"Hit 'em where they ain't" (1901)

[Laurence Horn]

At 2:07 PM -0400 5/11/05, Wilson Gray wrote:
>...> _Brooklyn Eagle_, July 29, 1901, p. 11, col. 3
>>  Willie Keeler is always prepared for anything that comes along. Yesterday
>>  a Philadelphia fan approached him and inquired what was the principal
>>  attribute of a successful batter.
>
>Do what?
>
>(a) "...inquired what was the principal attribute of a successful batter."
>
>and not
>
>(b) "...inquired what the principal attribute of a successful batter was"?!
>
>What's up with this? Isn't (b) the "correct" word order? Hasn't no
>less a light than William Labov stated that the (a) form is peculiar
>to the currently so-called "AAVE"? Or am I simply terribly, terribly
>confused? Naturally, I've seen and heard the (a) word order here,
>there, and everywhere, here of late. But that can be accounted for by
>ordinary language change since Labov made his statement. But, if Labov
>is or was correct about (a), why is it found in a white newspaper
>published before he was born? Did the (a) type die out among white
>speakers, only to be resurrected by black influence as a result of
>creeping desegregation? Or is my memory simply fubar?
>
>-Wilson Gray

Inversion in an embedded question is non-standard, but to my
knowledge not exclusively AAVE.  I remember a paper by Haj Ross,
"Where to Do Things With Words" (if memory serves; my copy is in my
office and I'm not), from _Syntax and Semantics 3: Speech Acts_
(1975, P. Cole & J. Morgan, eds.), which dealt in some detail with
the circumstances under which inversion does and does not occur for
speakers of certain non-AAVE dialects.  The generalization seemed to
be that if the speaker was intending the utterance as an indirect
question or the report of one, inversion occurs, so e.g.

He asked me where is the bathroom.
*He told me where is the bathroom.
She wanted to know where is the bathroom.
I'd like to know where is the bathroom.
*She didn't care to know where is the bathroom.

It was more complicated than this, and had to do with supporting an
argument for the performative hypothesis, but anyway the assumption
was that for Ross (and George Lakoff, who supplied much of the data;
this was the era in which they were still basically twinned)
inversion is fine in colloquial speech as long as there's a question
(and hence lack of knowledge) on the part of the subject of the
relevant verb.  The (a) example above, for the reporter playing
straight man to Wee Willie, fits the pattern.   (I don't think the
inversion in such cases ever died out, but it has remained
underground, impervious to prescriptive edicts.)

Larry