[I posted this to a linguistic anthropology listserve on January 16, 1997, in the wake of the Oakland brouhaha. Still relevant, I think.]
A language is a dialect supported by armed forces.
What does this mean? Everyone speaks a "dialect", i.e., a variety of human language. Those who happen to be in control get to call their variety a "language" , with all that implies, which is often negative for those who do not happen to speak that variety.
There is no -linguistic- difference between a language and a dialect. Each has all the features necessary for human language; or, each is a full realization of universal grammar, or the language bioprogram, or whatever, if you believe in such things and wish to state it in those terms.
Swedish, Danish, and Norwegian are all varieties of human language. They are labeled "languages" because in Sweden, Denmark, and Norway, respectively, the speakers of those varieties are in control. But, they are so mutually intelligible, they could easily be classified as "dialects" of "Scandinavian".. In fact, "Norwegian" did not become a "language" until "Norway" became a state separate from "Denmark" earlier in this century; prior to that, it was a "dialect" of "Danish".
On the other hand, the many varieties of human language spoken within the state of "China" are not mutually intelligible. Nevertheless, they are called "dialects" of "Chinese". Do I have to spell out why?
Another factor in language and dialect is speakers' attitudes towards each other. If people want to understand each other, they may think of their different language varieties as dialects of the same language. If, for some reason, they stop wanting to communicate with each other, they may begin thinking of them as separate languages. This may have nothing to do with the structure of the varieties of language under consideration.
Which brings us back to Ebonics (let's call it BE to make typing easier, OK?). BE is a variety of human language. It never was, strictly speaking, a regional variety; it is and always has been a social variety, existing alongside other sorts of English. It is not monolithic by any means, so there is some variation within what might be covered loosely by the term "Ebonics". There is some disagreement about just how it originated, but it's absolutely not, and never has been, "slang', "argot", "degenerate English", "cognitively deficient English", or "lazy English". And, of course, it is not "genetically based" as the Oakland School Board so boneheadedly put it (unless of course you mean that it's just like all other languages. And, by the way, isn't THIS a great example of why we need anthropology and linguistics in the public schools??). And, BE can be described using the same toolbox linguists use for any of the world's varieties of language: it is systematic, patterned, rule-governed, etc. For a while, it seemed to be assimilating to standard English, but then in the 80s ("Reagan Revolution"; catsup = vegetable) it started moving away again, perhaps because many of its speakers realized that there simply was no point.
Now, why might people want to label BE as a separate language? I think because for most of its history it has been labeled "bad English"; not even a "dialect". Recall that the educational psychologists of the 50s told us that its speakers were cognitively deficient because they produced sentences like 'The girl pretty' with no "linking verb". They had no real language. Apparently, these "doctors" had never studied Russian, which also allows predicates to consist of only an attributive, with no linking verb: 'Dyevushka krasivaya'. For the most part, even BE speakers themselves have been flamboozled into believing this (note that they are some of the most vocal critics of the Oakland proposal). What makes this situation so easy to maintain is that BE speakers (and others) so rarely have had the opportunity to look upon BE from a linguistic, i.e. scientific, perspective.
Calling BE a "language", then, is a political statement. It says (to me, at least) "our speech must be recognized and respected as a manifestation of the human Language potential, universal in all humans, but denied us for centuries by an ideology which claimed that it was merely a deformation of the speech of those who enslaved us, caused by our own (a) laziness and/or (b) subhuman linguistic ability."
Apparently, Oakland thinks it might be healthy to try breaking out of this pattern. Try something different, since the old racist notions about BE and the programs of action resulting from them don't seem to have worked. Let's give children (and others) explicit knowledge about both BE and standard English so that they know, consciously, what to do to move from one to the other. Note that this does NOT mean teaching the kids BE; they come to school already native speakers of it. Nor does it mean requiring teachers to speak BE. It does mean insuring that teachers have a scientific perspective on both BE and SE and the relationship between them. Perhaps most important, from my point of view, it means that teachers should know what a sentence like "The girl pretty" does NOT mean (linguistic deprivation, cognitive deficit, etc.).
Speaking as a linguist, I think that Oakland is on the right track. I wish they could have stated their aims better and thus given the inevitable attackers a bit less of a target. Are all linguists going to agree? Certainly not. From the perspective of Greenberg, a historical/comparative linguist, calling BE a separate language is probably absurd. From the perspective of the children affected, however, I think Oakland is correct. It's correct because it formalizes a respect for the intelligence and knowledge that these children come to school with, a far cry from what has usually happened to African American children in this country.
I should add that this is not "mere opinion". Studies from various parts of the world have shown that this is precisely what is needed to help people operate with a national language that is different from the home language. My own work in Grenada was along these lines, too.
To make something like this work, a lot of people (teachers, parents, children, others) have to learn a lot. Learning involves change: change in neural circuitry, in behavior patterns, etc.). But wait! Conservatives, who largely rule our country, tend to be opposed to change (I checked in my American Heritage). Therefore, conservatives tend to oppose learning!! (I don't think I need to defend this point too strongly, given recent postings here and over on sci.anthropology newsgroup, not to mention Gingrich's "history course".) This means it will not be easy, and in fact we have seen how the buzzards start circling as soon as there's a slip-up ("genetically based"?!?- Yikes!).