Tim,
   You do put it very succinctly: ":more than just a series of logical
relationships of word to word to word."  I could give you a lengthy
answer, but I'll try to be brief.
    If we start from the sentence, we can measure it against norms of
grammaticality>(whether it follows the natural sentence generating
rules of the language) or against norms of "correctness" (the usual
handbook stuff.) If we want to tell whether the sentence is effective,
however, we need to look at its discourse role.  And from that
perspective, it is possible to say at times that a sentence is
ungrammatical or incorrect or both, but highly effective. They are
different lenses.
   A lot of the quarrelling that goes on about grammar comes from the war
between descriptive and prescriptive camps, largely those who see the
rule driven nature of all language and those who think of nature as in
need of begin controlled.  But neither of these approaches can work in
harmony with composition, which inevitably needs to view revision as
context based.
   How does this work out in practice? One starting point is the notion
that  sentence is not a "complete thought", but a move in a series of
related moves.  Sentence boundaries are very flexible, and a sentence
can vary widely in the amount of information it contains and in the way
that information is parceled out or organized. Without a carryover of
meaning from one sentence to another, there is no coherence or
cohesion. Insofar as a text is a succession of complete, discreet
thoughts, it is almost impossible to follow. Most student writing
suffers from a decontextualized view of the sentence (lack of coherence
and cohesion), and "correcting" those sentences does not fix the larger
problems of the text. A whole other frame of reference comes into play,
some of that somewhat unfamiliar to many of us (given and new, theme
and rheme, intonation units, tonic prominence, nominalization patterns,
grammatical metaphor, and so on.) But The main idea is that you work
top down, asking yourself what sort of changes you would like to make
to a text (less argumentative, clearer, less stress on A and more on B,
and so on) and then look for the adjustments that would carry that out.
You don't just CORRECT sentences, but REVISE them, and grammar has an
enormously important (absolutely inevitable) role in all of that.
     Fish is not the first or last to see a radical split between content
and form. You can't solve that problem by choosing one side over the
other, though you can sustain a career by that kind of grandstanding.
The problem with form is that it CAN and, in fact, usually is
DECONTEXTUALIZED. The split between grammar and rhetoric will never
heal until we bring them back together.
    Like you, I have deep suspicions about the studies that show grammar
doesn't carry over to writing. You can make a good case that teaching
writing doesn't carry over to writing, depending on what approach is
being used. Conformity to decontextualized notions of form is not what
makes good writing effective, and that's as true at the level of the
paragraph as that is at the level of the sentence.  Ultimately, one
would hope that those choices are made in harmony with an evolving
purpose and in harmony with each other.
    I hope that makes sense.

Craig
Thanks, Craig,
>
> You're right-I probably over-do it, partly because I have been shocked
> to discover, in my recent dissertation research, how much that I thought
> I "knew" about this subject is not at all true. And I fully agree with
> your very appropriate emphasis on discourse, though one might argue (and
> some have) that meaningful discourse is built on meaningful sentences.
>
> I am curious, then, about the "much richer view of language" that you
> see coming from a discourse-based approach. Do you mean simply in the
> sense that it is more than just a series of logical relationships of
> word to word to word?
>
> Thanks,
>
> Tim
>
> Tim Hadley
> Research Assistant, The Graduate School
> Ph.D. candidate, Technical Communication and Rhetoric
> Texas Tech University
>
> -----Original Message-----
> From: Assembly for the Teaching of English Grammar
> [mailto:[log in to unmask]] On Behalf Of Craig Hancock
> Sent: Wednesday, June 01, 2005 8:57 AM
> To: [log in to unmask]
> Subject: Re: Response to Fish piece
>
> Tim,
>     Thanks much for once again making this important point concerning
> the lack of "solid research against grammar teaching."   One key, I
> think, is determining what  we mean or ought to  mean by "functional",
> "contextual", and "sentence based."  If you are right, then Fish's
> notion that we can understand grammar as a system of "logical
> relationships" and a system of forms outside the give and  take of real
> world application (content?) is wrong.  For that reason, I would take
> issue with "sentence based" and replace it with "discourse based," since
> so much of  the decision making that occurs at the sentence level is and
> ought to be discourse sensitive. We also find that far more than
> "logical relationships" (Fish's terms) are at play. In other words, Fish
> inherits students who know nothing about language and tries to remedy
> that in a short space. But I think it's wrong not to at least put that
> to practical use in the interpretation and creation of text, if not in
> his course, then in a following one. And when you do, a much richer view
> of language will come out of it.
>
> Craig
>
> Hadley, Tim wrote:
>
>>Point of order (or whatever type of point this should be) regarding the
> existence of "solid research against grammar teaching":
>>
>>In this discussion, precision in terminology is extremely important. As
> a matter of fact, there has never been any solid research against
> "grammar teaching." There has been, instead, some research (I won't get
> into the debate right now about whether it was "solid") against the
> teaching of "formal," "traditional," "linguistic," "Latin-based,"
> "non-contextual" (etc.--you get the point) grammar--which is an approach
> to grammar teaching that is more or less the opposite of a functional,
> contextual, sentence-based approach to grammar. The former (formal
> grammar) is believed, on the basis of some research, to be non-helpful
> to the improvement of writing. The latter (functional grammar) is now,
> and has been for more than 100 years, almost universally recognized as
> helpful, appropriate, and necessary to the improvement of writing.
>>
>>It is unfortunate that this distinction is often not made clear when
> this subject is discussed, and that many people, even among the English
> establishment, continue to believe that research has "proven" that
> teaching "grammar" (any and all grammar) is harmful.
>>
>>Tim
>>
>>Tim Hadley
>>Research Assistant, The Graduate School
>>Ph.D. candidate, Technical Communication and Rhetoric
>>Texas Tech University
>>
>>________________________________
>>
>>From: Assembly for the Teaching of English Grammar on behalf of John E.
> Dews
>>Sent: Tue 5/31/2005 12:24 PM
>>To: [log in to unmask]
>>Subject: Response to Fish piece
>>
>>
>>
>>    Below is a response to the Fish op-ed piece I forwarded to the list
> earlier. There are already strong feelings against his methods -- I've
> already heard one person today complain that he is ignoring all the
> "solid research against grammar teaching."
>>
>>
>>To join or leave this LISTSERV list, please visit the list's web
> interface at:
>>     http://listserv.muohio.edu/archives/ateg.html
>>and select "Join or leave the list"
>>
>>Visit ATEG's web site at http://ateg.org/
>>
>>
>>
>
> To join or leave this LISTSERV list, please visit the list's web
> interface at:
>      http://listserv.muohio.edu/archives/ateg.html
> and select "Join or leave the list"
>
> Visit ATEG's web site at http://ateg.org/
>
>
> To join or leave this LISTSERV list, please visit the list's web interface
> at:
>      http://listserv.muohio.edu/archives/ateg.html
> and select "Join or leave the list"
>
> Visit ATEG's web site at http://ateg.org/
>

To join or leave this LISTSERV list, please visit the list's web interface at:
     http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"

Visit ATEG's web site at http://ateg.org/