example, was considered an essential element of good singing until about the beginning of the 20th century: that is, singers connected notes ‘almost imperceptibly’ by gliding through the intervening pitches. The abuse of this technique at the end of the 19th century and the beginning of the 20th led to its abandonment, but the practice encouraged by Garcia in 1894 of hitting each note ‘purely’ has no basis in earlier history (see Coup de glotte). Singers not only connected notes, but approached initial notes from as far as a 3rd or 4th below the notated pitch (Cercar della nota). A consistent notation for the portamento and cercar della nota was never developed, in part because the practice was so normative that notation would have been redundant. Thus our record of this practice is largely limited to written treatises on singing and descriptions of voices, but early recordings also document its regular use (Crutchfield, 1983).
Also unnotated are such important ornaments as the Messa di voce, or a crescendo and diminuendo on a single sustained note. Caccini (1601/2) considered this practice ‘the foundation of passion’, the 18th-century castrato Farinelli was particularly renowned for his exquisite messa di voce, and Garcia (Traité, 2/1847) recommended the singing of scales with a messa di voce on every note (in order to unite the registers and develop volume). A gradual increase or decrease of volume over phrases was also recommended by tutors; Bernhard (c1649) writes, ‘Care must be taken not to shift too abruptly from the piano to the forte, but rather to let the voice wax and wane gradually’. So-called terrace dynamics, popular in the mid-century performances of Baroque music, are not supported by contemporary sources. Not only the treatises, but also scores from at least the early 17th century, indicate the use of crescendo and decrescendo rather than abrupt dynamic change.
Rhythm and metre have also been treated flexibly. Girolamo Frescobaldi (in the preface to his Toccate, 1615) speaks of performing instrumental pieces like madrigals, ‘taking it now slowly, now quickly, and even held in the air, to match the expressive effect’. Bacilly (Remarques curieuses, 1668) encourages singers to slow down in order to add embellishments, and Tosi (1723) describes ‘stealing the Time’ in order to avoid ‘a mechanical Method of going on with the Bass’. In the 19th century singers made liberal use of rubato, as the earliest recordings document. Rhythmic freedom was always of paramount importance in recitative and declamatory singing.
Ornamentation, in terms of the addition or alteration of the notes, is somewhat easier to document as composers and singers from the end of the 16th century to the present have left examples of their practice (see Ornaments and Improvisation). Caccini (1601/2) not only described the most important ornaments but published arias from his Il rapimento di Cefalo as they had been sung and ornamented by the bass Melchior Palontrotti, the tenor Francesco Rasi and the tenor-composer Jacopo Peri. Handel ornamented some arias in his own hand, and ornamentation used by the 18th-century singers Francesca Bordoni and Farinelli also survives. Many examples (including some by Haydn and Mozart) survive from the Classical era (see Crutchfield, ‘The Classical Era’, 1989). Rossini left many manuscripts illustrating the ornamentation of his arias, as did Verdi; such 19th-century singers as Cinti-Damoreau, Viardot and Kemble left notebooks with their ornamentation (see Crutchfield, ‘The 19th Century’, 1989). The invention of sound recording has further facilitated the comparison of earlier ornamental styles. In the 20th century, however, fewer composers were likely to assume or even desire rhythmic or melodic improvisation in performance. This is evident in the use of pre-recorded tape but also, for example, in Stravinsky's assertion that music should be not be interpreted but should rather be objectively executed, an attitude that found its way for a time into 20th-century attitudes towards earlier music. However, the application of such principles to earlier music, as has happened in some performances of Handel's and Rossini's operas, is not only inappropriate but often damaging.
The earliest known references to what might be called popular singing in the West are the work songs referred to in the works of Homer, which are assumed to have been sung in a ‘natural’ speech-related way that can still be heard in certain European folksongs. Medieval literature contains references to oral music which depended on singing that was untrained and inspirational, and which was identified with the singers who created and performed it rather than composers. From the 17th century onwards the growing commercialization of popular entertainment saw an increasing variety of popular singing styles ranging from speech-related folk singing to more stylized varieties derived from classical singing, which diverged significantly from popular varieties in the early 19th century with the evolution of the more efficient low-larynx technique and its associated breath control.
In the USA a multi-faceted popular oral tradition, with European folksong and African-American blues as major influences, was the earliest popular singing to interact with technology, first in the form of recordings in the early years of the 20th century and then by the use of amplification from the 1920s onwards. Early black American jazz singers such as Louis Armstrong and Bessie Smith were a crucial influence on white singers such as Bing Crosby, who used the microphone to take an intimate, speech-related style to large audiences, broadening the popular base of jazz-influenced singing from a black minority interest to something that caught the imagination of millions of people. Speech uses a higher larynx-position than classical singing, which modifies speech patterns as part of the projection process. Microphones enable singers to dispense with the mechanisms of projection and retain the nuances of speech, thereby seeming to create a feeling of a one-to-one relationship with their listeners: hence the misleading term ‘crooning’, which was applied to certain microphonic singers in the 1930s and 40s.
