Rendering the Sense More Conspicuous: Grammatical and Rhetorical Principles of Vocal Phrasing in Art and Popular/Jazz Music

Abstract

Singers two hundred years ago did not deliver the notated texts of recitatives, arias, and songs as literally as their modern counterparts do. Indeed, singers in the earlier era saw their role more as one of re-creation than of simple interpretation. Consequently, they altered the texts before them by varying the elements of expression known in their musical culture. Treatises from the period give detailed explanations of these elements, and a striking similarity exists between the method of phrasing that treatises describe and the practices exhibited on the recordings of modern jazz and popular singers. Holly Cole's recording of My foolish heart presents an aural image of a fascinating way of singing popular/jazz music that comfortably maps on to the verbal depictions and notated examples of grammatical and rhetorical pausing that survive from the eighteenth and early nineteenth centuries. Pauses are essential in sung lines, and a highly articulated style of phrasing results from the application of stops within sentences: before prepositions, conjunctions, relative pronouns, and vocatives; after nominatives and prepositional phrases; between substantives and verbs and their objects; and in other places where the sense of a sentence would be made more conspicuous through the introduction of a pause.

Recent research on past approaches to singing reveals that singers in the late eighteenth and early nineteenth centuries did not deliver the notated texts of recitatives, arias, and songs as literally as their modern counterparts do.¹ Indeed, singers today typically render the music of composers from Handel to Rossini much closer to the way it survives on paper than did singers two hundred years ago, the common modern practice being one of interpreting the notation within a narrow range of parameters so as to minimize the creative contribution of the performer, leaving the composer's score to speak for itself. A number of scholars have noted the inclination of performers in the second half of the twentieth century to revere the sanctity of the notated text, and Clive Brown, in outlining the development of a 'cult of the urtext', suggests that 'many modern musicians, including advocates of period performance, have invested these [urtext] editions with a mysterious, almost sacrosanct quality, as if the more literally the notes, phrasings, dynamics, and so on, which constitute the composer's latest ascertainable version of the work, are rendered, the closer the performance will be to the ideal imagined by the composer'.² Roger Freitas has observed this tendency in performances of Verdi's operas and concludes that 'modern performers have sometimes judged the only requirement for an authentically Verdian performance to be strict adherence to the musical text, as given in the best editions'.³ Moreover, Robert Philip, in a fascinating study of early recordings, documents the trend in the first half of the twentieth century towards, among other things, 'literalness' and 'evenness of expression', two important aspects of the then emerging modern style that were prized by musicians who desired greater clarity and precision in the delivery of notated text.⁴

In the late eighteenth and early nineteenth centuries, however, singers viewed notation quite differently. In order to transform inexpressively notated recitatives, arias, and songs into passionate musical declamation, they treated the score freely and personalized the music through both minor and major modifications.⁵ In other words, they saw their role more as one of re-creation than of simple interpretation.⁶ Because the final shaping of the music was the responsibility of the singer, the music an audience heard often differed substantially from what appeared in print, and Domenico Corri, writing around 1781, characterized the relationship between performance and notation candidly: 'either an air, or recitative, sung exactly as it is commonly noted, would be a very inexpressive, nay, a very uncouth performance'.⁷ Corri's comments resonate throughout the period, and tutors regularly advised performers to pay more attention to expressing the passion of the subject than to mechanically following the notation. Charles Smyth stated in 1817 that 'singers are apt to deliver the words [in recitative] too strictly according to the time of the notes to which the composer has adapted the words. It was necessary for the composer to fill up his bar: but he never intended the singer should pay mechanical attention to his notation',⁸ and around 1850, John Addison considered the composer's notation to represent only 'the Skeleton of his ideas'. The rest, he continued, 'is left to the Singer, who must give the finish according to his taste and judgment'.⁹ Moreover, Manuel Garcia, the younger, suggested in 1857 that performers alter pieces in order to heighten their effect or render them suitable to the power and character of an individual singer's vocal capability.¹⁰ Clearly, performers were expected to modify the music they sang, and composers routinely left the nature of the enhancements, whether minor or major, to the sagacity and discretion of the singer.¹¹

The principles governing this re-creative style of singing are described in treatises from the era; and even though we must rely on verbal depictions of the performer's art to generate an aural image of what was once a living tradition, the extant treatises provide a wealth of information on how singers altered the texts before them. Performers, we learn, achieved a persuasive delivery by varying not only the elements of expression which formed the basis of both spoken and sung discourse (emphasis, accent, tone of voice, phrasing (pauses and breathing), and gesture) but also those resources that are peculiar to singing (legato, staccato, portamento, messa di voce, tempo (including rhythmic rubato), register, vibrato, and ornamentation).

