Yes, it's ironic, but predictable, that people express ambivalence toward second or heritage languages: They attack languages such as Spanish that they perceive as a threat, while defending "cute" languages like Boontling (you and I are may be the only ones in the audience who know what that is) or "antique" languages like Cornish.
Yet even that picture is not so clear: If Anglo parents in New York are choosing to have their very young children begin the study of Spanish, as the article in today's New York Times indicates, does this mean, as the Times suggests, that they have come to value multilingualism, or does it mean they are just being practical? And how does that stack up against research indicating that second-generation Latinos and Latinas are often monolingual English speakers? I've collected lots of anecdotal newspaper stories in which employers in Miami, for example, claim they can't find Spanish speakers, let alone bilinguals, to hire to deal with their Spanish-speaking clientele.
Navajo has long been touted by sociolinguists as a language whose speakers maintain a great deal of language loyalty, that is, the Navajo are reluctant to give up their heritage language. Yet the need to establish a high-school Navajo class suggests that off the reservation, that is exactly what young Navajo speakers are doing. The Hasidim of New York and the Amish are two other groups known for preserving their speech--when, that is, they maintain their social isolation from the mainstream. But once that isolation is gone, language may soon follow.
Cultural-loyalty movements like the class in Navajo tend to spring up when cultural preservation is in danger, and their record of success is not impressive. Here at the University of Illinois, we run special sections of first- and second-year Spanish for Spanish-heritage students, and the reason is that their Spanish is too weak to meet academic requirements. That in turn means they may be able to speak Spanish (even that is often not the case); but they can't read it well or write it fluently, and their grasp of its grammar may be weak, since they haven't actually studied it as a language. They may speak some Spanish, but they can't pass a test in it. Students also use college language classrooms to try to recapture the heritage languages they may have lost (for example, Hindi or Urdu), or ones they may never have acquired (Yiddish).
What interests me about these minority language and dialect matters is why ebonics was perceived as so great a threat when the whole issue arose in late 1996 in Oakland, Calif. We are living in a time when English serves as an international language. Furthermore, living as we do in a (largely) postcolonial world, we tend to recognize that each national variety of English, that of the United States, of Canada, of Australia, of India, of Nigeria, of Ireland, has its own local standards, and we (that is, linguists) have begun talking about world Englishes, not simply world English, to emphasize this fact. Yet within a community, group variation in standards still tends to be rejected or regarded with suspicion. Hence ebonics was seen by the American public at large as a declaration of independence in need of quashing, an internal threat, a wild mistake, while American English, itself a partly revolutionary, postcolonial product, is seen as simply "the way things are" (that is, unless you're Prince Charles).
Someone posted an e-mail response to your morning essay asking whether it's true that ebonics speakers can't do math because their language isn't capable of rendering the subtleties of math. Sometimes it seems to me that every language except my own is capable of rendering those mathematic subtleties, and my colleagues in math sometimes insist that the subtleties of math cannot be rendered into English, or at least English that is understandable by anyone unfamiliar with math. A high-energy physicist once told me, when I asked her what she was working on, "I couldn't possibly put it into words"--yet she continues to receive NSF funding, which suggests someone understands what she is up to on her grant applications. The math case may really be what the computer folks call COIK, "clear only if known." But this is the kind of thinking about language that pervades our society. How would you answer that question about ebonics and math? Language varieties are flexible and adaptable. But they also exist in contexts, and while these contexts don't necessarily bind them, they do affect them. So if the context that some ebonics speakers exist in is one that does not bring them up against the differential calculus, what follows? Low math scores, for one thing. But that's not the same as saying ebonics speakers can't do math because they can't speak math.