the thin line between error and mere variation II

[Arnold M. Zwicky]

On Jun 29, 2004, at 7:55 PM, Wilson Gray notes the following phenomenon:

> There are people who, in their unmonitored speech, always mispronounce
> a
> given (class of) word. However, when this mispronunciation is called to
> their attention, they deny that said mispronunciation is part of their
> idiolect and "demonstrate" this by giving the word in question its
> standard pronunciation. Then they go right back to their idiosyncratic
> pronunciation. E.g.
>
> A. I'm goin' up the skreek. You want anything?
> B. Do you know that you always say "skreek" instead of "street"?
> A. (Annoyed) What the hell are you talkin' about? I don't say "skreek"!
> I say "street"!
> B. Oh. Okay. My bad.
> A. Like I said, I'm goin' up the skreek. You want anything?

the generalization is that, in a great many settings, people tend to
believe that they say what they think they're supposed to say.  it's a
species of earnest self-delusion.

so wilson's example is a lot like people's insisting that they never
"drop their g's" -- and then, of course, do so as soon as their
attention is focused elsewhere.

some years back we had a discussion here about r-omission by otherwise
rful speakers, and i noted that lots of people, a fair amount of the
time, didn't have an r in the first syllable of "quarter".  as the
discussion went on, people who protested that they'd never heard of
such a thing, etc. suddenly caught *themselves* saying "quater"
(exactly the response i had when my first linguistics teacher, the
classicist samuel atkins of princeton, pointed out that this was my
usual pronunciation).

more recently I've had a certain number of excellent writers (like
louis menand) and linguists (like larry trask) and authorities on
writing claim to me that they never used possessive antecedents for
pronouns, though in fact they did, they did.

one of those things that can make self-reports, including acceptability
judgments, perilous.

arnold (zwicky at csli.stanford.edu)