SCIENCE
By Edward Willett
It’s not over ‘til…
Although I’m not terribly familiar with Rossini’s La Cenerentola, Opera Saskatchewan’s current production, I think it’s safe to say that it does, at some point, feature a soprano singing very high notes. I think it’s also safe to say that many people will find it hard to understand the soprano’s words during that high bit.
This is not a reflection on the soprano; it’s a matter of physics, according to recent research by Professor Joe Wolfe, Dr. John Smith, and graduate student Elodie Joliveau (herself a soprano) at the University of New South Wales, Australia.
A human consists of a fundamental frequency mixed with a number of harmonics–higher frequencies generated at regular intervals. What we hear as the pitch is produced as much by the spacing of the harmonics as the fundamental frequency.
The vocal tract, meanwhile, resonates at several different frequencies. When one of these resonant frequencies matches a frequency present in the voice, that frequency is amplified. (The effect can be heard in bathrooms; when you sing in one at just the right pitch, the whole bathroom booms–you’ve managed to find one of the space’s resonant frequencies.)
One reason it is difficult to understand sopranos when they sing high notes is that most of the information our brains need to recognize vowels in other people’s speech is provided by the relative amplitudes of harmonics that aren’t present in a high-singing soprano. Consonants also become harder to understand because the vowel sound that comes before a consonant helps us to identify it.
The other reason is that in order to maximize volume–so she can be heard over the orchestra–and to insure that all notes sound equally loud, a soprano changes the shape of her vocal tract so that its resonant frequency matches the pitch she is singing. That resulting shape, however, is not necessarily what would normally be used to create a specific vowel sound, thus making it harder for listeners to figure out what vowel is intended.
If sopranos didn’t do this “resonance tuning,” not only would they be harder to hear, but whenever one happened to hit a note that matched the resonant frequency of her vocal tract, that note would sound much louder than the others, resulting in uneven loudness and uneven voice quality.
Swedish acoustician Johan Sundberg first suggested that sopranos were tuning the resonance of their vocal tracts to the notes they were singing, but at the time there was no way to study the acoustics of the human vocal tract while it was being used for singing.
The New South Wales researchers recently developed a way to do so, however, and decided to test Sundberg’s hypothesis. They invited a group of classically trained sopranos into their lab. They produced a carefully synthesized sound just outside each singer’s mouth as she sang. A microphone recorded not only each soprano’s voice, but also the way her vocal tract interacted with the synthesized sound, which told the researchers a lot about the tract’s acoustics.
They found that sopranos do what we all do during speech and ordinary singing in the low range of their voices: the pitch and the vocal tract resonances are mostly independent. In the high range, however, they did indeed tune the lowest resonance of their vocal tracts fairly precisely to the pitch they were singing, by gradually lowering their jaws as the pitch ascended, and/or by “smiling” more. The result: uniform loudness and vocal quality, but also a tendency for all vowel sounds to become very similar.
That’s hardly news to composers; in the 19th century, Berlioz wrote a book about orchestration that warned opera composers about this effect. Many composers write the highest parts of a soprano’s aria to be sung on a single word, an unimportant phrase, or no words at all, for that very reason.
Of course, there are other reasons opera-goers may have trouble understanding singers. The hall’s acoustics make a difference: a hall designed for orchestral music will have a fairly long reverberation, which may make it less than ideal for opera, which will be harder to understand there. The orchestra may also be a problem; much opera was written centuries ago when string and wind instruments weren’t as loud as modern ones.
And then, of course, some singers simply have better diction than others. Joan Sutherland, for one, was famous for never singing a consonant if she could help it.
In any event, intelligible or no, opera continues to delight and thrill us with its magical marriage of human voice and orchestra. If you can’t understand the words on stage–well, that’s what surtitles are for.
