To chime in on Jamelle's post about the DEA's effort to hire Ebonics speakers, I'd like to point out that African American Vernacular English (AAVE) -- like all languages and dialects -- doesn't just refer to vocabulary differences. For some background: Linguistic differences tend to arise when groups are socially isolated. Over time, these difference can diverge so much from the original they are considered a different dialect or language (the litmus test is mutual intelligibility, so depending on whom you talk to, AAVE is either a dialect of English or a separate language). AAVE shares many features of the Southern dialect of American English, though as with standard English, there are regional differences.
Unfortunately, discussions about AAVE are generally limited to slang terms -- in the case that Jamelle's addressing, terms related to the drug trade. But in fact, there are a lot of other linguistic features that characterize AAVE.
On the syntactic front, AAVE speakers have a more granular tense-marking system. In standard English, for instance, "James is happy" can mean either that James is happy at the moment or that he is habitually happy. AAVE uses the verb "to be" to mark the habitual form, but omits it otherwise:
James happy = James is happy right now
James be happy = James is usually happy/a happy person
In terms of pronunciation, many speakers of AAVE have replaced the sound "th" -- as in someTHing -- with "f," so you get "roof" instead of "Ruth." AAVE speakers also pronounce vowels higher in the mouth when they precede an "m" or "n," leading "empty" to sound more like "Impty" (this is common throughout the South).
These are just some of the features of AAVE that have been widely studied by linguists (for a look at others, you can go here). Not every speaker of AAVE needs to exhibit all of them, nor do they only occur in AAVE. For instance, in Martin Luther King's "I Have a Dream" speech, you can hear him omit the "r" from the word "later" -- a common feature of AAVE -- but he otherwise uses the syntax and vocabulary of Standard American English. And omitting the verb "to be" is common in the world's languages, including Hebrew, Russian, and Hungarian. Furthermore, speakers can switch between standard English and AAVE, a common phenomenon among bilinguals called "code-switching."
To relate the discussion back to the DEA's push to hire agents who speak AAVE, it seems that the DEA is more interested in agents who are familiar with the vocabulary of the drug trade than with those who exhibit the characteristic syntax, phonology, and morphology of AAVE. Recruiting people for this purpose is a lot harder to do than just hiring AAVE speakers, who may or may not be familiar with the vocab of the drug trade in a particular area. The broader point is that AAVE is a much richer and more interesting phenomenon than is generally acknowledged. In this case, that failure of understanding is the DEA's folly: Recruiting any old AAVE speaker to decode drug communications is like taking a random 80-year-old off the street and asking him to decode 14-year-olds' text messages.
-- Gabriel Arana