These elements probably are shared by many of the Western styles of singing practised over the past two hundred years, and particular manners of performance depend on the precise ways these devices are realized and the degree to which a single component or cluster of components is exhibited. The predilections of musicians to favour or disfavour these elements differ from region to region and change over time, and the exact mix of features singers adopt determines not only the collective style of an era or locale but also individual habits within an identifiable general style. Late eighteenth- and early nineteenth-century treatises inform us that European singers were inclined to favour all the elements listed above except vibrato. They sang in an emphatic way, accenting important syllables in words, matched register and the tonal quality of the voice to the emotional content of the words, employed a highly articulated manner of phrasing, varied their delivery with several types of legato and staccato, liberally applied more than one type of portamento, considered messa di voce to be one of the principal sources of expression, altered tempo frequently through rubato and the quickening and retarding of the overall time, introduced a wide variety of graces and divisions into the music they sang, and regarded gesture as a powerful tool for enhancing the effect of their delivery. They reserved vibrato, however, for heightening the expression of certain words and for gracing longer notes.¹²

But in the early twentieth century, as 'classically' trained singers began to narrow the range of expressive devices they chose to employ, interest in re-creating a work through substantial modifications, using the devices just listed, gradually waned, eventually reaching the point where a more literal reading of a text became, as it remains, the norm. Singers either abandoned altogether or minimized many of the elements of expression upon which the re-creative style of singing was based in the late eighteenth and early nineteenth centuries. This re-creative style continued to be fashionable throughout the nineteenth century,¹³ and during the first thirty or so years of the twentieth century, a significant decrease occurred in the use of ornamentation (particularly divisions), tempo alterations, rhythmic displacement, portamento, gesture, changes of register, and highly articulated phrasing. Vibrato, on the other hand, began to be applied continuously, instead of selectively.¹⁴ The 'evenness of expression', to borrow Robert Philip's words, that resulted from this shift in taste ushered in a new manner of singing art music that came to dominate the habits of most 'classical' musicians.¹⁵ Yet despite the ascendancy of clarity, precision, and evenness as performance ideals in the world of art music, a number of the principles upon which the earlier re-creative approach rests persist as central tenets of singing in other arenas. Jazz and popular artists, for example, tend to prize many of the elements of expression that had formed the basis of singing in the late eighteenth and early nineteenth centuries, and ornamentation, tempo fluctuation, rhythmic displacement, highly articulated phrasing, gradations of legato and staccato, portamento, messa di voce, changes of register and tonal quality of the voice, as well as selectively introduced vibrato, all figure prominently in jazz and popular styles of singing.

Naturally, without recorded performances of late eighteenth- and early nineteenth-century singers, one cannot prove that a continuity of practice exists between the two eras. Musical notation on its own is of limited value for studying performing practices, because scores, unlike recordings, which capture actual performance (whether live or in a studio), reduce expression to a set of symbols that at best can only approximate the unwritten nuances that constitute a performing style.¹⁶ Nonetheless, at least one editor from the late eighteenth century, Domenico Corri, endeavoured to notate some of the subtleties of performance as accurately as he could, and when his scores are coupled with treatises, a reasonably accurate understanding of certain aspects of singing, particularly phrasing, can be constructed.¹⁷ In fact, jazz and popular singers as diverse as Fanny Brice, Billie Holiday, Perry Como, Karen Carpenter, Herb Alpert, Holly Cole, Mariah Carey, and Faith Hill, to mention just a few, use some of the elements of expression listed above in ways that parallel the depictions of the techniques in late eighteenth- and early nineteenth-century treatises,¹⁸ and a striking similarity exists between the method of phrasing Corri and his contemporaries describe and the practices exhibited on the recordings of a number of these artists. An exploration of the similarities suggests that there may be a style of vocal phrasing, rooted in grammatical and rhetorical principles of speaking, common to the re-creative approaches of both eras. Although one cannot determine if these principles are employed in precisely the same ways by the two groups of singers, the practices of jazz and popular artists provide a fascinating model for performing late eighteenth- and early nineteenth-century art music in a manner that seems to correspond more closely with historical documents than the literal approach taken by today's 'classically' trained singers.

