Singing in French: between theory and practice.
Developing an accurate perception and understanding of any language requires at first a command of the rules that regulate its structure. French is no exception, and the extent of rules that await average English-speaking undergraduate voice majors on their first semester of French diction is there to prove it. However, it is no secret that there will always be room for arguments when it comes to French diction, and it is not atypical to witness two French-speaking singers or coaches debate issues such as liaisons or the proper pronunciation of a particular sound. This article first will summarize how linguists have understood and explained liaisons and vocalic harmonization, and then will survey the performance of French-speaking singers on recordings and compare their interpretations. It will analyze how well the outcome matches up with the rules, and if there is a method behind French singers' choices.
A LANGUAGE IN CONSTANT EVOLUTION
Evolution of liaisons
The history of modern liaisons looks back to the thirteenth, fourteenth, and fifteenth centuries, where final consonants were progressively dropped in popular speech. Testimonies from contemporary grammarians corroborate this fact: "About everywhere else, finale consonants were silenced at the beginning of the sixteenth century, and one pronounced 'ave,' 'soy,' 'fi,' 'mo' (but 'beaucoup') according to Palsgrave (p. 24) ; 'cle' (clef),'ro' (rost),'beaucou' (beaucoup), according to H. Etienne, at the end of the century." (1) To counteract the phenomenon and preserve the voicing of consonants, the grammarians agree that in sustained speech, "they are pronounced in liaisons before a vowel, as well as before a pause, but dropped before consonant." (2 )Despite their effort, throughout the seventeenth and eighteenth centuries, the majority of final consonants became silent in front of vowels. It was inappropriate then to execute liaisons, even between closely related words. "In the seventeenth century, the most educated persons spoke currently without liaisons, after the testimony of the best grammarians cited in Thurot (1883): vene(z) ici, je sui(s) assez bien, voyon(s) un peu." (3) It all changed in the nineteenth century. The use of liaisons became a means to show off the ability to spell, and the number of liaisons increased as the century unfolded. It hit its peak within the upper and aristocratic classes in the last quarter of the century. Littre, in his preface to the Dictionnaire de la langue francaise (1866) witnesses a phenomenon in evolution. Of course, as always when a grammarian comments on a linguistic trait in transition, his view is conservative, and for him a higher level of speech does not necessarily imply more liaisons.
There is still a point where our pronunciation tends to branch out from our fathers and our ancestors, I mean the people of the eighteenth and seventeenth centuries: it is the liaison of consonants. Formerly one would carry out a much lesser amount of liaisons; everyone remembers hearing older people pronounce not the Eta-z-Unis as we do it, but the Eta-Unis [etayni]. To this trend, I have nothing to object, but that it must be restricted to the principle of the tradition, which, in speech, does not extend liaison beyond a certain number of cases settled by practice that, in declamation, eradicate liaisons in all the cases where they would be harsh or offensive. One must conform to this saying of abbe d'Olivet [Pierre-Joseph Thoulier d'Olivet, dit l'abbe d'Olivet 1682-1768 ]: "the conversation of the honest citizens is filled with voluntary hiatus that are so authorized by usage, that if one would speak differently, it would be pedantic or provincial." (4)
Born the same year Littre wrote those lines, Maurice Grammont comments in 1914 on the decline of liaisons that has already taken place in speech. His insight only confirms how much the perception has changed within two generations, although both linguists share one opinion: they disapprove of the overuse of liaison.
The general rule is quite simple: one makes all liaison within a groupe rythmique, one does not make liaison from a groupe rythmique to a following one. In other words, one observes liaisons from an unaccented syllable into the next syllable; one does not observe liaisons after an accented syllable. One will say:
Ces petits enfants / ont perdu leur chemin; but ces petits / enfants is not French, and ces petits enfants ont perdu is even worse. This rule was still regularly observed in conversation thirty years ago [in 1884]. It is nowadays [in 1914] still almost absolute in reading and oratory speech; but in casual conversation, one observes fewer and fewer liaisons. The new generations tend to keep only the liaisons that hold a grammatical function, particularly those that make the distinction between singular and plural. One would not be able to fix a set of rules ... above all because the phenomenon is in thorough evolution. One can say that liaisons between groupe rythmique upset almost right away and reveal a person either conceited, lacking education, or provincial, whereas liaisons within a groupe rythmique never disturb, even when they are not necessary. (5)
Philippe Martinon (1913) is uncompromising and goes as far as harshly criticizing the practice of the revered Comedie-Francaise: "But the actors especially overexert strangely [liaisons], either because of hypercorrection, or to make themselves better understood, and that at the Comedie-Francaise as well as other places, more than other places, alas! ... And foreigners go to the Comedie-Francaise to learn to pronounce! I agree, except for the subject of liaisons." (6) Fernand Carton (1974) comments on what has become of liaisons: "the reduction to the minimum of liaisons is characteristic of popular and familiar speech. Intonation and 'accent d'insistance,' demarcating gestures, favor the progressive elimination of optional liaisons in contemporary French." (7) Carton, whose approach remains objective, uses a simple phrase to demonstrate a hierarchy of different levels of language among liaisons. The sentence
Des hommes importants ont attendus. offers the possibility for four liaisons: the first one between an article and its noun. Observing solely that liaison is appropriate in a conversation between friends or family:
Des hommes / importants / ont / attendus. The second liaison in rank is the one within the verbal group, and corresponds to a conversation between colleagues:
Des hommes / importants / ont attendus. The third liaison in rank, between the plural noun and its following adjective, would not normally be made unless the speaker has the opportunity to see the words written out. Such a liaison is appropriate for a public speech or national television:
Des hommes importants / ont attendus. The last liaison left, between importants and ont, is inappropriate in modern versification. It is only maintained in classical versification as an indirect liaison, for an adjective cannot be linked to a verb. (8)
Pierre Encreve (1988) codifies a new tendency in liaison: indirect liaisons. They are liaisons wherein the final consonant of a word is pronounced at the end of the word and separated from the following vowel by a glottal stop.
