Race and “Voice Quality:” A Skeptic’s Viewpoint

During an unrelated Google search the other day, I stumbled upon this Yahoo Answers query:

Can you tell someone’s race from the sound of their voice? I was wondering if you could tell if someone was white or black etc by hearing their voice?

This is obviously a controversial topic. Ethnolects are real phenomena, of course, but the notion that they evince genetic differences makes me uncomfortable. Such memes have spread through the internet for years, typified by this uninformed comment (on a different message board):

Black people’s voices have a definite “blackness” to them – I’m referring specifically to the timbre or sound, before grammar, inflection, diction, etc. even come into play. This rules out the sociocultural answers (“blacks are more likely to use slang,” or “blacks tend to be more relaxed with language”, etc.) that are often given.

Don’t get me wrong: one can sometimes deduce ethnicity by voice alone. In 2011, researchers from CUNY found that young New Yorkers could ascertain, from mere recordings, whether a speaker was East Asian, African-American, Latino or “White.” The study identified factors that “cue” a listener’s expectations, one of which you can hear in this clip of comedian Cheech Marin:

Marin’s speech, like that of many Chicano-Americans, is arguably more “syllable-timed” than that of General American English, meaning that (roughly speaking) each syllable is more equal in duration. Laypeople may associate this tendency with Latino or Hispanic speech, even if they’re unaware of the connection. To be clear, though, this has nothing to do with racial genetics, but rather the influence of Spanish.

I may have strayed from the Yahoo commenter’s question, however, since “voice quality” is arguably a different animal from intonation. Indeed, the CUNY researchers found other factors along those lines, such as a type of “breathiness” which people associate with Asian-American voices. A possible example might be cellist Yo-Yo Ma‘s soothing, professorial tenor:

But how are genetics involved? After all, many Americans with immigrant backgrounds exhibit vocal qualities indicative of their ancestry. It is hard to dismiss connections between the accent of Jewish New Yorkers and their forebears’ unique Germanic language, nor those between Irish-American speech and Hiberno-English. We know that the Jewish and Irish “races” are the stuff of Victorian pseudoscience, so why discard the impact of ancestral languages on broader ethnic categories?

To be fair, I have found at least one study suggesting possible ethnic differences in vocal tract anatomy.  But what about the genetic variation within “races?” For instance, latitude, diet, and random selection produce substantial height differences between sub-regions within East Asia, Europe, and Africa. Why would the human voice be any different? The burden of proof is upon those who seek interracial genetic explanations.

**An important note, in the form of a tongue-twister: the researchers are not suggesting  that all Asian-Americans speak with this quality, but rather that Asian-Americans who speak with this quality are more easily identified as Asian-American.

Reference: Newman, M., & Wu, A. (2011). “Do you sound Asian when you speak English? Racial identification and voice in Chinese and Korean Americans’ English. American Speech, 86(2), 152-178.

(Source: dialectblog.com)