In the late eighteenth and early nineteenth centuries, the influence speaking had on singing was profound, and treatises that circulated in England inform us of the close connection between the spoken and the sung word.¹⁹ Indeed, William Gardiner, writing in 1832, observed that in singing the elements that comprise music and language should be blended together:

To sing with taste and expression, many qualifications are required:-first, as music, voice, ear and execution; secondly, as language, enunciation, mind, and action. These, when combined with a just feeling, constitute the highest point of vocal excellence . . . to blend the singing and speaking voice together-to unite them artificially in song-is a great achievement.²⁰

Other authors, such as Richard M. Bacon, reinforce Gardiner's observations. Bacon comments:

the analogy between the elocution of reading or public speaking of any kind and singing, is very complete-they scarcely differ at all but in degree. The effects of reading or declamation are produced by the quality of tone, by inflexion, by emphasis, and by total cessations or pauses. Singing seems only to heighten these effects by using in a bolder manner the same agents. The principles of both are the same.²¹

Moreover, the singer Anselm Bayly,²² writing in 1771, noted that recitative 'is an expressive and elegant manner of speaking . . . let [singers] ask themselves how an orator would pronounce [the words], preserving the grammatical connection, touching lightly, without any appoggiatura, short syllables and unimportant words, and giving a due, but not fierce, energy to the emphatic'.²³ Other singers active in eighteenth-century England also applied these principles when singing, for Charles Dibdin, a composer of songs and operas as well as a singer, praised the ability of Elizabeth Sheridan and Gertrud Mara to sing as they would speak:

Those who get at the force and meaning of the words, and pronounce them as they sing with the same sensibility and expression, as it would require in speaking, possess an accomplishment in singing beyond what all the art in the world cannot convey . . . Mrs. Sheridan and Madam Mara were, according to my idea, the most accomplished singers I ever heard, because they were taught upon this principle.²⁴

All these authors describe the fundamental relationship between speaking and singing, and two of them, Bacon and Bayly, highlight the importance of pausing to both speakers and singers. With the words, 'preserving the grammatical connection', Bayly alludes to the role of punctuation (that is, the pauses associated with commas, colons, and full stops) in vocal delivery, and Bacon specifically refers to the use of 'total cessations or pauses' by singers. Furthermore, one of Bacon's contemporaries, Thomas Williams, recognizes that 'to become an Orator in Song, it is indispensably necessary that the true sense and meaning of the words should be strictly attended to, and the breath be taken according to the proper punctuation, just as if the vocal passage, instead of being sung, were read by [one] adept at elocution'.²⁵ Clearly, the introduction of pauses in singing would produce a highly articulated style of delivery, and Haydn Corri, son of Domenico Corri, identified the essence of the approach with this suggestion:

Let the language of the Song direct the judgment of the Scholar[;] whenever a Phrase will allow a division without destroying the sense of the Poetry, then take the opportunity for respiration; as it is better to have too much breath, than too little-therefore, if you can merely get through a line in one breath, it is better to divide it, and sing it in two respirations; as for instance. If you can sing eight notes tolerably in one breath, how much more fine quality-force-articulation and ease to yourself, must be ensured by singing only 4 notes in a breath.²⁶ (Ex. 1)

Ex. 1.