Indirect liaison is today an uncontested linguistic fact. Our survey allows us to think that the phenomenon has become a grammatical fact during the seventies, at least among the professional of public speech, which has allowed all the French to familiarize themselves unconsciously with these pronunciations. Today, indirect liaisons appear in full growth among the majority of highly educated orators. (9)
Alain Thomas (1998) acknowledges the same phenomenon, in addition to a newer development.
A number of new pronunciations, like the sounding of the final consonant of the words but, fait, cout, and quand, even in front of [a word beginning with a] consonant, within the political elite, suggest that the writing holds a non-negligible influence over the pronunciation of final consonants. (10)
It makes sense that the disappearance of liaisons in speech throughout the twentieth century on one side has triggered a reverse behavior in sustained speech on the other. Sounding final consonants that are unnecessary was qualified as being "pedantic" by d'Olivet (1737) and Littre (1866), and as "conceited" by Grammont (1914). At the dawn of the twenty-first century, the sounding of a highest proportion of final consonants in liaisons or even in front of a word beginning with another consonant has become a trademark of political leaders and the intellectual elite. It has lost its negative connotation. It is now perceived as a prerequisite for scholarship and leadership.
Groupe Rythmique or Phonetic Word
Leslie De'Ath already explained the notion of rhythmic group in his article from the November/December 2007 Journal of Singing, "Double Consonants and Gemination in Sung French." (11) It is recognized and introduced in the earliest treatise on modern French pronunciation. Grammont borrowed the term from Edouard Bourciez (1889). Pierre Fouche (1959) labels it accent de groupe. Fouche quotes Leonce Roudet's Elements de phonetique generale (1910) as part of his discussion on the subject. What characterizes the French Language is not its word accentuation, but its group accentuation. The French sentence lets itself cut out into a succession of groups that show very little respect to the individuality of the word, and "which each group creates an impression of phonetic unity, to the point that those who ignore the language believe a group to be a single word." (12)
Fernand Carton (1974) uses the term mot phonetique (phonetic word) together with groupe accentuel. He relates the phonetic word to a syntagma, a unity of syntax, that is, in a given sentence, the subject, the verb, and the objects, with their relevant determiners.
Jean-Claude Milner and Francois Regnault in their essay "Dire le vers" (1987) elaborate a definition that corresponds best to the features of the phonetic word in lyric diction." It will rely on the concurrence of three unifying criteria: in general, a phonological [phonetic] word is what constitutes at once a flow unit, a grammatical unit, significance unit." (13) The following description respects the guidelines established in Milner and Regnault's essay and recognizes four categories of phonetic word.
1. Nominative group
Noun, its determiners (articles, adjectives placed before), its complements
beaux yeux aux flammes douces ("Puisque l'aube grandit," P. Verlaine)
des couples blonds d'amants heureux ("L'ile heureuse," E. Mikhael)
tes yeux leves au ciel ("Soir," A. Samain)
2. Verbal group
Verb, its determiners (auxiliaries, pronouns, adverbs, negation, etc.), its closely linked complements (predicates, direct and indirect objects, straightforward objects of the preposition)
j'ai longtemps habite ("La vie anterieure," C. Baudelaire)
je chanterai des airs ingenus ("Puisque l'aube grandit," P. Verlaine)
et porte au coeur un humide eclair ("Le colibri," Leconte de Lisle)
3. Adjective group
Adjective, its determiners and objects
douloureusement incertain ("J'allais par des chemins perfides," P. Verlaine)
beau comme une sombre rose ("Cantique a l'epouse," A. Jounet)
isoles dans l'amour ainsi qu'en un bois noir ("N'estce pas?" P. Verlaine)
D. Prepositional group
Preposition, its linked nominative group
dans le golfe aux jardins ombreux ("L'ile heureuse," E. Mikhael)
Pres des remparts de Seville (Carmen, Meilhac and Halevy)
vers les vastes cieux enchantes ("Le jet d'eau," C. Baudelaire)
A phonetic word can also correspond to a single word.
