By Kenneth Kreitner, University of Memphis
The editors of the ICB have asked me for a few practical words about the pitch of Renaissance choral music, and it would be lovely to answer with an email along the lines of “A=409. Thanks for asking.” But things are never as simple as that, and in the case of Renaissance choral music, which was normally rendered a cappella, the whole idea of a rigid pitch standard seems to begin to border on nonsense. So yes, there is much that we don’t know, but some things which maybe we can; let us start in the relatively clear water and swim out a bit from there.
Trained singers in the Renaissance, with hours and years of solfège practice starting when they were small children, were unafraid of clefs and had little use for ledger lines. This means that the original clefs of a piece of music (found, in good modern editions, on prefatory staves at the beginning) are a pretty good indication of its ranges. A part originally in treble clef is unlikely to go much above a g” or much below a d‘, one in soprano clef not much above an e” or below a b, and so on; you can work this out on music paper if you want, but what it amounts to is that the name of a clef corresponds reasonably well to a comfortable choral range (even if maybe a little low for a solo range) for a singer of that name. This in itself is a handy thing for choral directors to know: that these original clefs are a quick way, when browsing through the monuments looking for something to sing, to grasp the ranges of individual pieces. But there is a little more to the story than that.
The sacred vocal music of the late sixteenth century tends to fall into two patterns of clefs: normal clefs, or chiavi naturali in Italian, usually with soprano on top and bass on bottom, and high clefs, or chiavette, usually with treble on top and tenor or baritone on bottom. Palestrina’s Pope Marcellus Mass, treble-mezzo-alto-alto-tenor-tenor, is a familiar example of high clefs; his Sicut cervus, soprano-alto-tenor-bass, is in normal clefs. Look through a volume of the Palestrina or Victoria collected works sometime and see how strictly this distinction is kept – and it’s an odd thing because the two clef combinations outline ranges, in all the parts, about a third apart, and at least in the case of Palestrina we know that the music was all written for the same choir.
We happen to know some useful things about Palestrina’s choir, the choir of the papal chapel, in the sixteenth century. We know that their number was officially 24, but in practice around 30. We know that they often sang one-on-a-part. We know that they were adult males: boys were not allowed and women were always, of course, out of the question. There were a very few castrati in the Sistine Chapel in Palestrina’s time, but mostly we are talking about, in modern terms, countertenors, tenors, and basses. We know that they sang with no instruments at all: the Sistine did not have even an organ. And we know that they were professionals, trained in the choir schools and singing this music – music, in large part, written for them personally – for several hours a day, every day, with virtually no rehearsal. All this adds up to a reasonable suspicion that the two patterns of clef-codes actually mean something important.
Feel free to work this out on music paper too if you like, but the upshot is that music in normal clefs generally works well around A=440 for a group of falsettists, tenors, and basses today, and that music in high clefs doesn’t. As I say, it’s about a third higher, in all the parts, and this interval is a lot, especially over the course of a long piece: if you have ever tried to sing the entire Pope Marcellus Mass at its written pitch, you know how fatiguing it is especially for the basses. The theory, then, is that music in high clefs was meant to be transposed down about a third (or, more accurately, intoned about a third lower) so that all this music would be approximately in the same range, and that composers, singers, and chapel masters everywhere – not just in the papal chapel – knew this and did it as a matter of course. The reason the music wasn’t written down a third in the first place is that this would require a sharp key signature, and sharp key signatures were not permissible under the rules of solmization and musica ficta.
I am oversimplifying quite a bit here, of course: there are many complications still a good deal of debate about the details (see especially Andrew Johnstone’s article, and its bibliography, in the Additional Readings below). But the general idea of moving high-clef music down is supported by the advice of theorists and by, for example, surviving organ parts from times when the music was not performed a cappella. And so, in sum, it seems abundantly clear that Palestrina and a lot of his contemporaries wrote their music with some such clef-code in mind.
