Friday, March 16, 2012

The life cycle of a peeve

There's an interesting report in tomorrow's Wall Street Journal by my former Globe Ideas colleague, Christopher Shea, about a word study based on Google Books (and published in Science Scientific Reports*). The paragraph that caught my eye:
The authors even identified a universal "tipping point" in the life cycle of new words: Roughly 30 to 50 years after their birth, they either enter the long-term lexicon or tumble off a cliff into disuse. The authors suggest that this may be because that stretch of decades marks the point when dictionary makers approve or disapprove new candidates for inclusion. Or perhaps it's generational turnover: Children accept or reject their parents' coinages.
The dictionary hypothesis seems unlikely to me, a confusion of cause and effect; how many people consult the dictionary for permission to add a word (or new usage) to their lexicon? But in my years as a language columnist, I often told peevers to be patient, because every innovation they hated would either disappear or become inoffensive, sooner or later. (The category that Bryan Garner calls "skunked terms" would be the outliers that remain open wounds -- because of all the devoted scab-pickers? -- far longer than the norm.)

I've often wondered why some changes attract disproportionate ire while others blend into the language without opposition. A century ago, Ambrose Bierce and his crowd were sweating the distinction between admission and admittance, proposition and proposal, necessaries and necessities -- all forgotten now. But their worrying about anxious vs. eager, try and vs. try to, and figurative literally live on. I doubt that we'll actually find out why some peeves die and others flourish, but it's going to be fun trying.

*Update 3/17: The WSJ credited the paper to Science, but Mark Liberman's post today at Language Log correctly cites Scientific Reports.

12 comments:

Kay L. Davies said...

Hoorah for your headline. I love it. Not only that, I understood it.
What if I outlive my peeves? Is that good? Or if they outlive me? Is there no hope for me then?
Words, all words. Such fun.
K

Ø said...

Here's a good exampl (a headline on page A6 of today's edition of your own Boston Globe) of why the peevers who sweat distinctions sometimes have a point:

"Gingrich withdrawal may not aid Santorum"

I had to do a double take. Has Gingrich withdrawn? No, read "might" for "may".

Ø said...

I'm not suggesting that therefore the may/might peeve is likely to live longer than, say, the fewer/less peeve, where there is little or no danger of confusion from the loss of distinction. But I think that may/might deserves to live. I could imagine caring much less about this.

I might even write a letter to the editor. Or do I mean "may"?

Jan said...

Ø, I had the same problem with that Globe headline; I thought Gingrich had dropped out and I'd somehow missed the news. Like you, I have the may/might distinction deeply embedded -- so deeply that I truly don't understand how so many people can't hear it. And yet, despite its usefulness in distinguishing fact from possibility, it seems to be disappearing rapidly. I don't think I go a week without seeing "may" (in a reasonably respectable publication) where my dialect requires "might."

Gregory Lee said...

I could follow this discussion better if I understood the difference you find, or want to find, between "may" and "might".

Ø said...

The more I look at it, the harder it is for me to explain myself. I haven't really got a good handle on this.

But to me, "The Gingrich withdrawal might not aid Santorum" and "The Gingrich withdrawal may not aid Santorum" sound very nearly synonymous.

And "A Gingrich withdrawal might not aid Santorum" and "A Gingrich withdrawal may not aid Santorum" do not sound synonymous. The latter is more ambiguous as to whether Gingrich is known to be withdrawing or whether one is speaking in a "what if" way..

The lack of article in the headlines "Gingrich withdrawal might not aid Santorum" and "Gingrich withdrawal may not aid Santorum" leads to even more amibiguity.

Are the following all grammatical, and how do they differ in meaning?

"If Gingrich withdrew, that may not aid Santorum."

"If Gingrich withdraws, that may not aid Santorum."

"A possible Gingrich withdrawal may not aid Santorum."

"A Gingrich withdrawal may not aid Santorum."

Gregory Lee said...

I have a clear intuition about the example ""If Gingrich withdrew, that may not aid Santorum.". Which is that it is not acceptable. The "if" part is counterfactual, and means the same as "if Gingrich were to withdraw", and that assumes that he has not withdrawn. The "then" part is equivalent to "it is possible that the withdrawal will not aid Santorum", which is equally unacceptable. Instead, a subjunctive "would" form is needed: "it is possible that the withdrawal would not aid Santorum", which in turn is like "that (= the withdrawal) might not aid Santorum.

So, just considering this example, it appears that "may" is equivalent to "possibly will" and "might" is equivalent to "possibly would".

Marc Leavitt said...

A good example of a peeve that's more or less accepted now is "normalcy," attributed to President Harding back in 1920, as opposed to the "preferred" normality. Unless I miss my guess, and I haven't checked the corpora, the former is now probably more common.

Bryan White said...

It's funny. I was going to say that whether you use "may" or "might", the sentence would still seem to give me the false impression that Gingrich's withdrawal was certain, but I didn't want to look like a fool again after the last post. I'm glad to see that I'm not alone.

Obviously, newspapers word things this way for the sake of economy, but sometimes I think these newspapers almost make these things deliberately misleading to grab your attention. "What!? Gingrich is withdrawing!!??? Well, look like I'm gonna have buy the paper and read up on this over breakfast."

Ø said...

Yeah, except it wasn't on page 1.

I did write the letter.

Bryan White said...

Hmmm, not being on the front page does kind of kill my theory :)

Unknown said...

Yule and other linguists talk about the "past" forms as being remote, which with the modals leads to unlikely or irrealis readings.
So, "I will do it" presents the act as quite likely while "I would do it" presents the act as unlikely or contingent upon some other event.
Is the same true of may and might? I think that "it may happen" and "it might happen" retain only a hint of that distinction because they both suggest that the likelihood is sort of 50/50.
But the larger problem with "may help" and "might help" is that the distinction has to do with what happens to Santorum. The modals don't actually say anything about what Gingrich did.
As others have mentioned, a definite/indefinite NP would clear things up - "a Gingrich withdrawal" says he hasn't withdrawn while "the Gingrich withdrawal" suggests that he has.