For example, in Hebe s'eloigne ("Hebe," Ackermann),
Hebe is a nominative group and s'eloigne is a verbal group. In vont une a une ("Les heures, "Mauclair), vont is a verbal group, une a une an adjective group (it is neither a direct nor an indirect object to the verb). One grammatical unit may hold two significant units, as does le temps des lilas / et le temps des roses. They are treated as two phonetic words. In the examples cited above, every liaison must be maintained.
beaux yeux aux flammes douces [bo zjo zo flame\ duse\] j'ai longtemps habite [ze lota zabite] douloureusement incertain [dulurozema teserte] dans le golfe aux jardins ombreux [da le g[??]l fo zarde zobro]
Understanding the notion of phonetic words will benefit a better insight on liaisons.
Theories on trial
Three melodies and three cycles were reviewed: "Villanelle" and "Sur les lagunes" (Berlioz / Gautier); "Automne" (Faure / Sylvestre); L'horizon chimerique (Faure / de Mirmont); Chansons de Bilitis (Debussy / Louys); Chansons madecasses (Ravel / Parny). Their texts are set below, although to save space the poetic lines that did not display a liaison were omitted. In addition, to avoid any confusion between liaisons and elisions, the latter, since they are not part of this study, are left out altogether. In cases where all singers agree, the symbol indicates liaisons and the symbol | identifies hiatus. Bold characters highlight instances where the outcomes are conflicted. (See Appendix for survey results.)
"Villanelle" // Quand h auront disparu les froids, Tous les deux nous irons, ma belle, Pour cueillir le muguet | aux bois. // Sous nos pieds egrenant les perles Que l'on voit, au matin trembler, Nous irons ecouter les merles // Le printemps | est venu, ma belle; C'est le mois des amants beni; Et l'oiseau, satinant son aile, Dit ses vers au rebord du nid. // Pour parler de nos beaux amours, // Loin, bien loin | egarant nos courses, Et le daim, | au miroir des sources | Admirant son grand bois penche; // Puis chez nous, tout heureux, tout aises, En paniers, enlacant nos doigts, "Sur les lagunes" // Mon ame et mes amours. // Que mon sort est amer! Ah! Sans amour s'en aller sur la mer! // Sur moi la nuit immense S'etend comme un linceul, "Automne" Automne au ciel brumeux, | aux horizons navrants. Aux rapides couchants, | aux aurores palies, // Sur l'aile des regrets mes esprits emportes, // Parcourent, hen revant, les coteaux enchantes, | OU jadis sourit ma jeunesse! Je sens, au clair soleil du souvenir vainqueur, Refleurir en bouquet les roses deliees, Et monter a mes yeux des larmes, qu'en mon coeur, Mes vingt ans avaient oubliees! L'horizon chimerique La Mer est infinie et mes reves sont fous. //De danser sur la mer comme des oiseaux souls. Le vaste mouvement des vagues les emporte, La brise les agite et les roule en ses plis ; Jouant dans le sillage. Ils feront une escorte *** Je me suis embarque sur un vaisseau qui danse //Mes pieds ont oublie la terre et ses chemins ; Les vagues souples m'ont appris d'autres cadences Plus belles que le rythme las des chants humains. A vivre parmi vous, | helas ! avais-je une ame ? //Pour me bercer, comme un h enfant, | au creux des lames. //Les larmes du depart ne brulent plus mes yeux. Je ne me souviens pas de mes derniers adieux ... O ma peine, ma peine, oU vous ai-je laissee ? *** Injurieuse au trouble vain des pauvres ames, Et mon coeur, toujours las | et toujours agite, *** Vaisseaux, nous vous aurons aimes en pure perte ; Le couchant | emporta tant de voiles ouvertes Que ce port et mon coeur sont a jamais deserts. La mer vous a rendus a votre destinee, // Nous ne pouvions garder vos ames enchainees ; // Car j'ai de grands departs inassouvis en moi. Trois chansons de Bilitis Pour le jour des Hyacinthies, // Unis avec la blanche cire // Il m'apprend a jouer, | assise sur ses genoux; // Mais je suis un peu tremblante. // Que je l'entends a peine. // Nous n'avons rien a nous dire, Que je suis restee si longtemps | A chercher ma ceinture perdue. *** // Je les caressais | et c'etaient les miens; Et nous h etions lies pour toujours ainsi, // Tant nos members etaient confondus, // "...Ou que tu entrais en moi comme mon songe." // Quand il eut acheve, Il mit doucement ses mains sur mes epaules, // Que je baissai les yeux | avec un frisson. *** // Et mes sandals etaient lourdes // Ses petits pas fourchus alternent Comme des trous dans un manteau blanc. // "Les satyres et les nymphes aussi. Depuis trente ans, il n'a pas fait un hiver aussi terrible. La trace que tu vois | est celle d'un bouc. Mais restons ici, oU est leur tombeau." // Il regardait au travers. Chansons madecasses // Le lit de feuilles est prepare; je l'ai parseme de fleurs et d'herbes odoriferantes; il est digne de tes charmes. // J'entends le froissement de la pagne qui l'enveloppe; c'est elle, c'est Nahandove, la belle Nahandove! // Reprends haleine, ma jeune amie; repose-toi sur mes genoux. Que ton regard est