Einsiedeln, Stiftsbibliothek, Codex 121(1151), p. 204 – Graduale – Notkeri Sequentiae

Einsiedeln, Stiftsbibliothek, Codex 121(1151), p. 204 – Graduale – Notkeri Sequentiae

Some Details of Notations and Performance

While the details to be mentioned are properly part of the general subject of signs of nuance, they may be identifiable by chant connoisseurs only in specific traditional terms; these are listed here, with my explanation of my own procedure. 

Quilisma: I respect most of the irreconcilable opinions about the meaning of this notation, usually described as ornamental. For years I sang it as a mordent. I find that too elaborate, yet I am persuaded by David Hiley's argument that such an elaborate sign ought to have some degree of elaborate melody; but I have had to balance that opinion off against the fact that L often omits the sign—and apparently the thing—altogether. I find that what is needed is a tone that moves forward to fulfilment, whether (a) b c or (F) G a, or some other configuration. This melodic function leads me sometimes to move through the b or G quickly (the Solesmes solution) but not necessarily lightly, other times to lengthen and stress it as an anticipation; and I do not decide consciously between these alternatives.

Oriscus: clearly a sign used for a repetition of the immediately preceding pitch; but sometimes apparently for the tone above. That irregularity (usually indicated by the oriscus sign being placed slightly above the preceding note), plus other contextual evidence such as its elaborate grapheme, suggests that it has additional significance. I find that lenthening and increased expressiveness are often appropriate, and that the oriscus may in those respects differentiate this instance of repeated pitches from others that are notated differently.

Bivirga, trivirga, tristopha: most specialists have agreed that repeated pitches should be individually articulated (the word repercussion is often used) rather than rendered as one prolonged pitch. Often, however, modern singers have not found a way of articulating that did not sound either strange or awkward (“goat trill”). From my vocal coach I learned how to “repercuss” without sounding awkward; it is after all, a regular part of European vocal practice. Whether it sounds strange is a matter of personal opinion.

Hence, the only problem with repercussions (aside from that inappropriate term) concerns their relative lengths and emphases—the rhythm of the figure as a whole. I find this to be almost complete a matter of performer's option, and I try various arrangements without considering any of them final.

Liquescents: signs of liquescence are very close to signs of nuance in that they suggest ways of modifying the melody that is to be sung to the given words. These signs are, therefore, ad hoc rather than systematic: they differentiate in response to context; hence, rules a priori are hard to find. It seems to me that an intuitive, pragmatic approach is the most useful one. At the same time I find liquescents of all types to be an extremely valuable component of rhythmic shape, and I use them with enthusiasm at every opportunity. One of the interesting uses of adjunct letters is to direct the liquescence either up (s) or down (i).

There is a case of a liquescent sign that can take on alarming proportions: some moderate examples are at hand in the Gradual Universi (First Sunday in Advent). More than one kind of sign may be involved, or, it may be subject to various interpretations. There is a famous mark in Chinese notation that winds around in a similar fashion over a large part of the page. The Gregorian mark is relatively restrained, but even so, Hiley's paradox applies. I conclude that the sign calls for merely an extravagant enunciation of the liquescent consonant.

Breathing: not usually notated in any European music before the twentieth century. Did they not breathe for ten centuries? Yes, they breathed; but career singers would have been embarrassed by any notation whose presence implied that they did not know when to breathe, or that suggested that they might run out of breath.

It has long been observed that in Gregorian chant the musical phrase at the highest level is syntactic: we can always breathe at the colon or period—indeed, it would be contrary to sense not to breath there. The problems come in florid chant using persistent groupings of four, five, six or more pitches per syllable, even if no melisma is involved. In this kind of florid chant the syntactic phrase may be stretched further than is practical for a musical phrase. In such cases there is usually a subphrase structure, and the question is whether a breath after a subphrase makes that phrasing too obvious; also, whether such a breath breaks up the diction in an awkward or inappropriate way.

Choral practice uses staggered breathing to solve these problems; but I find something artificial in the effect of staggered breathing. There seems to me that a more obvious connection is required between breathing on the one hand, and the rhythm of phrasing on the other: phrases that are too long to be sung in one breath need to be acknowledged as such, or else such phrases run the risk of sounding disembodied. The acknowledgment may, however, take an artful form, such as a breath in mid-phrase, taken in such a way as not to create a terminal articulation. Since for solo performance I had perforce to find places for such breaths, I found them; and it seems to me that the places are often marked with a t (for tenete, which I regularly read as a held pitch that slightly interrupts the rhythmic flow), or some equivalent sign.