The speech-song of Armstrong and the early Crosby became more singer-like and sophisticated with their successors, culminating (after a digression into the virtuosity of bebop) in the work of Frank Sinatra, who even described his own singing as ‘bel canto’. Sinatra sang with speech-related word shaping, but used efficient breath control and was concerned with tone-colour (and, especially in his later recordings, with vibrato) in a similar way to classical singers. Postwar rock and roll, vocally a synthesis of country music and rhythm and blues, was very much a reaction to this ultra-sophistication, Elvis Presley in particular representing the visceral ebullience of a youthful return to speech-related singing in which the voice was the servant of textual rhetoric. Rock and roll's rather limited musical potential was given new life by the Beatles and others during the 1960s. The Beatles’ early catholic taste embraced black American music and musicals as well as conventional rock and roll. Their second album, With the Beatles (1963), included Paul McCartney singing ‘Till there was you’, a ballad from the musical The Music Man, which extended the range of speech-related singing to music that was originally conceived for the sub-classical singing of previous generations. The punk phenomenon of the mid-1970s was in part a reaction to the all-embracing stylistic and commercial tentacles of rock. Before they, too, were subsumed into the mainstream, punk singers briefly outraged the establishment with their aggressive recitatives performed in a kind of heightened speech. The real revolution in the late 1970s came once again from the African-American community, when New York black youths began to have some commercial success with rap. Rap is heightened rhymed speech; many variants have developed under the general term ‘hip-hop’. Ironically, the declamatory style has theoretical echoes of the ‘classical’ dramatic rhetoric of ancient Greece.
Classical singing has a rigid classification of voice-types. Pop singers tend to be loosely categorized according to genre (folk, soul, rock, rap and so on). The evolution of rock has generally seen a narrowing of vocal ranges, with men singing in the upper part of the voice and women in the lower (Kate Bush is an exceptional high soprano, Tom Waits a rare bass), the overlap of tessitura between the sexes perhaps signifying a certain ambiguity towards gender in the late 20th century. There is little formal pedagogy associated with pop singing, which is able to use any rhetorical means to deliver its message, unencumbered by a systematized technique (a contrast vividly demonstrated by the duetting of Freddie Mercury and Montserrat Caballé on the 1988 album Barcelona). Popular singing is potentially the most democratic means of music-making (anyone who can speak can sing) and has always shown a rich diversity of styles (from bebop to doo-wop, techno to hip-hop). It is increasingly enhanced by technology: the sound system on which commercial pop music depends has evolved from simple amplification to a creative tool which can modify tone-colour and create simultaneous harmonies, but the major stylistic changes of the 20th century were connected with the need for popular singing not to stray too far from its primary purpose: to express a text in a speech-like way that is relevant to all.
See also entries on individual voice-types and Acoustics, §VI.
GroveO [comprehensive bibliography to 1992]
H. Goldschmidt: Die italienische Gesangsmethode des XVII. Jahrhunderts und ihre Bedeutung für die Gegenwart (Breslau, 1890)
T. Gérold: L'art du chant en France au XVIIe siècle (Strasbourg, 1921/R)
V. Ricci: Il bel canto (Milan, 1923)
R. Duhamel: ‘Le chant moderne: sa matière, sa nature, sa forme et ses moyens’, RMI, xxxii (1925), 80–97, 219–38, 599–620
R. Duhamel: ‘La question de la vocalise dans le chant français’, RMI, xxxiii (1926), 363–85
H. Biehle: Die Stimmkunst (Leipzig, 1931–2)
H. Osthoff: ‘Der Gesangstil der frühdeutschen Oper’, Singen und Sprechen: Frankfurt am Main 1938, 11–16
N. Fortune: ‘Italian 17th-Century Singing’, ML, xxxv (1954), 206–19
F. Husler and Y. Rodd-Marling: Singing: the Physical Nature of the Vocal Organ: a Guide to the Unlocking of the Singing Voice (London, 1965, rev. 1976)
H. Pleasants: The Great Singers (New York, 1966)
R. Celletti: ‘L'insegnamento del canto nell'ottocento’, Educazione musicale, v/3 (1968), 71–7
R. Celletti: ‘Origini e sviluppi della coloratura rossiniana’, NRMI, ii (1968), 872–919