enchanteur! Que le mouvement de ton sein | est vif et delicieux sous la main qui le presse! // Meurt-hon de volupte, Nahandove, o belle Nahandove? // Le plaisir passe comme un eclair. Ta douce haleine s'affaiblit, tes yeux humides se referment. Tu pars, et je vais languir dans les regrets et les desirs. *** Les Blancs promirent, | et cependant ils faisaient des retranchements. // Leurs pretres voulurent nous donner un Dieu que nous ne connaissons pas, // ils parlerent enfin d'obeissance et d'esclavage. Plutot la mort. Le carnage fut long et terrible; mais malgre la foudre qu'ils vormissaient, et qui ecrasait des armies entieres, ils furent tous extermines. // Nous avons vu de nouveaux tyrans, plus forts et plus nombreux, // il a fait tomber sur eux les pluies, les tempetes | et les vents h empoisonnes. *** Il est doux de se coucher, durant la chaleur, sous un arbre touffu, et d'attendre que le vent du soir amene la fraicheur. // Femmes, approchez. Tandis que je me repose ici sous un arbre touffu, occupez mon oreille par vos accents prolonges. // Elle chasse les h oiseaux avides. Le chant plait a mon ame. // Que vos pas soient lents; qu'ils imitent les attitudes du plaisir et l'abandon de la volupte. Le vent du soir se leve; la lune commence a briller au travers des arbres de la montagne.
French singers are remarkably consistent in observing all the obligatory liaisons within phonetic words. There are only two exceptions (briller au travers des arbres de la montagne, de grands departs inassouvis), but in those two cases all the singers who do not maintain the liaison take a breath. When all singers concur to leave out a liaison (on fifteen occasions), it always occurs between two words without any grammatical connection between them. Unsurprisingly, some inconsistencies stand out, which will require a closer look.
A first example occurs in the last stanza of "Automne," Je sens au clair soleil du souvenir vainqueur / Refleurir en bouquet les roses deliees. There, the object of the preposition au clair soleil du souvenir vainqueur relates to the verb refleurir and not sens. Both words clearly belong to two distinct phonetic words. That did not stop five singers out of the thirteen surveyed to maintain the liaison. The next examples are taken from Berlioz'Nuits d'ete. What makes them particularly relevant is that they display three contradictory handlings of the same letter t in liaison between two distinct phonetic words. There is no possible link between voit and au in the line que l'on voit au matin trembler. Yet nine singers out of eleven let the t of voit sound. Then, in the line Sur moi la nuit immense, immense is the determiner to the noun nuit. A liaison between a noun and its adjective is permissible only in the singular when the adjective precedes the noun. Four singers chose to maintain it. Finally, all eleven singers omit the liaison between muguet and bois in nous irons, ma belle, pour cueillir le muguet au bois. Pierre Bernac indicates all three examples as forbidden liaisons in his book The Interpretation of French Song. French singers would not erratically ignore rules that are inherent to the basic grammatical structure of their own language. It is clear that a third factor comes into play and prevails over the rule.
Georges Le Roy labeled consonants that bear a grammatical function in liaison," flexional consonants." (14) Few linguists have elaborated on the subject because such categorization makes a difference for the treatment of liaison only in versification. Milner and Regnault enumerate them: "there are liaisons that maintain a consonant with grammatical function: the s or the x of the plural, the s [or z] of the second person, the t of the third person, the r of the infinitive." (15) Going back to the previous examples, the liaison in Je sens au clair soleil links the s of the first person, and the t of voit in voit au matin indicates the third person. The liaison will sound natural to the French ear; however, omitting it would enlighten the structure of the text and help the listener understand au clair soleil and au matin as inversions right away. Both options are defensible, and the result will reflect the singer's attentiveness to the grammar. (16) With the exception of the /z/ of the plural, all flexional consonants entail a verb form. Consequently, the t of muguet, in muguet au bois, cannot qualify as a flexional consonant. Besides, au bois is the object to the verb irons, which is the first person of the plural. In nuit h immense, nuit is also a noun, but this time nuit is subject, and its t matches with its third person. One only has to replace nuitwith sourcil and try to link sourcil h immense to realize the preeminence of the flexional consonant. Et cependant ils faisaient des retranchements, from Chansons madecasses, also displays t as a flexional consonant. Six singers chose to ignore it and follow the rule, but seven of them observed the liaison.