Breathing to give 'superior elegance of delivery [to] the language' (H. Corri, Delivery, 34)

In order to reconstruct the principles of phrasing that singers in the late eighteenth century, as well as writers such as H. Corri, Bacon, and Bayly, took for granted, an understanding of the role of pauses in speaking is essential, for the singer's method of articulating the ideas presented in songs is derived directly from the principles employed by speakers. Hence, a reconstruction that unites the worlds of speaking and singing not only places singing in a broader cultural context but also helps us understand how singers controlled the final shaping of the music they sang. Because, as we have seen, composers of this period did not write down their ideas exactly as they intended them to be expressed, singers frequently had to create space in a vocal line for pauses.²⁷ The introduction of pauses, then, is one of the ways singers liberated both airs and recitatives from their inexpressive notation. Of course, there is no way for us today to know exactly how this highly articulated approach might have sounded, but a number of modern singers, especially, though not exclusively, some of those listed above, have adopted a comparable style of phrasing. One of them, the jazz artist Holly Cole (b. 1963), is as rhetorically expressive in this regard as any of her historical counterparts, and her recording of My foolish heart (Ned Washington, words, and Victor Young, music) presents us with an aural image of a fascinating way of singing popular/jazz music that comfortably maps on to the verbal depictions and notated examples that survive from the earlier period.²⁸ The similarities between late eighteenth- and early nineteenth-century performing practice and recent jazz and popular singing can be shown quite convincingly through this facet of the singer's craft, and although modern singers specializing in art music no longer sing in a highly articulated way, the frequent introduction of pauses was as customary in the late eighteenth and early nineteenth centuries as it is in Holly Cole's recordings. The analogous approaches to pausing taken by singers from these two widely separated historical moments demonstrates the essential nature of pauses in sung lines and provides us with a way of bridging the gap between two seemingly disparate musical cultures.

The principles of pausing treated here have been derived from the works of Joseph Robertson, John Walker, and Lindley Murray²⁹ and are based upon their discussions of grammatical and rhetorical pauses. The two ways in which orators articulated the structure of sentences were through the observance of notated punctuation (the grammatical pause) and the application of a pause in a place where, although a stop was not indicated, the sense of the sentence called for one (the rhetorical pause).³⁰ By introducing grammatical and rhetorical pauses of varying lengths (a hierarchy of commas, colons, and full stops existed), speakers were able to establish the sense of a text clearly by distinguishing, to quote James Greenwood in 1711, 'Words from Words, and Sentences from Sentences'.³¹ A graceful pause, when well timed, gave speakers an opportunity to draw in a fresh supply of breath, to reinvigorate exhausted spirits, and to awaken the attention of the listener. Through these pauses, listeners were given the opportunity of recollecting ideas that are past and anticipating those that are to come.³² In short, pauses allowed speakers to convey a clear conception of the subject matter to the hearer.³³

Then, as now, notated punctuation consisted of the following elements:³⁴ comma [,], semicolon [;], colon [:], period [.], interrogation [?], exclamation [!], and parenthesis [()]. However, even though some writers maintain that no general rule can be given for the duration of each pause, primarily because the structure of sentences and the ideas they convey are so immensely various,³⁵ many writers describe the hierarchy of stops outlined here. The comma, the shortest stop in reading, was pronounced with a little pause while the speaker counted to one (about the time of a quaver). It signified that the sentence was unfinished and momentarily suspended the sense in such a way that it gave the expectation that much more was to follow. The semicolon required a pause long enough for the speaker to count to two (about the time of a crotchet) and was used when the foregoing member of the period was perfect in sense, but the following member was so dependent on the previous one that the latter made no sense without the former. In pronouncing a colon, the speaker counted to three (about the time of a minim), and like the semicolon, the colon was used when the foregoing member was perfect in sense. However, the following member did not so immediately depend upon the former, for the colon divided the sentence into halves or distinguished it into several limbs, clauses, or lesser sentences. In fact, both the semicolon and the colon implied that the thought of the sentence was finished only partially. The period was used when the sense of the sentence was fully completed, and it was marked by a pause long enough for the orator to count to four (about the time of a semibreve). Such a pause composed the mind and enabled the listener to reflect upon the entire sentence in order to gain a full comprehension of it.

Interrogation and exclamation were related to the period, interrogation denoting a question and exclamation being used when a sentence, to paraphrase the words of John Walker, showed that the mind laboured with some strong and vehement passion. Exclamatory outbursts generally were indicated by interjections, such as O! Oh! Ah! Alas! and the like (the signs of the figure), and were delivered with far greater energy than normally was employed for expressing passion or emotion, that is, with greater force, loudness, and vehemence.³⁶ The pauses associated with both interrogations and exclamations were indeterminate in length, some writers remarking that the stop may be equivalent to a comma, semicolon, colon, or period, as the sense of the sentence demands.³⁷ One writer, however, Joseph Robertson, noted that in questions the stop should be longer than that of a period, because 'an answer is either returned or implied; and consequently a proper interval of silence is necessary'.³⁸ Parenthesis, on the other hand, enabled writers to insert some other matter into a sentence which was independent enough that its omission would not harm the sense of the rest of the sentence. It was marked by a short stop at the beginning and the end, and these pauses, especially when the words enclosed in parenthesis were spoken with a quieter and quicker voice, allowed listeners to perceive where the main thought broke off and where it resumed.

