One of the nicer perks of my job is our departmental subscription to Science, which comes to me first each week. I even occasionally get time to read it before somebody else bugs me for the last issue. Since I had a little downtime yesterday afternoon, I was perusing the March 28 issue when I stumbled across this little item (subscription and/or membership required).
It seems that there may have been a reason why composers in the Middle Ages and the Renaissance composed the way they did. And that reason may have had something to do with the size and shape and construction of the buildings in which their works were performed--which may in turn have had something to do with the way that cathedrals and chapels were designed. An architectural historian took the choir of St. John's College, Cambridge (along with a bunch of acoustics people and some musicologists) on a very special choir tour to Venice, where they sang a variety of works in 11 different churches and chapels around the city.
Polyphonic works (for example, a Byrd Mass--though in some cases that might be a special case--or a Tallis motet) get mushy in great big spaces like the Basilica di San Marco. There's too much room for the sound to bounce around, and it makes it very difficult for the audience to make out the words. Simpler melodies, like plainchant (or Gregorian chant, or simple polyphonic works like Byrd's Masses) work better in the big spaces. The best spot to sing a really complex polyphonic work is in a smaller space, like a side chapel off of the bigger cathedral. The historian speculates that composers may well have tailored their works for the building in which they would be performed (which, really, any good composer should have done). She even speculated that it might have been possible, with a big enough choir, to achieve something like "surround sound"--which may explain some of the really gigantic works in the early polyphonic repertoire, like Tallis's Spem in alium, composed ca. 1570 for 40 voices. That's 40 voice parts (eight choirs of five voices apiece), not just 40 people singing the same three or four parts, by the by.
Having spent some time singing in spaces big and small, and having attended services in some pretty doggone big churches, the group's findings, which were reported, according to Science, at the Cambridge Science Festival last month, don't surprise me much. I distinctly remember a small chapel on the campus of a choir college or a seminary (I don't remember which it was) in Nashville, where we stopped on choir tour my junior year in college. Obviously, since at least one of the functions of the place was to train singers and musicians and directors for choral work, the chapel at the school was designed with singing in mind: all stone and glass and wood, hard surfaces that keep even a small space vibrant.
One of the pieces we were singing that year was C. P. E. Bach's 1749 Magnificat which, while it's a middle-period piece by chronology and should therefore be more in the classical style, is written in a style much closer to the Baroque or Rococo style favored by C. P. E.'s more famous father. (What that means to non-music buffs is that it's a little more florid and has a lot more weaving of the various lines than you would expect in a classical-period piece.) And I do remember that our director had the darnedest time reining in the soloists (including me), because that little space did such wonderful things for a voice that we were all hanging on to notes longer than we should have, just to enjoy the sound a little more.
By contrast, it's really been an experience going to Mass in places like the Basilica of the Holy Sepulchre in Jerusalem, the Church of the Transfiguration on Mount Tabor in the Galilee, or Notre Dame in Paris. All of those spaces are simply huge--and at Notre Dame, just as would have been done in medieval times--while Mass is going on in the central nave, the tourists and the pilgrims continue to go in and out, oohing and aahing as they gawk at the spectacle, in the side aisles. Even with modern microphones and loudspeakers, it can sometimes be difficult to follow what's going on in a space as big as that--particularly if the speaker or the singer(s) isn't careful about enunciating and allowing for the reverb time.
It's nice to know there may be a scientific basis for those empirical observations.
Fascinating, Michael, and consistent with my experiences in various church settings here and in Europe.
I have performed in spaces large and small, live and dry, and indeed one has to choose both the literature and the disposition of forces carefully, especially in large, live spaces. When one doesn't get to choose both, the results can sometimes be less than satisfying.
In one particularly large, live church in Austria, the challenge in performing a polychoral work was finding the right location for the conductor... yes, you read that right... so that a person in the pews somewhere in the middle received a more-or-less simultaneous sound in the work's vertical passages (i.e., most of them).
As performers, we were strictly instructed to sing or play simultaneously by sight of the conductor's beat, not by ear. I'm glad the fellow in charge knew what he was doing.
(I also have two stories set in Catholic churches in Houston, but I'll save them for another time.)
Posted by: Steve Bates | Tuesday, 08 April 2008 at 15:54