The r of the infinitive has a lesser influence than t on liaisons. All singers maintain the liaison when it occurs within a phonetic word (donner un Dieu, briller au travers des arbres, monter h a mes yeux). (17) They all omit the liaison in Il m'apprend a jouer / assise sur ses genoux, its only occurrence between two phonetic words.
In contrast, the s or x of the plural (/z/) is the strongest and foremost among the flexional consonants, and the only one that can outweigh grammar and even logic. In "Villanelle," two singers pronounce dit ses vers au rebord du nid and en panniers enlacant nos doigts, and three singers pronounce sous nos pieds egrenant les perles, while in Chansons madecasses two singers pronounce Le lit de feuilles est prepare. These liaisons are forbidden. One cannot link a plural noun to an unrelated verb or preposition. On the other hand, a liaison connecting a plural noun to its corresponding verb exemplifies the higher level of speech required for singing. Such liaisons took place four times (mes pieds ont oublies, nos members etaient confondus, mes sandals etaient lourdes, nos vingt ans avaient oublies). It is the only case where all singers agree to maintain a liaison between two different phonetic words.
Three examples from L'horizon chimerique display one of the only incidences in the songs surveyed where French singers are not only in disagreement with each other, but inconsistent with themselves. They are nous vous aurons aimes en pure perte, vous a rendus a votre destinee, and j'ai de grands departs inassouvis en moi. The first two consist of a past participle followed by a preposition and could arguably be accepted as one extended phonetic word. Only six singers out of thirteen maintain the liaison in both cases. In the third example, en moi is a phonetic word on its own. It is an object to the preposition to the verb ai, and has no grammatical ties with inassouvis. Of the three examples, it presents the weakest case for liaison and yet, nine singers maintain that liaison, among them three had omitted the liaison in both of the two previous examples.
The line ses petits pas fourchus alternent comme des trous dans un manteau blanc from Louys' "Le tombeau des naiades," set by Debussy, stands apart.Here the words are a quotation spoken by an ingenue. Inexperienced and unschooled, she would not know of maintaining a liaison between fourchus and alternent, above all in casual speech. In this case, maintaining the liaison brings the language to a level that contradicts the character of the song.Words are rarely spoken by popular characters in melodies, but when it happens, the level of speech needs to be lowered a rank or two, therefore affecting the options for liaisons and putting restraints to the voicing of the plural. Only six women out of twelve leave out this liaison. (18)
Consonants that do not bear a grammatical function, /k/ and /p/, meet with little tolerance when it comes to liaison. Their occurrence is scarce in poetry, since their voicing sounds unnatural and awkward to a French ear. The line le carnage fut long et terrible from the second Chansons madecasses displays one example of compulsory liaison with the sound /k/. Only one singer makes a liaison, yet inaccurately with a /t/.
Rejection of vocalic harmonization
Vocalic harmonization is a relatively recent development of French speech. Linguists make no mention of the phenomenon before the twentieth century. Maurice Grammont (1914) introduced the term in his eminent essay La prononciation francaise. (19) Throughout his treaty, Fouche (1959) always specifies, when explaining a case of vocalic harmonization, that it correspond to colloquial speech and does not affect sustained diction. His perception reflects the general agreement among linguists. Fernand Carton (1974) summarizes, "Vocalic harmonization in French, which only play in open syllables, is related to what is occurring in Turk or in Finnish, although communication is not affected: it is only a tendency, which constitutes a criteria for the level of language." (20) Table 1 compiles the pronunciation of key words as recorded by French-speaking singers and sheds some light on the usage of vocalic harmonization in singing. Column 1 lists the pronunciation of the determiners des, as well as les, mes, or ces exclusively when preceding baisers; column 2 the pronunciation for the word baiser. Column 3 lists the pronunciation of apaiser (inapaiser), and column 4 the pronunciation of words of similar vowel structures to apaiser that are less recurrent (connaissez, delaisser, eclairer, caresser, laisser, j'ai reve, sceller, enchainer, maltraiter). Column 5 lists the pronunciation of eveiller (reveiller), egaye, or ensoleille, where a 'j' glide follows the unaccented 'e' vowel of the middle syllable. Column 6 lists the pronunciation of /oe/ when followed by /[empty set]]/ in heureux (malheureux), and column 7 the pronunciation of the schwa when followed by /[[empty set]/ in cheveux. The excerpts considered displayed the selected words set in middle range, no higher than [A.sub.5] for women's voices and [C.sub.4] for men's voices, to avoid any inconsistency due to vocal manipulation. For example, the opening line of Duparc's "Lamento" was surveyed in men's voice only, and the same line set by Berlioz in Nuits d'ete was ignored altogether because of its range. As a result, it was not always possible to find an occurrence for each word for all thirty singers surveyed.