These ordinary pauses, as Gilbert Austin also calls them,³⁹ serve principally for grammatical discrimination, whereas rhetorical pauses are placed either before or after important matter in order to leave it impressed on the memory with stronger effect. By suspending the sense in an unusual manner and in an unexpected place, rhetorical pauses arrest the attention, for even though these pauses interrupt the sound, something further is expected. Austin provides us with a description of the objective:

The speaker appears full of his subject and rather to wait for the expression. He appears to take time for reflection, to exercise thought, to doubt, to resolve, to be alarmed. When he speaks after such pauses judiciously made, he seems to utter the persuasions of his mind at the moment[;] he seems to speak as nature dictates, and makes, on that account, the stronger impression.⁴⁰

Indeed, a natural manner of speaking required the frequent insertion of pauses, and Joseph Robertson tells us just how often they occurred:

An ingenious writer has observed, that not half the pauses are found in printing, which are heard in the pronunciation of a good reader or speaker; and that, if we would read or speak well, we must pause, upon an average, at every fifth or sixth word.⁴¹

The common practice was exemplified by John Walker:

Foolish men * are more apt to consider * what they have lost, * than what they possess; ** and to turn their eyes * on those * who are richer than themselves, * rather than on those * who are under greater difficulties.⁴²

In the passage, a single asterisk indicates a subordinate pause and a double asterisk denotes the longest or principal pause, that is, the one placed between the two main constructive parts of the sentence.⁴³ The stops are located after a nominative (Foolish men), and before pronouns (what, who), conjunctions (than, and), and a preposition (on), but grammarians also recognized that pauses may be introduced in other ways:

1. a. Following one word after a conjunction: 'But this * I confess * unto thee.' (Maittaire, English Grammar, 203)

2. b. Between substantives: 'Reason, virtue, answer one great aim.' (Murray, English Grammar, 161)

3. c. Before vocative to distinguish expressions in a direct address: 'Beware, Amasia, of the artful sycophant.' (Robertson, Essay, 26)

4. d. To distinguish words or phrases in apposition: 'The butterfly, child of the summer, flutters in the sun.' (Murray, English Grammar, 163)

5. e. Whenever a verb is understood: 'Solitude makes us love ourselves; conversation, others.' (Robertson, Essay, 62)

6. f. To set off parenthetical expressions or intervening clauses that may be omitted without injuring the construction: 'Very often, while we are complaining of the vanity and the evils of human life, we make that vanity, and we increase those evils.' (Murray, English Grammar, 164)

Ordinarily, speakers would never separate an adjective from its noun, but one of the greatest actors on the eighteenth-century British stage, David Garrick, commonly suspended his voice at precisely these moments. This, along with other aspects of Garrick's delivery, displeased a number of critics, and some of them wrote of their displeasure in letters sent either to the press or directly to Garrick. Several of the letters that appeared in the Craftsman were gathered together and published in 1760, and these letters detail more than fifty of Garrick's 'errors' in pausing and emphasis.⁴⁴ This collection, together with Garrick's response to a private letter written to him in 1762,⁴⁵ provides us with important information on spoken delivery, for Garrick frequently paused in unexpected places. He defended his practice of occasionally severing an adjective from its noun by differentiating between a stop, which in this instance might be considered false, and a suspension of the voice. A stop closes the sense of the text, but a short suspension of the voice helps actors naturally portray the emotional state of the character they are representing, especially when grief or horror might warrant such an interruption.⁴⁶ Garrick delivered the line from Shakespeare's Macbeth 'My thought, whose murder yet is but fantastical, shakes so my single state of man' with a suspension of the voice between the words 'single' and 'state' in order to paint the horror of Macbeth's mind.⁴⁷ Similarly, in the line from Hamlet, 'I think it was to see my mother's wedding', he suspended his voice between the words 'see' and 'my' because the grief he felt caused him to sigh before he finished the sentence.⁴⁸ Moreover, Garrick introduced this last type of articulation, that is, one between a verb and its object, on such a regular basis that even a simple sentence such as 'He would drown the stage with tears' (Hamlet), carried a pause after 'drown'.⁴⁹

