Relating to this question about what tag to use for questions about acceptability, I thought I would ask what is the best way to answer these kinds of questions, so as to address both prescriptive and descriptive viewpoints comprehensively.

What issues should good answers to questions of acceptability address?

3 Answers 3


For any question of acceptability I think it is important to cover both descriptive and prescriptive viewpoints, regardless of the views of the answer’s author.

I would suggest that really good answers to questions of usage would address all these issues:

  • Would anyone have a basis for calling a usage wrong, based on criticism from significant grammar commentators?
    • If so, is the justification for calling it wrong reasonable and shared by many people?
    • What counter-argument is there, if any, for calling it right?
    • If not, why was the question asker concerned about it?
  • Is the usage in common contemporary use?
    • If so, is it commonly used only in certain levels or formality or topics of discourse?
    • If not, is it because it used to be common but has fallen out of favor, or was the usage always marginal?

Other interesting questions that answers might address are:

  • How has usage varied over time?
    • Is the questioned usage recent or old?
    • Is the questioned usage on the rise, on the decline, or in a steady state?
  • What alternative usages are there?
    • How does frequency of the questioned usage compare to suggested alternatives?
    • Are suggested alternatives more common at different formality levels or in different topics of discourse?

My own policy about "acceptability" is that it's best to try to answer these questions in the context of what the person asking the question actually wants. It's quite likely that someone learning English as a foreign language would not be at all happy to get an answer intended for someone studying linguistics, and vice-versa. I'd also be careful about overloading answers with information.

  • 3
    I think it is important to remember that the answers are only secondarily for the person who asked the question. They are primarily for the legions of future visitors who will stumble upon the page from a search engine results page, and if answers only narrowly answer exactly the question posed, they are less likely to be useful to someone who finds the page with a similar but not identical question.
    – nohat Mod
    Sep 18, 2010 at 18:33
  • It's too easy to get carried away with that kind of notion. I'm not sure we need to consider these legions when answering questions and it seems to be leading to people trying too hard to be clever.
    – delete
    Sep 18, 2010 at 22:20

I have felt for a while that we need a canonical question/answer (here, or in the FAQ) to which to point people who want to talk in quasi-moralistic terms about things being 'right' or 'wrong' (as opposed to 'common and understandable' or 'weird and downright ugly').

I say this because I think that lots of questions from people with English as a second language* tend to be looking for an equivalent 'authority' to one in their mother tongue. (I'm thinking of French in particular, with the Académie Française, which pretty much decides what is French or isn't).

I don't think it's helpful to these people to perpetuate the illusion that such an authority exists in English.

As far as I am concerned, there is nothing subjective about this. You can love it or bemoan it: language evolves and you can't stop it.

Does that mean that anyone can just arbitrarily decide to change the meaning or spelling of a word or invent new ones? (panic!!!!) Well, yes, I guess it does; though at the risk of being completely incomprehensible (Finnegans Wake anyone?). Apparently teenage girls are doing it all the time.

So I'll take nohat's COCA statistics anytime, rather than some prescriptive advice from a decades-old book written by some (cough) grammar nazi.

Talking of statistics, I think it would also be good to have some canonical question/answer on using Google fights as tests of common usage. Maybe nohat could include some info about getting the best out of COCA, and other alternatives.

*(Maybe also some with English as a first language but with memories of particularly prescriptive teachers.)


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .