Coding grammatical errors in language samples

[Beverly Flanigan]

At 02:09 PM 9/6/2002 +0100, George Hunt wrote:
>The forms could of, might of, should of etc are also common in adult
>writing, including that of education students [and linguistics students!].
>I had always assumed that
>the writers were simply making a graphophonic analogy with forms like kind
>of, sort of, some of, many of. When these are spoken, the vowel of the
>last syllable is reduced to schwa, making these syllables identical with
>those of the abbreviated modal forms. When these writers were mentally
>reconstructing the full forms for spelling, perhaps they drew upon the
>'kind of' forms. However, this does not explain the direction of the
>analogy - why don't we as often see forms like kind have, many have etc?
>
>It might be interesting to talk to the writers to see whether or not they
>are making conscious choices between possible spellings in the process of
>writing these non-standard forms. I once tried to correct a student by
>pointing out the difference between the full forms of the modal
>constructions, and contrasting them with kind of etc. She replied, 'I know
>that, but in my dialect the full form is should of'.

"I know that" is the key.  At some underlying level, I suspect most people
know the difference between 'have' and 'of' in these phrasal sets.  Your
student may not understand what "full form" means, and she's confusing
dialect and style.  A way to test this is to ask the writer to turn the
statement into a question (cf. Labov's test of black kids' underlying
knowledge of the copula).  Similarly, if someone is asked what "your
welcome," or "your tired," or "your not my boss" means, I suspect they will
at some point articulate the copula fully.

Beverly Olson Flanigan
Ohio University