Writers on punctuation and grammar, particularly John Walker, thought of these ungrammatical pauses as subordinate ones that could be introduced wherever their omission would harm either the sense or the delivery,⁵⁰ and, of course, they were exactly the types Gilbert Austin described as rhetorical, that is, pauses required by judgement and feeling.⁵¹ They figured prominently in theatrical delivery, and even one of Garrick's critics had to admit that 'stage declamation, which being impassioned, frequently to a degree of enthusiasm, does not admit of rigid correctness'.⁵²

Undoubtedly, a good knowledge of both the grammatical and the rhetorical pause was essential to eloquent spoken delivery. But it was equally important to singing, and in 1771 Anselm Bayly insisted that in speaking and singing a 'just observation of stops' establishes the sense of the text but an improper use of stops obscures the meaning.⁵³ This view was echoed repeatedly by later writers, Gilbert Austin going so far as to declare that pausing between the members of a sentence, or in any other place that will admit a momentary suspension of the voice, is that 'beautiful point of art [in which] the singers of Italy excel all others'.⁵⁴ Normally, these pauses were not written in the score, for composers relied on singers to insert them into the vocal line at the time of performance. However, one early nineteenth-century writer, William Kitchiner, implored composers not to leave such an important matter entirely to the discretion of singers and their accompanists:

If Composers would first attend to the accurate punctuation of the Words,-and then, over the several Stops-introduce Rests of defined value,-equivalent to the Stops;-it would in a great degree prevent that playing at cross-purposes which now so often occurs, to the great perplexity of both the Singer and the Accompanist.⁵⁵

In carrying our understanding of the relationship between speaking and singing one step further, William Cockin, writing in 1775, noted that in both types of delivery, pauses, like blank spaces in pictures, set off and rendered more conspicuous whatsoever they disjoined or terminated.⁵⁶ In rendering the sense of sentences 'more conspicuous', singers inserted pauses just as frequently as speakers, and Domenico Corri exemplified the typical practice in his Select Collection, a publication in which the places singers should pause (breathe) have been marked meticulously.⁵⁷ On the title page, Corri explains that the symbols he uses to indicate breathing are equivalent to the punctuation found in sentences: 'the Music in this Work is divided into phrases, as, in reading, sentences are marked by points'. But he goes much further than this in his notation, for he specifies pauses not only in the same places that speakers would introduce grammatical stops but also in those locations where rhetorical pauses would arrest the attention of listeners. Corri's edition of Michael Arne's 'This Cold flinty Heart' (Cymon) demonstrates how grammatical and rhetorical pauses were applied in practice (Ex. 2).⁵⁸

Ex. 2.

Michael Arne, 'This cold flinty heart', Cymon (D. Corri, Select Collection, ii. 30)

Corri's mixture of the grammatical and the rhetorical produces 'ordinary' stops after nominatives ('my senses * have Charm'd'; see Table 1) and preposed extensions comprising prepositional phrases ('In vain against Merit and Cymon * I strove'), as well as before prepositions ('sweet Passion * of Love') and between substantives in succession ('sweet Passion * sweet Passion * sweet Passion'). His rhetorical gestures suspend the sense in unusual and unexpected places through pauses placed between

Table 1.

Corri's pauses in 'This Cold flinty Heart'

Table 2.