Baiser [beze] becomes [beze] for 19 of the 24 singers surveyed (79%). It is clear that for words that are archetypal in poetry like aimer and baiser the use of vocalic harmonization causes no problem. Their pronunciation with /e/ is time-honored even in proper levels of speech. However, contrary to the explanation in diction books, French singers do not endorse the vocalic harmonization of les and other determiners in the plural. Out of the 24 singers, 22 (92%) maintain /e/ and pronounce [de], [te], [me], or [se]. Apaiser reveals a weaker result than baiser: nine singers maintain vocalic harmonization (38%), and 15 singers choose not to. That word, although not as common as baiser, is a recurring word in poetry, and again the pronunciation [apeze] has earned a certain tolerance. It is very instructive to notice that the proportion of singers that chose to maintain vocalic harmonization decreases as we move closer to the present time and that the youngest generation of singers is more inclined to restore a proper pronunciation.
The treatment of words such as connaissez, with an unaccented e vowel spelled ai followed by an accented [e], is less flexible: 26 out of 29 singers (90%) maintain [e] for ai and therefore ignore vocalic harmonization.
A similar proportion (88%) pronounces [eje] in eveiller, which again confirms French singers' reserve to import vocalic harmonization into poetry. The fact that heureux is pronounced [oero] by only 8 out of 26 of them (31%) is not a contradiction. The noun from which the adjective heureux is derived, heur [oer] (luck), has become an archaism in modern French and no longer holds a meaning on its own: [oer] will be understood as heure (hour) or heurt (shock). Consequently, [oro], like [beze], finds acceptance in all levels of language. Cheveux [favo] creates a stronger consensus (96%). The French muted 'e'cannot be treated as an unstressed vowel. When maintained in lyric diction, its sound is unaffected by the surrounding vowels." When it [the muted e] falls, it is not diminished and reduced, but suppressed completely; when it remains, its pronunciation is as full as any of the unaccented vowels." (21)
French singers do not maintain vocalic harmonization in singing. This is especially true of French melodie, aristocratic art among all. In opera, vocalic harmonization will be relevant to typify an ingenue or any character from the popular and middle classes. The aristocratic characters would then stand out more efficiently.
Are French singers consistent with the rules? This study suggests the affirmative. They demonstrate a surprisingly strong consensus regarding liaisons. (22) The divergences on the matter result mainly in their choice to give flexional consonance precedence over the grammatical syntax and vice versa. It is impossible to tell, however, if their choices are led by their instinct, or if they made their decisions being fully aware of the grammatical structure of the text. It becomes obvious that for the English-speaking student, the learning and memorization of sets of rules alone are deficient, and the thorough understanding and recognition of the grammatical syntax in French is essential to achieve a competence in carrying out liaisons and elisions.
With vocalic harmonization, the outcome is even more compelling. If one considers the results only for the singers born within the past sixty years, the tendency to ignore vocalic harmonization in singing becomes the norm. All the singers pronounce [de] or [le], [apeze], [kcnese], [eveje], [fevo], and only 43% pronounce baiser [beze] while 33% pronounce heureux [oro]. For the English-speaking student, the difficulty lies mainly in the ability to recognize vocalic harmonization in the dictionary, where the pronunciation is given according to the way in which it is most commonly spoken and realizing that it does not correspond to the appropriate pronunciation in singing.