A few more general questions of performance can be noted here. There is little medieval documentation for any of these questions—if, indeed, there is any at all. My solutions are all obvious in the way I sing, and I discuss them here only to respond to the possible curiosity of scholars or singers.

Choice of pitch: there is no hard evidence concerning the pitch level of medieval music. There are some casual comments, hard to correlate or corroborate, to the effect that some modes are higher, some lower; that seems to concern tone or character more than absolute pitch. Modes are notated on D, E, F, or G, but with many more transposed notations than are used in modern editions, primarily for the sake of avoiding accidentals. Medieval chant notation, especially of the eleventh and twelfth centuries, transposes whole phrases by a fourth or a fifth, without warning, simply to indicate the use of a flat in that phrase. Only in the numerical office is there any suggestion of careful maintenance of notated pitch level among consecutive items. In Mass formularies there is a persistent variety in the choice of modes throughout each formulary, suggesting that maintenance of a pitch level was to be used in such a way as to produce variety rather than uniformity. When singing at Mass I observe variety scrupulously.

For study purposes, however, this Archive is arranged to facilitate comparison; hence, items of chant are grouped together by mode, rather than in order of liturgical performance; and the items grouped in one mode are sung at the same pitch—as close as I can manage. My choice of this pitch, however, is expedient: for instance, I sing the graduals in mode 2 on an E final, a fourth below their usual notated level, a step higher than the generic level for other chants in mode 2. These mode 2 graduals, which use one model melody, would never be heard in this relationship except on Ember Saturday in Advent, a Mass which, by exception, has four Graduals and all in the same mode.

Tempo (pace): I believe this is almost completely a matter of performer's choice, and depends on judgment concerning occasion, ambient, listeners, liturgical function, size and nature of performing group. Choice of pace is of course almost indistinguishable from choice of tone on one hand, and choice of rhythm on the other. What is involved is, it seems to me, a global notion of how the piece should go under conditions of performance.

Rhythm: long ago I learned Dom Mocquereau's rhythm, and have used it ever since, faster and in my own way. I find in it much truth about musical rhythm, quite apart from its application to Gregorian chant; but I also find some of the applications of his rhythm to Gregorian to be inappropriate. Because of differences in many details, as well as general differences of tone and pace, I imagine my performance will not reveal how much it owes to Dom Mocquereau.

I take very seriously a basic observation of Frieder Zaminer, that in Gregorian chant, rhythm cannot be considered separately from melody or other dimensions.

Tone: I strive for the least encumbered, most “natural” sound that I can produce as a singer, and I credit whatever success I may have in achieving such tone to my long-time vocal coach Lilian Loran. As a music historian I can present a long, complex argument that supports the use of that kind of tone for modern performance of Gregorian chant, but this is not the place to present it. I try to make the chant sound lyric, melodically expressive, and appropriate to the words being sung, without much sense of drama or rhetoric.

I have sung Gregorian chant at Mass in a variety of acoustical environments. For the purpose of this recording—a study edition—I thought that the most appropriate performance would be the one that was clearest in word and tone, with as much musical quality as I could achieve. I have experimented with different kinds and degrees of electronic “reverb.” My model would be the sound as heard a few feet away from a cantor in a moderately large, moderately resonant church.

Pronunciation: several studies have appeared giving alternatives to “Liber-usualis Latin”; these alternatives are clearly more appropriate to Gregorian and medieval chant than is the abstract Italianate pronunciation identified with the Vatican hierarchy. Documentation for tenth-century pronunciation from the Carolingian heartland, however, seems lacking, at least in a form sufficiently specific to guide a singer. In order to learn to sing in a foreign language, a singer needs to hear native speakers, not just read directions in a manual. Lacking such speakers as models, I content myself with making the words as intelligible as I can, avoiding anything that obtrudes on or interferes with the musical effect. I believe that throughout its history, the chant's Latin pronunciation has accommodated itself to the singer's native language; mine is Eastcoast North American English.