Cole's pauses in 'My foolish heart'

two substantives connected by the conjunction 'and' ('Merit * and Cymon'),⁵⁹ a preposition and its object ('without * Passion'), and a verb and its object ('you waken'd * my Passions'). He also separates a transposed objective completion from the pronoun it precedes ('This Cold flinty Heart * it') and introduces pauses between elements in the figure epizeuxis ('Passion * sweet Passion * of Love? * sweet Passion * sweet Passion * sweet Passion * of Love').⁶⁰

In the score, all these stops are indicated either by rests written into the melodic line or through the two signs Corri devised for the purpose (see Ex. 2). Generally, longer pauses distinguish musical periods, to use Corri's words, and mark the principal grammatical divisions in the text. This type of pause gives listeners an opportunity to reflect on the sentiments and thoughts expressed, and Corri normally places it at the ends of periods (Ex. 2, passim), the main exception being the breath that separates the two substantives connected by the conjunction 'and' ('Merit * and Cymon'; Ex. 2, bb. 18-19). He also uses the longer pause to set apart the elements in the passage involving epizeuxis ('Passion * sweet Passion * of Love? * sweet Passion * sweet Passion * sweet Passion * of Love'; Ex. 2, bb. 22-8), and the anticipatory effect created by these lengthy pauses enables singers to draw attention to the increased vehemence (that is, the intensification of the passion concerned) that is to be placed on the reiterations. Corri employs shorter pauses to articulate other aspects of grammatical structure (for example, the separation of nominatives from verbs: 'my senses * have Charm'd'; Ex. 2, bb. 6-8) and for those situations in which the singer wishes to arrest the attention by uttering, to borrow Gilbert Austin's words, 'the persuasions of his mind at the moment' (see in Ex. 2 the pause between preposition and object in bb. 21-2: 'without * Passion').⁶¹ But, of course, arresting dramatic effects also could be created through the suppression of pauses, and Corri confounds listeners' expectations by omitting the pause after the nominative on the repetition of the phrase 'my senses have Charm'd' (Ex. 2; compare bb. 6-8 with 10-12).

Clearly, when eighteenth-century actors and singers coupled the rhetorical with the grammatical, they had at their disposal a powerful arsenal of devices to assist them in communicating the meaning of the text, and Holly Cole, like her counterparts in the eighteenth century, uses both grammatical and rhetorical pauses to establish the sense of My foolish heart for the listener. She places pauses after nominatives (night, lips; Table 2), and before prepositions (between, on, in, of), conjunctions (for, when, then, or), and a pronoun (that). But apart from these grammatical stops, Cole separates expressions in a direct address ('Beware, * my foolish heart'), pauses when the verb is understood ('How white * the ever constant moon'), and introduces rhetorical pauses in the second verse to draw attention to certain ideas in the text. Specifically, she suspends her voice between a transposed adverbial phrase and the pronoun it precedes ('this time * it isn't'), separates a verb from its complement ('it isn't * fascination'), and interrupts the vocal line after the words 'But should' and 'then let' in order to allow listeners a moment to anticipate the ideas that follow. This latter manner of delivery is similar to Michael Maittaire's practice of joining conjunctions to pronouns and pausing after the pronoun in clauses beginning 'but this *', 'and they *', and 'because that *'.⁶² Like Maittaire, Cole pauses one word after a conjunction, but in My foolish heart the stop follows an auxiliary verb.

When inserting these articulations into the vocal line, Cole creates what eighteenth-century listeners would have considered a natural manner of delivery. She not only stops frequently (that is, after every second, third, fourth, or fifth word) but also varies the length of her pauses to generate variety, the final shape of the music being a product of her taste and judgement. A transcription of her performance is shown in the upper stave of Ex. 3, and the lower stave presents a reading of the song similar to the type found in many published anthologies.⁶³ Most publishers, in the eighteenth and nineteenth centuries as well as now, never try to notate songs the way singers actually sing them (D. Corri's Select Collection being a rare exception), for their editions simply record skeletal versions that are not meant to be reproduced mechanically in performance. As has already been noted, Domenico Corri observed over two hundred years ago that music sung exactly as it is written would be 'a very inexpressive, nay, a very uncouth performance',⁶⁴ and few singers in the modern world of jazz would perform My foolish heart as it is represented in the lower stave of Ex. 3. Nonetheless, this rather 'uncouth' rendering of the song is useful as a 'standard' version to which Cole's phrasing may be compared.

