When the Stanford sociolinguist Penelope Eckert read my “Lying About Writing” post, she was just approaching the end of her writing-in-the-major course, so she already had her mind on what undergraduates need to learn about writing well. Agreeing with my criticisms of a silly list of don’t-do-this maxims handed out by an unidentified English department, she commented that “this kind of no-no advice is not just stupid but de-skilling”—because “the question is never whether, but when.”

That hits the nail on the head, in just seven words. I wish I could put things so incisively.

Grammatically, whether-clauses are closed interrogatives, expressing the content of questions with fixed and usually very short lists of potential answers: whether it’s raining or not, whether tenure was or was not approved, whether Ljubljana is in Slavonia or Slovenia or Slovakia—choices from a stipulated list, frequently including just yes and no.

By contrast, open interrogatives contain words like who, what, which, when, where, why, or how, and express questions with wide-open ranges of potential answers: what kind of animal that is; who was really responsible for the banking crisis; when there will be peace in the Middle East.

Traditional grammar advice is mostly framed in terms of answers to closed interrogatives: No, never split an infinitive. No, never use negative auxiliary verbs like can’t or don’t. No, don’t ever begin sentences with but.

Eckert’s insight is that students need answers to open interrogatives about when to do what, and why, and how. She cites the example of less with count nouns (as in “there were less errors than in the control condition”), which has been steadily gaining popularity:

They need to know that less with a count noun is like nails on a blackboard to me, but that it’s clearly a change in progress and they have to decide where to place themselves with respect to it.

That where question is the key point. It’s all about where you want your writing to position you in the huge space between the stiff phraseology of a pompous old crank (“Have we to conclude we may communicate with whomever we please?”) and the racy, informal, chat of a talk-radio host (“So we can just talk to whoever the hell we wanna talk to?”). There are degrees of formality and informality, and linguistic changes in progress, and signals by which you can show that you are leaning one way or another.

Neither stranding prepositions (“the conclusion that we arrived at“) nor dragging them to the front of the clause (“the conclusion at which we arrived”) should be an invariant practice; you need to decide when it sounds right and when it doesn’t. You have to steer between the Scylla of sounding like a tweet by an excited high-schooler and the Charybdis of sounding like a great-grandfather in a starched collar.

I clearly veered too close to the latter when I was an undergraduate. I only ever got one piece of direct advice about my clueless freshman writing: After reading my first embarrassingly pompous screed, my professor, David Reibel, told me: “You write like an 18th-century clergyman.” I could see that he did not intend this as praise, so I stopped doing it.

Eckert sees the other aspects of writing up results in an empirical discipline in a broadly similar way:

They need to know that writing up research is a semi-cynical exercise in erasure of one’s personal process: I do want to know how many tokens you had, but I don’t want to know that you put them in an Excel spreadsheet. You do want to tell your reader that you found that gender and age interact, but you don’t want to admit that you started out with no clue that they could.

When should you reveal something about the course of the investigation as opposed to just announcing its results, and why? Which aspects of the investigative process is it essential to mention, and which are superfluous? As with grammar or prose style, students have endless judgment calls to make about what and how they should write. Those calls turn on open questions, not closed ones.

Reducing open grammar and style questions (like “When does a modifying adverb improve a sentence?”) to closed ones (like “Should I ever use adverbs?”) is the fatal oversimplification made by the don’t-do-this-don’t-do-that brand of usage bullying that I often refer to as stupid grammar advice.