When written signs are given life in sound, the untrained reader or singer does more than communicate a naked text: he unconsciously imbues it with meanings unrelated to its content. For he may also reveal his background, his place in society, even his personality. Learning to pronounce a language well is not a dry, mechanical exercise; rather is it the acquisition of a rich palette of vocal articulation, that the skilful reader will control deliberately, subordinating his own person to the requirements and embellishment of the text. My aim here is to show that, even though these conventions may become modified over time, they nevertheless maintain a coherence and consistency that go far beyond those of the natural language from which they originate, and here it is tradition that plays a vital rôle.
Good pronunciation is, so to speak, the substructure of singing, the articulatory frame without which singing would be nothing more than a sequence of sounds, tuneful perhaps, but still shapeless. Good pronunciation is an essential component of «style», the first of the keys to good singing — first, because it is closest to language.
It has been said that every language has its peculiar rhythm, often described as its stress-pattern. The French language seems to have evaded such an analysis; for you would look in vain, either in ancient or in modern literature, for any generally agreed theory about stress-pattern in French. Yet the rhythm of French is just as real, though its native speakers may produce it unconsciously and feel it only vaguely, while it is often vulgarized, whether deliberately or not, by non-native speakers.
How far is poetical recitation consistent with the natural rhythm of a language? It is better to describe it as a sort of stylized form, since it amplifies certain elements at the expense of others. You could say that recitation was a biased version of the rhythm of a language rather than a faithful reproduction of it. Further, when it comes to singing, the rhythm of recitation is not simply copied, but stylized anew. Even there, it may depart from its model or even conflict with it. That the relationship between the rhythm of a natural language and that of singing is so difficult to grasp and to describe is simply because, as with engravings made of paintings, we are dealing with two successive stylistic modifications.
From an historical point of view, there are widely differing aspects to the rhythm of singing. There are, for example, hardly any points in common between a monodic lai by Machaut and a recitative by Lully. Yet treatises on recitation often pay no attention to the history of the language and seem to put the Chanson de Roland on an equal footing with free verse of the twentieth century. They try to extract, from chronologically disparate examples, timeless principles of the rhythm of French recitation. Taken out of their particular contexts, such principles run the risk of remaining merely abstract. To gain form and substance, rhythm needs to be embodied in a language itself anchored in real sound and in the history of the language.
According to the particular character of the articulatory framework that it encounters, one and the same rhythmic idea can be realized in extraordinarily contrasting ways. Whatever may be its rhythmic ancestry, a line of verse of the twelfth century cannot be spoken in the same way as a romantic alexandrine. In singing, rhythm is fundamental, but not autonomous; it is derived from an harmonious agreement, based on careful scholarship, among the rhythm of language, the rhythm of verse and of music. This is the second of the keys of singing.
Writing is the least appropriate way of giving an account of the voice. Metaphors are unavoidable and quite ineffectual. Not through speech, but through interaction with a teacher do you learn to train the voice. There is, besides, a too frequent confusion nowadays between the study of singing and of the voice. Training the voice is, of course, indispensable for good singing, but is not enough. All too often you hear beautiful voices that make you wish they were not burdened with a text! The tendency to develop the voice as an instrument, rather than as a vehicle for singing, leads to the error of forgetting that singing is essentially a way of speaking.
You cannot, though, neglect this aspect of singing once you have adopted an historical outlook. The few available documents hardly ever offer a positive picture of the aesthetic of the voice at any given place or time. The voice in history is always framed negatively by the unknown and the unknowable. How can we know how far Franco-Flemish singers used vibrato? It is nevertheless logical to allow that a vibrato so broad as to compromise the intelligibility of the polyphonic lines could not have been current in that music. Likewise, the historical approach to language casts a negative image of the voice: from an historical analysis of the vowels of cultivated French, one may conclude that the nasal resonances that characterize certain types of folk-singing were never current in learned French vocal music.
Generally speaking, teachers and singers with a good understanding of the language and its history should be able better to control the trained voice, the third of the keys of singing and the closest to music, to put it to the service of the art of singing.
Translation : Michael Bulley