Grammar Nazis and deflected/projected racism

My mother, who was very racist but sincerely believed herself to be not racist, said that she was not personally opposed to intermarriage, but she was opposed to it, on the grounds that it was so hard on the children. In other words, she supported a racist practice (social shaming of “intermarriage”) while still feeling herself not racist because she could tell herself that her racist practice was necessitated by the racism of other people.

Teachers—all teachers, at every level—are far too often my mother. We teach in a racist way, all the while claiming that we, personally, aren’t racist, but our racist practice is necessary because of the racism of others. We do it when it comes to teaching “standard Edited American English” (a particular dialect) as though it is better than other dialects.

English has a lot of different dialects, and many of those dialects are grammatically different. Standard Edited American English (SEAE), for instance (a dialect no one speaks), prohibits the comma splice (The cats ran, the dogs barked), but Standard Edited British English doesn’t. In spoken English, sentence fragments are fine, and are also fine in much published writing (depending on formality), but generally prohibited in very formal writing (except resumes or cv, where they are required). It would be inappropriate for someone to use full sentences in a resume, and therefore equally inappropriate for someone to mark a resume as “wrong” for using sentence fragments. Sentence fragments aren’t therefore “worse” than complete sentences–they’re appropriate or not; that’s how language works.

However, in any language there are dialects that are stigmatized for racist, classist, historical, or various other bigoted reasons. They’re stigmatized as “bad” English (or French, or German, or whatever). In American English, one use of the double negative is stigmatized and the other accepted because one is associated with Black English. “She don’t know nuthin’ about nuthin’” is a perfectly clear sentence, but “The argument is not unclear” takes math to understand. Yet, it’s the first that gets called “bad English.” (Which is funny, if you think about it–calling something “bad English” is itself an instance of using the wrong term, so it’s “worse” English than a double negative.)

So, it’s important to separate out two kinds of grammatical errors: a violation of a dialect from within that dialect (such as someone trying to write SEAE who violates rules of that dialect, such as the muddled Black English of The Help), ones that are correct usage within that dialect but not accepted in the dialect a reader is expecting. (A third category would be uses of language that aren’t grammatically incorrect at all, but people think they are–ending a sentence with a preposition, for instance.)

Here’s what I mean by the second kind of error. It would be bizarre for someone to chastise someone speaking German for ending a sentence with a preposition—that’s how German works. (It’s also how English works, but that’s a different post.) It would also be sheer bigotry to say that French is better than German because French doesn’t allow ending sentences with prepositions. Dialects and languages are all equally good at communicating; none is better than another.

I’ll mention something about the first toward the end of this post, but, for the most part, I want to focus on what we do about stigmatized dialects. The problem is that, since, for instance, Black English is stigmatized, and Standard Edited American English is rewarded, should teachers require that their students learn Standard Edited American English?

The advice for years (ever since the National Council of Teachers of English and Conference on College Composition and Communication issued the Statement on Students’ Right to Their Own Language”) has been to advocate code-switching. To say that a student should know SEAE because it’s useful, not because it’s better, is like saying that it’s useful to know French if you intend to live and work in France. From within this model, German is no better than French (nor is French better than German), and a student might be speaking perfect German in a French class. A person shouldn’t give up German, but add on the knowledge of French. Students should learn SEAE as an additional dialect that is useful under some circumstances.[1]

Unfortunately, too many teachers and professors and employers and people in power use the language of code-switching in order to enforce the message that Black English is inferior.

A few years ago I found myself in an argument on the internet with a white teacher in a predominantly African American school who banned Black English in her classes. She was proud that she told her students that Black English would hold them back. She wasn’t racist, she insisted; she was helping them. There’s what might seem like a subtle difference between what she was doing and what “Students’ Right to Their Own Language” advocates, but it’s an important one. She was clear that SEAE was better than Black English, that Black English was something they should be shamed for using. I then noticed that I often had the same problem with training people in the teaching of writing—they made a bigger deal about perfectly clear uses of stigmatized language than they did about about grammar problems that interfered with communication. They did so, they said, because other people would be racist.

It’s my mother opposing “intermarriage” because other people would be racist. That’s racist.

Granted, we’re in a racist world, and using a stigmatized dialect will hurt a person in terms of job or housing applications, getting good scores on standardized tests, or dealing with racist teachers who deflect their racism onto others who might be racist. So, I understand, and still support, the idea that we should teach code-switching, but if (and only if) we give students the ability to choose whether they want to learn to code-switch, we do so by making it absolutely clear that no dialect is better than another, and we make a bigger deal about violations of grammar and usage within (rather than across) dialects. I don’t know that we can do the second, and if that’s the case, then teaching code-switching is racist.

I mentioned that violations within a dialect are worth looking at carefully, largely because they can signal issues with thinking. For instance, mixing metaphors can indicate that we haven’t decided on the underlying model, or that we’re appealing to troubling models, or that we just aren’t thinking. I once heard a facilitator say, “We’re on a fast train flying out of the box.” She was describing a train wreck, as far as I could tell, but I think she meant it as a good thing. I don’t know. Had she said, “We ain’t done nothin’ about nothin’” I could have understood her perfectly.

Unclear pronoun reference can mean we haven’t really decided how causality works. For instance, if I say, “There are bunnies eating kale in the backyard, which is weird,” it isn’t clear whether the weird part is that there are bunnies, that they’re eating kale, that they’re doing it in the backyard.” In other words, it isn’t clear what “which” is referring to. What’s interesting to me about these sorts of errors (predication error or mixed construction is another one along these lines) is that “correcting” the error means first figuring out what I’m trying to say. These are interesting and significant errors.

Whenever I get into this topic (or when it comes up even on scholarly mailing lists), people advocating my position (the position of most if not all linguists, btw) get accused of thinking that anything goes, and that we shouldn’t care about clarity or correctness of any kind. That isn’t what I’m saying. I’m making four points. First, no dialect is better than any other (it might be more useful, inappropriate, effective under certain circumstances). Second, what grammar Nazis worry about are often not “grammar” issues at all (but style preferences, hypercorrectness, misunderstandings of rules, misapplications of rules), and are almost always not issues of clarity, but are class or race markers (e.g., comma splices, double negatives, subject-verb agreement, ending with a preposition). Third, we should worry about certain issues of usage, but it should be the ones that are violations within a dialect, especially ones that signal muddled thinking. Fourth, the conventional wisdom among experts for years has been that we should teach code-switching (that is, the ability to switch between dialects), but that’s still racist unless we do so in a way that makes it clear that we aren’t privileging one dialect over another, and we offer it as a choice to students.


[1] Another way to put this is to say that prescriptivism is perfectly fine, as long as it’s taught qua prescriptivism.