Is it really about Sounding Black?

The Black Blog Buzz is all about a recent social science research study about Speech Patterns and Wage Inequalities in the US work place, by Dr. Jeffry Grogger at the University of Chicago.

I like how Cobb framed the discussion as an ability/inability to Code Switch. Code switching is an important tactic among ambitious people (race, class, and nationally doesn't matter). And in my interactions with urban youth and young adults, I believe lacking this skill frustrates their lives, especially when they encounter civil servants like police, social workers, or even customer service personally at stores and service providers.

One commenter on YPB Guide thought this study was a waste of time and energy, but as noted by the author, though this subject has received much attention there has never been an empirical analysis of the phenomenon. In fact, this study "represents the first attempt to determine empirically whether native-language differences between blacks and whites help explain racial wage differences." That is a big deal. To not only document but demonstrate how perceived differences in people, based on language patterns, can have a big impact on their employment status.

Now, I've always thought that the idea of 'sounding black' had more to do with the sound and rhythm of the speaker's voice. But according to linguistic standards there are variants of speech based on indices, one called Standard American English (SAE) and African-American English (AAE). Both are characterized by several several variables including syntax, grammar rules, phonology, and acoustics - basically, at every linguistic level.

The author had recordings of young people being interviewed for a summer program. There were a team of volunteers who listened to the recordings and were asked to indicate which group (black or white) they think the caller belonged to. They were also asked to indicate the speaker's sex, education level, and region of origin they thought he/she belonged to.

Listeners were able to accurately identify the sex of the speakers (98%). 84% of white speakers were accurately identified and 77% of black speakers were accurately identified by the listeners. The listeners were not able to identify education level of speakers and did not do well at identifying the region of origin of speakers, either.

The author then applies a statistical model and asks - Could these perceived differences in race, based on language impact wage differences between these groups. So the rest of the paper explores how such perceptions can the reason for wage inequalities between these groups. Though I'm not about to get into all of the statistical modelling stuff, it is an interesting thing to connect together. Could speech be a basis for discrimination? It certainly has been demonstrated in housing discrimination.

But a better question is: What do we do with this information? Help people speak better SAE or how to code switch or help people understand that using AAE doesn't mean you less intelligent, productive or profession? I think a little of all three.