Skeptoid PodcastSkeptoid on Facebook   Skeptoid on Twitter   Skeptoid on Stitcher   iTunes   Google Play

Members Portal

Store

 

Get a Free Book

 

SKEPTOID BLOG:

Singing voices: a classification or a convention?

by Bruno Van de Casteele

July 21, 2013

Share Tweet Reddit

Donate Traditionally, singing voices have been classified in six groups: three for men (bass, baritone and tenor) and three for women (soprano, mezzo-soprano and contralto). But is there something behind it, or is it only convention?

Hugo Lycke asked the same question, and tried to answer it in his doctorate this year at the University of Louvain. Lycke was probably trying to get some record, too, as he is 75 years old and funded his doctoral research himself. But the advantage that he got were his decades of experience as a vocal coach and in logopedics.

His initial starting point, as he explained it to the University of Louvain's newspaper (link in Dutch), was the sometimes very subjective discussions during auditions whether a candidate was, say, a bariton or tenor. I understand that problem, too. I think I'm more of a bariton, but when I used to sing in a choir I was always grouped with the basses. One day, just for fun, I decided to go sit next to my brother with the tenors, and could quite hold myself.

Lycke would say that is not so weird. Given training and exercise, it is indeed possible to extend your range. I was probably switching to falsetto when trying to match my brother, and because it was a choir, that switch was hidden within the sound of the other tenors. But it is probably not optimal, and Lycke warns against damaging your voice by singing the "wrong" part.

But how do you know your part? Lycke took fonetograms of 500 people, and analysed them statistically on 34 attributes.

[caption id="" align="alignnone" width="283"] Example of a phonetogram (illustration Lycke)
Example of a phonetogram (illustration Lycke)[/caption]


After statistical analysis, both for men and women three distinct groups appeared. This indicates that the distinction made since several centuries, seems to be based on a natural phenomen.

However, Lycke is quick to point out (in the university magazine article linked above) that it is still a long way for a standard ID for a voice. It is a statistical analysis (allowing for outliers), and furthermore more research is needed to see if those three clusters really align to the traditional singing voices. As the title of the doctorate indicates, this is an exploratory study, giving some solid footing to further work. But the last word, or rather, the last note hasn't been sung yet. Lycke himself has retired to France, where he still remains active as a vocal coach (aside from his bed & breakfast Maison Conchette). Given the average life expectancy, maybe there are some follow-up articles we can expect...

 

by Bruno Van de Casteele

Share Tweet Reddit

@Skeptoid Media, a 501(c)(3) nonprofit

 

 

 

Donate