APPENDIX 1. "Villanelle" and "Sur les lagunes." pieds Singer egrenant voit au vers au 1910 Suzanne Danco no yes yes Gerard Souzay yes yes yes 1920 Regine Crespin no yes no 1930 Bruno Laplante no yes no 1940 Jose Van Dam no yes no Colette Alliot-Lugaz yes yes no Francoise Pollet yes yes no 1950 Brigitte Balleys no no no 1960 Isabelle Vernet no no no Veronique Gens no yes no Elsa Maurus no yes no paniers nuit Singer en immense 1910 Suzanne Danco no no Gerard Souzay yes yes 1920 Regine Crespin no yes 1930 Bruno Laplante no yes 1940 Jose Van Dam no no Colette Alliot-Lugaz yes yes Francoise Pollet no no 1950 Brigitte Balleys no no 1960 Isabelle Vernet no no Veronique Gens no no Elsa Maurus no no APPENDIX 2. "Automne." Singer sens au ans avaient 1880 Ninon Vallin yes no (br) 1890 Charles Panzera yes no (br) 1900 Hugues Cuenod yes no 1910 Camille Maurane yes yes Irma Kolassi no yes 1920 Berthe Montmart no no (br) Gabriel Bacquier yes no (br) 1930 Bernard Kruysen no no (br) Jacques Herbillon no br no (br) Rachel Yakar no no (br) 1940 Michel Piquemal no no (br) 1950 Francois Le Roux no yes 1960 Nathalie Stutzmann no no (br) APPENDIX 3. L'horizon chimerique. toujours Singer agite aimes en 1890 Charles Panzera yes yes 1900 Hugues Cuenod yes no Martial Singher yes yes Paul Derenne no 1910 Camille Maurane yes yes Gerard Souzay yes/no yes 1920 Pierre Mollet yes yes 1930 Bernard Kruysen yes no Jacques Herbillon yes yes 1940 Michel Piquemal yes no 1950 Francois Le Roux no yes 1960 Laurent Naouri no yes Thierry Felix no yes inassouvis Singer rendus a en 1890 Charles Panzera yes yes 1900 Hugues Cuenod yes no Martial Singher yes no (br) Paul Derenne no yes 1910 Camille Maurane no yes Gerard Souzay yes yes 1920 Pierre Mollet yes br 1930 Bernard Kruysen no yes Jacques Herbillon yes yes 1940 Michel Piquemal no yes 1950 Francois Le Roux no no 1960 Laurent Naouri no yes Thierry Felix yes yes APPENDIX 4. Trois chansons de Bilitis. l'entends toujours fourchus Singer a ainsi alternent 1870 Jane Bathori yes yes yes 1900 Flore Wend yes yes no 1910 Suzanne Danco yes yes yes Irene Joachim yes yes yes Irma Kolassi no yes no 1920 Regine Crespin yes yes no 1930 Ettel Sussman yes no yes Irene Jarsky yes yes no 1940 Danielle Galland yes no no Michele Command yes no yes 1960 Nathalie Stutzmann yes yes no Veronique Gens yes no yes satyres regardait Singer et au 1870 Jane Bathori yes yes 1900 Flore Wend yes yes 1910 Suzanne Danco yes yes Irene Joachim yes yes Irma Kolassi yes yes 1920 Regine Crespin yes yes 1930 Ettel Sussman yes yes Irene Jarsky no yes 1940 Danielle Galland yes no Michele Command yes yes 1960 Nathalie Stutzmann yes yes Veronique Gens yes yes APPENDIX 5. Chansons madecasses. feuilles fleurs regrets Singer est et et 1890 Madeleine Grey no no yes 1900 Martial Singher no no yes 1910 Gerard Souzay no yes yes Irma Kolassi no no yes 1920 Pierre Mollet no no no Jean-Christophe Benoit yes no yes 1930 Bernard Kruysen no yes yes Jacques Herbillon yes no yes 1940 Michel Piquemal no no yes 1950 Didier Henry no no yes Francois Le Roux no yes yes 1970 Claire Brua no yes yes Isabelle Cals no yes yes cependant Singer ils long et forts et 1890 Madeleine Grey no no no 1900 Martial Singher no no yes 1910 Gerard Souzay yes no yes Irma Kolassi yes no no 1920 Pierre Mollet no no no Jean-Christophe Benoit no no yes 1930 Bernard Kruysen no no yes Jacques Herbillon yes no yes 1940 Michel Piquemal yes no yes 1950 Didier Henry yes yes (t) no Francois Le Roux yes no yes 1970 Claire Brua no no yes Isabelle Cals yes no yes
(1.) Christine Marchello-Nizia, La langue francaise aux XIVe et XVe siecles (Paris: Armand Collin, 2005), 109.
(2.) Mireille Huchon, Le francais de la renaissance, 2nd ed. (Paris: Presse universitaire de France, 1998), 91.
(3.) Philippe Martinon, Comment on prononce le francais (Paris: Larousse, 1913), 356.
(4.) Cited in Pierre Encreve, La liaison avec et sans enchainement (Paris: Seuil, 1988), 284.
(5.) Maurice Grammont, La prononciation francaise (Paris: Librairie Delagrave, 1914), 130.
(6.) Martinon, 357, 385.
(7.) Fernand Carton, Introduction a la phonetique du francais (Paris: Bordas, 1974), 219.
(8.) Ibid., 201.
(9.) Encreve, 271.
(10.) Alain Thomas, "La liaison et son enseignement: des modeles orthoepiques a la realite linquistique," in The Canadian Modern Language Review 54, no. 4 (June 1998): 549.
(11.) Leslie DeAth, "Double Consonants and Gemination in Sung French," Journal of Singing 64, no. 2 (November/December 2007): 206-8
(12.) Pierre Fouche, Traite de prononciation francaise (Paris: Klincksieck, 1969), LVII.