The problem is that we today have a kind of unspoken clef-code of our own for this music, and it goes in the opposite direction. Our choirs are mixed, not all-male, and are apt to be loaded towards women. Our best sopranos are trained as soloists and like to sing high, our altos are female, and our tenors are few and precious. In general, then, high-clef music works better for the choirs we have today; and the result is that when we sing Renaissance music, we tend to choose high-clef music or to transpose normal-clef music up. (The classic 1922 Schirmer octavo of Victoria’s O magnum mysterium, which so many of us cut our teeth on, raises the motet, soprano-alto-tenor-bass clefs in the original, a fourth.) Nor, to be honest, has the cause – if cause there be – been helped by the number of professional mixed choirs specializing in Renaissance repertory that habitually, and with glorious results, transpose up from where it’s written.
More along that line presently, but three quick thoughts before we go on. First, I have focused on Palestrina in part because he is a popular and familiar composer today and in part because his music and his situation together add up to produce a relatively clear-cut case; the lessons learned there do seem to apply quite naturally to his contemporaries on the Continent like Victoria, Lasso, and Guerrero, and it is safe to suppose that some sort of clef-code was understood there too, and presumably in repertories like the Italian madrigal. But it is less safe to apply this notion to English composers like Byrd and Tallis, whose music presents its own problems, more complex than we can get into here. Secondly, it is hard to know how far back to go with the whole high- and normal-clef idea: the music of the generation of Gombert, Willaert, Clemens, et al. does not seem to show such a clear differentiation of stereotyped clef combinations, and even less the music of Josquin’s time, or Ockeghem’s, or Dufay’s. And thirdly, as I started to say at the beginning, exact pitch standards matter only when you’re singing with, or in alternatim with, an organ or other instruments: in a fundamentally a-cappella world, some fluctuation is going to happen.
There is a temptation to put all this together and decide that things are messy enough to absolve us from fretting over any decisions we might feel like making about pitch – that since pitch varied back then, there is no point worrying about it now. That, I think, would be a mistake. Written pitch in the Renaissance may not have indicated an exact performing pitch in choral music, but it did mean something, as the clef-codes prove and as, for that matter, the very existence of key signatures proves. Composers and scribes were at some pains to adjust the level of their written pitches, and in general, clef-codes and exceptional cases (e.g., again, the English) aside, they placed them in a way that works well with adult male singers, countertenor through bass, within a step or so of A=440. Flexible pitch, in short, does not mean whimsical pitch: if we move music up and down freely, we really do risk misrepresenting it. And there, to return to where we were a few paragraphs ago, is the problem that we have to face honestly and bravely. Any choral director who doesn’t love Renaissance music and want to sing it should probably find another line of work; but we also have to recognize that in the modern mixed choir we have a rather different instrument from the one the composers had in mind. The trick, then, is to negotiate that difference in the way that does the least damage to the intended sound of the music. And this means, for one thing, not messing with the indicated pitch level of a piece, insofar as we understand it, any more than we have to.
So: a few practical words. For most sacred music of the late sixteenth century, if it was originally in normal clefs, don’t transpose; if it was in high clefs, bring it down somewhat. If this causes trouble for your singers, let me tell you I have had good luck moving a tenor or two onto the alto part and a baritone or two onto the tenor part. For English music and music before Palestrina, think of the rules more as guidelines: anything that clearly looks like high clefs might be brought down, but otherwise, it’s worth trying to stick close to the written pitch. If we love this music, we have to respect how it was supposed to sound; we should be adapting ourselves to it, not it to us.
Additional Readings
- Fallows, David. “The Performing Ensembles of Josquin’s Sacred Music.” Tijdschrift van de Vereniging voor Nederlandse Muziekgeschiedenis 35 (1985): 32–64.
- Haynes, Bruce. A History of Performing Pitch: The Story of “A.” Lanham: Scarecrow Press, 2002.
- Johnstone, Andrew. “‘High’ Clefs in Composition and Performance.” Early Music 34 (2006): 29–53.
- Kreitner, Kenneth. “Very Low Ranges in the Sacred Music of Ockeghem and Tinctoris.” Early Music 14 (1986): 467–79.
- ———. “Renaissance Pitch.” In Companion to Medieval and Renaissance Music, ed. Tess Knighton and David Fallows, 275–83. London: Dent, 1992.
- Kurtzman, Jeffrey. “Tones, Modes, Clefs, and Pitch in Roman Cyclic Magnificats of the 16th Century.” Early Music 22 (1994): 641–664.