R. Celletti: ‘Il vocalismo italiano da Rossini a Donizetti’, AnMc, no.5 (1968), 267–94; no.7 (1969), 214–47
E.V. Foreman: A Comparison of Selected Italian Vocal Tutors of the Period circa 1550 to 1800 (diss., U. of Illinois, 1969)
D. Galliver: ‘Favolare in armonia: a Speculation into Aspects of 17th-Century Singing’, MMA, iv (1969), 128–46
D. Galliver: ‘Vocal Colour in Singing in 17th-Century Italy: the Contribution of Caccini’, IMSCR XI: Copenhagen 1972, 385–7
D. Galliver: ‘Cantare con la gorga: the Coloratura Technique of the Renaissance Singer’, SMA, vii (1973), 10–18
D. Galliver: ‘Cantare con affetto: Keynote of the bel canto’, SMA, viii (1974), 1–7
L. Manén: The Art of Singing: a Manual (London, 1974)
J.B. Steane: The Grand Tradition: Seventy Years of Singing on Record (London, 1974, 2/1993)
A. von Ramm: ‘Singing Early Music’, EMc, iv (1976), 12–15
J. Dyer: ‘Singing with Proper Refinement: from De modo bene cantandi (1474) by Conrad von Zabern’, EMc, vi (1978), 207–27
A. Ransome: ‘Towards an Authentic Vocal Style and Technique in Late Baroque Performance’, EMc, vi (1978), 417–19
H.M. Brown: ‘Embellishing Eighteenth-Century Arias: on Cadenzas’, Opera &Vivaldi: Dallas 1980, 258–76
R. Bowers: ‘The Performing Ensemble for English Church Polyphony’, Studies in the Performance of Late Mediaeval Music: New York 1981, 161–92
D. Fallows: ‘Specific Information on the Ensembles for Composed Polyphony, 1400–1474’, ibid., 109–59
M. Uberti: ‘Vocal Techniques in Italy in the Second Half of the 16th Century’, EMc, ix (1981), 486–95
G. Buelow: ‘Vocal Ornamentation in the Sacred Music of the Schütz era’, American Choral Review, xxiv/2–3 (1982), 5–13
W. Crutchfield: ‘Verdi Performance: Restoring the Color’, High Fidelity/Musical America, xxxiii (1983), 64–6, 100–01
E. Carei: ‘Le techniche vocali del canto italiano d'arte tra il XVI e il XVII secolo’, NRMI, xviii (1984), 359–75
G. Newton: Sonority in Singing: a Historical Essay (New York, 1984)
D. Fallows: ‘The Performing Ensembles in Josquin's Sacred Music’, TVNM, xxxv (1985), 32–64
D. Laing: One Chord Wonders: Power and Meaning in Punk Rock (Milton Keynes, 1985)
S. Mamy: ‘L'importation des solfèges italiens en France à la fin du XVIIIe siècle’, L'opera tra Venezia e Parigi: Venice 1986, 67–89
R. Bowers: ‘The Vocal Scoring, Choral Balance and Performing Pitch of Latin Church Polyphony in England, c.1500–58’, JRMA, cxii (1987), 38–76
J. Sundberg: The Science of the Singing Voice (Dekalb, IL, 1987)
G. Moens-Haensen: Das Vibrato in der Musik des Barock: ein Handbuch zur Aufführungspraxis für Vokalisten und Instrumentalisten (Graz, 1988)
E. Vecchiarelli: Il canto rossiniano e la critica (Pesaro, 1988)
E.T. Harris: ‘Das Verhältnis von Lautstärke und Stimmlage im Barockgesang’, Aufführungspraxis der Händel-Oper: Karlsruhe 1988 and 1989, 157–71
W. Crutchfield: ‘The Classical Era: Voices’, Performance Practice, ii: Music after 1600, ed. H.M. Brown and S. Sadie (London, 1989), 292–319
W. Crutchfield: ‘The 19th Century: Voices’, ibid., 424–58
W. Crutchfield: ‘Some Thoughts on Reconstructing Singing Styles of the Past’, Journal of the Conductors’ Guild, x/3–4 (1989), 111–20
S. Frith: ‘Why do Songs have Words?’, Music for Pleasure (Cambridge, 1989), 105–28
E.T. Harris: ‘The Baroque Era: Voices’, Performance Practice, ii: Music after 1600, ed. H.M. Brown and S. Sadie (London, 1989), 97–116
E.T. Harris: ‘Hoch und zart: das Verhältnis von Lautstärke zum Tonumfang im barocken Kunstgesang’, OW, xxx/11 (1989), 58–60
W. Friedwald: Jazz Singing: America's Greatest Voices from Bessie Smith to Bepop and Beyond (London, 1991)
R. Barthes: ‘The Grain of the Voice’, Image-Music-Text (London, 1993), 179–89
D. Higginbotham: Performance Problems in Contemporary Vocal Music and some Suggested Solutions (diss., Columbia U. Teachers College, 1994)
J. Potter: ‘The Singer, not the Song: Women Singers as Composer-Poets’, Popular Music, xiii (1994)
J. Morehen, ed.: English Choral Practice, 1400–1650 (Cambridge, 1995)
S. Carter, ed.: A Performer's Guide to Seventeenth-Century Music (New York, 1997), ‘Vocal Issues’, 3–66 [incl. sections by J. Baird, A. Harrington Heider, S. Sanford, G. Towne]
J. Potter: Vocal Authority: Singing Style and Ideology (Cambridge, 1998)
D. Toop: Rap Attack 3: African Rap to Global Hip-hop (London, 1999)
J. Potter, ed.: The Cambridge Companion to Singing (Cambridge, 2000) [incl. articles by S. Banfield, J. Dyer, F. Laurence, D. Mason, R. Middleton, J. Potter, D. Toop, R. Wistreich]