The Word

Mighty likely

Might vs. may: A battle of possibilities

By Jan Freeman
September 13, 2009

E-mail this article

Invalid email address
Invalid email address

Sending your article

Your article has been sent.

  • Email|
  • Print|
  • Reprints|
  • |
Text size +

“I may finish the project today.” “I might finish the project today.” Which one is a more optimistic prediction?

The standard line, repeated in usage books for some 200 years, is that “I may finish” implies a greater likelihood than “I might finish.” In fact, some critics have drawn a sharp distinction: In his 1839 handbook, optimistically titled “The Difficulties of English Grammar and Punctuation Removed,” John Best Davidson explained that “I may visit London this summer” means “it is probable I shall,” while “I might visit London” means “it is not probable.” As late as 1957, Bergen Evans, in “A Dictionary of Contemporary American Usage,” concurred: “I may” means it’s possible, he said, and “I might” means it’s not likely.

Not everyone agreed. Josephine Turck Baker, in her 1907 guide, “Correct English,” said there was “no essential difference” in meaning between “I may go” and “I might go” (though one had to use might in the past — “He might have gone”). And over the past half-century, even usagists who observe the distinction have softened it: Bryan Garner, in the new “Garner’s Modern American Usage,” says merely that “may expresses likelihood … while might expresses a stronger sense of doubt.”

This may (or might) be a real distinction. But how would we know? The usage mavens don’t say, but to Gabe Doyle, who blogs at Motivated Grammar, that was the obvious puzzle. He was baffled, he writes, by a post at the New York Times grammar blog, After Deadline, asserting that may was the more optimistic word. Not for him: In fact, he said, “if there is a sustained difference between may and might for me, it’s that may expresses less certainty than might does.”

So Doyle asked his readers: In their usage, did “I may” or “I might” (or neither) express a greater likelihood? Three commenters agreed with the traditional notion: “I may go” is more promising than “I might go.” Two thought may and might differed in formality — but they disagreed on which verb was more formal. And “at least three agreed … that there wasn’t any clear difference.” The words, he concludes, are “essentially interchangeable, because different people don’t agree on what the difference between them would be.”

At least one usage authority noticed this several decades ago. In his 1980 usage guide, “Words on Words,” John Bremner observed that though “some lexicographers” thought may was stronger than might, “If such distinction exists in common language, the distinction is even thinner than nuance.”

And once you take a skeptical view, it’s clear that he and Doyle are correct. When I say “I may finish today,” you don’t know whether I’m using may as a stronger alternative to might, and I don’t know whether you hear it that way. May and might, in this use, are entirely subjective; there’s no way you can say that I’ve chosen the “wrong” word.

That doesn’t mean that may and might have lost all power to perplex. In recent decades, critics have begun to take note of may’s substitution for might in past-tense counterfactuals: “If he’d survived, he may have been a contender.”

Unlike the may and might of subjective possibility, this usage can still be marked wrong. Even the infinitely tolerant Merriam-Webster Dictionary of English Usage doesn’t approve: “We advise you to use might in all contexts where the past tense is appropriate or where a hypothetical or highly unlikely situation is being referred to.” If you know he didn’t survive, you can only say what he might have done; he no longer has access to the world of may.

. . .

CLOTTED CREAM: Last week, in a discussion of charter schools on WBUR’s “Radio Boston,” at least two people used the phrase “cream the crop,” meaning “take the best students from public schools.”

Now we all know “skimming the cream,” even if we’ve never seen unhomogenized milk. And “cream of the crop” is just one metaphorical application of cream, which has meant “the best part” since the 1500s. But “creaming the crop” was new to me.

That’s not to say it’s new to you. In a quick Google Books search, I found an example of “creaming the crop” dating to 1976. And cream as a verb is also kosher: “to cream off,” meaning “skim off the top,” dates to the 1600s and remains in use in Britain: “Arsenal can cream off all the young talent from across Europe, but the top youngsters are never going to go to Stoke or Fulham.”

So “cream the crop” isn’t outlandish. But it may be a sign that the figurative cream is going sour. Do people using “cream the crop” still connect it with “skimming the cream”? Or is the blended phrase a sign that progress, this time in the form of homogenization, has killed off one more metaphor?

E-mail Jan Freeman at For past columns, go to