Cole's pauses range from those as short as a semiquaver to those slightly longer than a minim, and she usually creates space for them by abbreviating the note preceding the pause and delaying the note or notes following it. This approach, similar to procedures employed in the eighteenth century,⁶⁵ allows Cole to approximate spoken delivery, for instead of words being unnaturally elongated, as often occurs in other styles of singing, especially the modern operatic style, they are more speech-like in duration. In establishing the sense of the text for her listeners, Cole provides large spaces for the comprehension of the ideas expressed in the song. The pauses applied to the opening line of the first verse ('The night * is like a lovely tune. *'; Ex. 3, bb. 1-2) and the first two lines of the second verse ('His lips * are much too close to mine. * Take care, * my foolish heart. *'; Ex. 3, bb. 16-20) allow listeners time either to anticipate ideas to come or recollect those which are past. Her shorter pauses, normally a quaver or dotted quaver in length, often precede prepositions and adjectival phrases ('on an evening * such as this' and 'when you're lost * in the magic * of a kiss'; Ex. 3, bb. 11-12 and 14-16) or separate adverbial phrases from the pronouns they precede and verbs from their complements ('for this time * it isn't * fascination'; Ex. 3, bb. 24-6). Cole even introduces a short suspension of the voice reminiscent of the type employed by Garrick to portray emotional states realistically. In the third line of the song ('How white * the ever constant moon'; Ex. 3, bb. 4-6), she interrupts the flow of the thought momentarily between 'ever' and 'constant', and this unexpected hesitation has the effect of setting off and rendering more conspicuous, to use William Cockin's words (see above), the final sentiment of the line, 'constant moon'. She also confounds the listener's expectations in the second verse

Ex. 3.

Ned Washington/Victor Young, My foolish heart.

© 1949 (Renewed) Chappell & Co., Largo Music Inc. and Patti Washington Music. All Rights Reserved. Used by Permission. Warner Bros. Publications U.S. Inc., Miami, FL 33014

through her suppression of the pause before the conjunction 'for' in the line 'then let the fire start, for this time it isn't fascination' (Ex. 3, bb. 22-6). This approach finds historical precedent not only in the work of Domenico Corri (see above) but also in the treatises of Manuel García, who writes of increasing the effect of a musical phrase by suppressing the pause and joining the two members together through a crescendo (Ex. 4).⁶⁶ This is precisely the way Cole delivers the passage, and had she adopted the normal approach to pausing at this point, she would have lost the opportunity of augmenting the emotional impact of the text in a dramatic way.

Ex. 4.

Suppressing the pause; Donizetti, Anna Bolena (García, New Treatise, 48)

Without a doubt, Cole's highly articulated manner of delivery parallels earlier practices, for her intuitive instincts seem to be identical to those of singers two hundred years ago. It is almost as if a strong sense of grammatical correctness, combined with an impassioned enthusiasm for the text, has compelled her to introduce pauses of flexible length in the same places that eighteenth- and early nineteenth-century speakers and singers would have done. Cole is, of course, part of a rich tradition of jazz singing in the second half of the twentieth century, and several other jazz singers, particularly Ernestine Anderson, Gabrielle Goodman, Abbey Lincoln, Carmen McRae, Dinah Washington, and Monica Zetterlund, along with the popular singers listed earlier, employ a similar manner of phrasing on some of their recordings.⁶⁷

But as noted at the outset, in the world of art music, this style of pausing began to disappear in the early twentieth century. The last vestiges of it appear on the recordings of opera singers such as Adelina Patti (1843-1919), who continued to sing in the older ways at the dawn of the recording era.⁶⁸ Curiously, no one has studied why traditional phrasing ceased to form part of a singer's training in art music when it continued to play an important role in the practices of jazz and popular artists. Perhaps, as Robert Philip has noted, taste changed so much during the first half of the twentieth century that some of the traditional elements of expression, including highly articulated phrasing, melodic rubato, portamento, and flexibility of tempo, became distasteful to practitioners of art music, even to those musicians in the late twentieth century who advocate historically informed performance.⁶⁹ Nonetheless, as we have seen, the manner of pausing discussed here figures prominently in the depictions of performing styles found in late eighteenth- and early nineteenth-century treatises, as well as on the recordings of a number of jazz and popular artists,⁷⁰ and although the discussion has concentrated on just one facet of the singer's craft,⁷¹ it does reinforce a suggestion made by Richard Middleton more than a decade ago that 'the differences setting apart twentieth-century popular songs from the lineages of European music are less than commonly thought'.⁷²