I come to you with a deep sense of confusion. I read today’s post at the After Deadline blog at the New York Times, and the crux of it is that there is a subtle difference between might and may when they’re expressing possibilities. For instance, compare (1a) and (1b):

(1a) I may build a gravy fountain.
(1b) I might build a gravy fountain.

I’m not sure I see a significant difference between these two sentences. I believe I use may and might more or less interchangeably, and I suspect I use might more than I use may. Then again, maybe I don’t; I was about to write a sentence starting “post-modernists might argue X”, and in that sentence I would definitely, at the present moment, find may a poor substitute. This may be due to my relative certainty in the sentence; I am saying that I suspect that the post-modernists would argue X, not merely that they could argue X. And then, in the sentence I just wrote, I strongly prefer may to might, because I am less sure that the relative certainty is to blame. So, in summary, if there is a sustained difference between may and might for me, it’s that may expresses less certainty than might does — but I am not confident that I or the average English speaker consistently distinguishes between the two.

Some of my doubt has been sown by Philip Corbett, the After Deadline blogger, who favors precisely the opposite distinction. He writes:

“Trouble arises mainly when “may” and “might” convey possibility. Both words can carry this meaning, but there’s a difference in nuance. “May” simply states the possibility or likelihood, while “might” emphasizes the conditional nature of the possibility, introducing a greater level of uncertainty.

He may go to the theater tonight (stating the possibility).
He might go to the theater tonight (raising some doubt).”

It sounds in his post like Corbett has run an informal poll of some of the Times’ editors before reaching this conclusion, so I’m inclined to give it some credence. But, out of competitiveness and a gnawing curiosity if my usage is exactly backward, I wonder, first, whether there is any consensus amongst English speakers on this matter, and second, which one of us is closer to the consensus if it exists. What do you think? Is there a difference between may and might in sentences like these? If there is, is it one of likelihood or some other dimension(s)? How separated, if at all, are the two in your mind?

Post your opinion in the comments, and I’ll compile the results at the end of the week in a follow-up post. Then we’ll see whether it’s me or Corbett who has his finger well off the pulse of modern usage. (Of course, if it is me, I’ll bury the results in some rambling exposition on case assignment so that no one will ever find it.)

[Update: I forgot to add the link to the follow-up post. Here it is.]