Consider a quote like "I am fascinated at1 the way some African-Americans speak to each other in an English I struggle to understand, then switch to standard English when the situation requires." Any linguist, or indeed any undergrad who took a sociolinguistics class for a gen-ed credit and stayed awake for the lecture on William Labov, would take one look at that and say, "Aha! Code switching!" Because that's what it is: changing registers at the drop of a hat, moving from a covert-prestige mode to an accepted prestige dialect when the situation occasions it. This is a long-studied and well-understood phenomenon.
But the undergrads who skipped that day, along with all the grammar noodges who perpetually aced English but never studied a day's worth of our fair discipline, say "Huh? No, it's just bad!"
(I doubt a single one of them -- Christie Vilsack, Rob's mother, my mother, William Safire, take your pick -- has ever been taken to task by a stodgy old BBC doctrinaire who cannot abide this horrifying American perversion of the Queen's English. But I digress.)
Meanwhile, folks like the article's author, David Guarino, get all up in arms (some of my rural Southern relatives might say 'het up') about this perceived slight on the character of blacks, Northeasterners, Southerners, and indeed anyone not from the Land of the Free and the Home of the Radio-Neutral Dialect, my adopted home, the Midwest. Vilsack is eeeee-vil, y'see, because her opinions on speech must necessarily extend to people, right? Right, and people are only Good if all their opinions about other people are Good and otherwise they must be vilified in public in the hopes that they will renounce their ways and become Good again. (Ahem.)
I submit that every group involved has a different definition of "good." The prescriptivists know what the liberals' definition of "good" is, and they think it's wrong, and vice versa. The problem we face is that jack no one but the linguists knows how we define "good" ... and I rather doubt anyone's willing to listen.
"Good," for a linguist, generally falls nicely in line with the idea of "robustness." Can a language, no matter what it sounds like, convey between its speakers all the ideas that a speaker might want to discuss? It can? Righto, then it's a good language. That's it, end of story. C++ is a great language for what it does, but it's a "bad" language qua human language (it is certainly not natural), because I cannot communicate in it anything I bloody well please.2
Parallel to this, I often hear linguists use "good", in reference to human language, as synonymous with "interesting." If a syntactic mechanism is particularly quirky and hard to understand, if a set of OT constraints are especially descriptive, elegant and simple, this is worthy of much excitement and praise. "Russian has a genitive of negation! This is great!" Which means there are plenty of linguists who get righteously wet over AAVE3 because of its neato syntactic and phonological similarities to Bantu languages.
It is precisely this sort of geekery that prescriptivists are not geared to understand. (I say this having grown up with English teachers for mothers and aunts, having known an awful lot of English teachers over the years, and having spent a number of years as a newspaper copy editor; perhaps the prescriptivists you know don't think this way, but the ones I know do.) Your average prescriptivist, in my experience, is looking for one simple stand-alone system with rules she can memorise, enumerate, and rely upon. Linguists, by contrast, thrive on synthesis. We think it's cool that there isn't just "one English." It's not about tolerance; it's not about diversity; it's about the "Oooh, neato!" factor.
But these politicians and journalists are Serious, you see, because on the one hand this is an Education Issue, and on the other it is a Race Issue, and no one has room for the gripping hand at all. There is no room for "Oooh, neato" in Serious Debate. And so dumb things get bandied about on the DNC floor, and everyone goes round and round in circles once again.
That said, I have no quibble with Rob's point, which is that it is a Valuable Thing when language is understandable. The difficult thing is that mutual intelligibility is often a dearer (and more confusing) thing than anyone expects. English speakers are actually lucky that our various dialects -- Wilde's remark about England and America being two nations separated by a common language notwithstanding -- are as mutually intelligible as they are. Good bloody luck being someone who understands Classical Arabic and trying to get along in Egyptian Arabic (FWIW, as I understand it, most of the modern dialects are by-and-large mutually unintelligible too), or knowing both Cantonese and Mandarin and finding yourself somewhere in the rural Chinese countryside. Not too long ago, I did a bunch of research on an Austronesian language called Rukai, which is a minority language in rural Taiwan; it's spoken by fewer than 8000 people and has three divergent dialects. To add even more confusion to the mix, you also have the politically split languages that are still mutually intelligible, e.g. Norwegian and Swedish or Serbian and Croatian (formerly Serbo-Croat, now divided by lines on a map and lots of guns). America is big, as in "you may think it's a long way down the road to the chemist" big, and it's not too surprising that some of our dialects may be difficult to recognise as belonging to the same language4 at all. So we do the best we can with what we have ... and sometimes, our best necessarily sucks.
But, hey, Christie? For what it's worth, code-switching fascinates the living hell out of me too. Screw Guarino and his types; you go on being fascinated, okay? Watch it, listen to it, study it, have fun with it.
And if you happen to find yourself fascinated by other aspects of speech and human behaviour, then I can happily promise you, we linguists will be waiting to welcome you aboard.
1 Perhaps this is a regionalism on my part, but I've always heard "fascinated with" or "fascinated by," never "fascinated at."
2 Someday real soon I should talk about the idea I had over dinner tonight about Universal Turing Machines, the halting problem, using natural language to solve just such thorny computational problems, and why this probably means that attempting to implement a perfect parser on a Von Neumann architecture is doomed, doomed, doomed.
3 African-American Vernacular English.
4 Don't believe me? Have a look at Gullah.