(13.) Jean-Claude Milner and Francois Regnault, Dire le vers (Paris: Seuil, 1987), 28-9.
(14.) Georges Le Roy, Traite pratique de la diction francaise (Paris: Delaplane, 1911).
(15.) Milner and Regnault, 61.
(16.) Milner and Regnault, 12. In his introduction, Milner writes that the great eighteenth century French actress Adrienne Lecouvreur took advice, to declaim her role, from the foremost grammarian of her time, Cesar Chesneau Dumarsais.
(17.) Unless they choose to breathe.
(18.) See Debussy's "Noel des enfants qui n'ont plus de maisons." Les ennemis / ont tout pris present the same case.
(19.) Grammont, 13, 41.
(20.) Carton, 87.
(21.) Grammont, 105.
(22.) There is one exception. Plural word-endings in 'rs' displayed the most erratic treatment of liaisons, be it obligatory, forbidden, or debatable. Three instances with toujours presented a case of obligatory liaison ignored each time by a fraction of singers. In Chansons madecasses, one example of forbidden liaison, parseme de fleurs et d'herbes odoriferantes, and one example of arguable liaison, plus forts et plus nombreux, yielded also a split result.
Leslie De'Ath, Associate Editor
Quebec-born pianist Martin Neron moved to the United States to complete his doctorate on scholarship at the Manhattan School of Music and presently lives in Princeton, New Jersey. Praised as "an attentive partner" (Opera News), Dr. Neron is an active recitalist, vocal coach, and educator. He has recorded different collections of French, Greek, and English art songs on the One Soul label, and a program of songs by Mikis Theodorakis on the Romanos label.
TABLE 1. Vocalic harmonization. Singer 1 2 3 4 des baisers apaiser connaissez 1870 Reynaldo Hahn le beze apeze karese 1880 Ninon Vallin le beze apeze eklere 1890 Charles Panzera te beze apeze eklere Pierre Bernac le beze eklere 1900 Hugues Cuenod le beze apeze karese 1910 Suzanne Danco apeze kanese 1910 Camille Maurane le beze apeze kanese Gerard Souzay te beze apeze kanese Irma Kolassi te beze apeze 1920 Pierre Mollet te beze apeze lese Gabriel Bacquier apeze zereve Jean-Christophe Benoit te beze apeze konese Michel Senechal le beze apeze delese Regine Crespin apeze eklere 1930 Mady Mesple de beze karese Bernard Kruysen te beze apeze konese Jacques Herbillon te beze apeze eklere Bruno Laplante se beze apeze konese 1940 Jose Van Dam le beze apeze konese Colette Alliot-Lugaz le beze apeze sele 1947 Rachel l'akar te beze karese Francoise Pollet apeze afene 1950 Gilles Cachernaille de beze eklere Didier Henry de beze eklere Francois Le Roux le beze apeze eklere Jean-Francois Gardeil de beze apeze delese Jean-Paul Fouchecourt le beze delese 1960 Isabelle Vernet le beze apeze maltrete Franck Leguerinel apeze karese Veronique Gens apeze konese Singer 5 6 7 (r)eveiller heureux cheveux 1870 Reynaldo Hahn reveje oro fovo 1880 Ninon Vallin oro favo 1890 Charles Panzera eveje oro Pierre Bernac oro favo 1900 Hugues Cuenod eveje oro 1910 Suzanne Danco eveje oro favo 1910 Camille Maurane eveje oro favo Gerard Souzay eveje oro favo Irma Kolassi eveje oro favo 1920 Pierre Mollet cero favo Gabriel Bacquier eveje favo Jean-Christophe Benoit caro favo Michel Senechal reveje orca favo Regine Crespin eveje favo 1930 Mady Mesple egcje cero favo Bernard Kruysen eveje oro favo Jacques Herbillon eveje oro favo Bruno Laplante eveje oro favo 1940 Jose Van Dam eveje oro favo Colette Alliot-Lugaz eveje favo 1947 Rachel l'akar eveje cero Francoise Pollet eveje favo 1950 Gilles Cachernaille reveje cero Didier Henry reveje cero favo Francois Le Roux eveje cero favo Jean-Francois Gardeil eveje cero favo Jean-Paul Fouchecourt cero favo 1960 Isabelle Vernet eveje oro favo Franck Leguerinel asaleje oro favo Veronique Gens eveje oro favo
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||LANGUAGE AND DICTION|
|Publication:||Journal of Singing|
|Date:||May 1, 2008|
|Previous Article:||An appeal for patience and longsuffering by singing teachers in their assessment of the value of voice science.|
|Next Article:||Music theater singing ... let's talk. Part 2: examining the debate on belting.|