"teaching grammar doesn't work"

Camped out in the faculty lounge post-Sandy, reading articles on "basic writing":

At a recent workshop for high school and community college teachers, an earnest young high school teacher explained forcefully to an experienced community college teacher that grammar was of no use in teaching writing. The high school teacher cited the now-famous Braddock, Lloyd-Jones, and Schoer quotation. She said that knowing grammar had no effect on writing ability, insisting that "all the research" counterbalanced any intuitive and experiential evidence the older teacher might have to offer. The young teacher had, however, misquoted the passage; it says: "the teaching of formal [emphasis ours] grammar has a negligible or, because it usually displaces some instruction and practice in composi- tion, even a harmful effect on the improvement of writing" (37-38).

Taking the words teaching of formal grammar to mean knowing grammar is a serious mistake. What the research cited by Braddock et al., indicates is that instruction in traditional grammar over a limited period of time (a semester or less in the research studies being discussed) showed no positive effect on students' writing. In fact, several research studies and much language and composition theory argue for certain types of grammar instruction, when effective methods are used for clearly defined purposes. When writers learn grammar, as opposed to teachers merely "covering" it, the newly acquired knowledge contributes to writing ability.

In separate essays on grammar, both Kolin (139) and Neuleib (148) point out that the often-quoted passage in Braddock et al. was preceded by "Uncommon, however, is carefully conducted research which studies composition over an extended period of time" (37). Few people seem to pay attention to the qualification, however. Also, another 1963 study, one that Kolin reviews, has attracted much less notice than Research in Written Composition. Yet that other study, by Meckel, is more extensive and thorough in its conclusions and recommendations than is the Braddock work. Meckel's work shows that major questions still existed in 1963 about the teaching of grammar.

Meckel points to three crucial issues (981): First, none of the grammar studies up to 1963 extended beyond one semester-"a time span much too short to permit development of the degree of conceptualization necessary for transfer to take place." Second, none of the studies had to do with editing or revising, that is "with situations in which pupils are recasting the structure of a sentence or a paragraph." Finally, none of the studies makes comparisons between students who had demonstrated knowledge of grammar and those of equal intelligence who had none.
Teaching Grammar to Writers by Janice Neuleib and Irene Brosnahan

So basically:

• None of the studies checked to see whether students being taught formal grammar actually learned formal grammar.
• None of the studies included students deploying formal grammar in an effort to copy-edit sentences of their own.

And on this basis, schools have been refusing to teach grammar for a good thirty/forty years now.

"Teaching grammar doesn't work."