http://fish.blogs.nytimes.com/2009/08/31...ch-part-2/

Stanley Fish Wrote:... That research, according to Dee, shows that “teaching grammar out of context” is ineffective, in part because, as writer teacher observes, “students are afraid that they aren’t abiding by the rules.”

If that is what is meant by teaching grammar — memorizing rules and being always afraid of breaking them — I agree. If the effect of instruction is to make students fear that they are walking through a minefield of error and that at any moment they are going to step on something that will wound them, the odds of their learning anything are small.

The research that led Richard Braddock, Richard Lloyd-Jones and Lowell Schoer to conclude in 1963 that “the teaching of formal grammar has a negligible . . . even a harmful effect on improvement in writing,” is no doubt correct but beside my point; all it proves is that “drilling students on parts of speech” (McLoughlin) doesn’t work. What does work, I have found, is something quite different: drilling students in the forms that enable meaning; and these are not inert taxonomic forms, but forms of thought.

A small example. Let’s say I’m teaching the neither/nor form. I begin by producing a simple neither/nor sentence. “Neither his age nor his disability prevented him from competing.” I then ask my students to write their own sentences on that model. Most of them are able to do it, and they produce sentences with 20 different contents, but only one form. The next step is for the students to figure out what that form is. Just how does a neither/nor sentence organize items and actions in the world?

It takes a while to work that out, but in not too much time students are able to explain that the form organizes three components: two conditions (age and disability) and the resolution (to compete) in relation to which they may have presented an obstacle, but did not. The important thing students learn, in addition to being able to generate neither/nor sentences forever, is that the relationship among those components, whatever they are, is always the same: two assertions that have a relationship to each other combine to highlight the unlikeliness of an action or an attitude.

Notice that this is an abstract and purely formal account of the matter and that it will fit innumerable narratives (and a narrative is what a neither/nor sentence is). While the content is variable and abundant — as David Berman says, “content is everywhere” — the form is unvarying. It follows then that what students must learn are the forms; the content will follow. A neither/nor sentence, or an even-though sentence or a nevertheless sentence, or a thousand other forms that can be studied and mastered — these do not clothe an antecedent content; they make it possible; they are not brought in to adorn a story; they are the story. In short — and I borrow this phrasing from my book editor Julia Cheiffetz — in learning how to write, it’s not the thought that counts.

I'm pretty much in agreement with what's quoted above, if I interpret it as an example of how to learn grammar formally yet descriptively, using genuine content/context as a tool to supplement the identification and deconstruction of these forms so that they can be reproduced easily in new contexts. Of course, the question goes back to how or whether to integrate this into classrooms, how to define and combine the 'expert' and the 'user'.

Other thoughts: http://forum.koohii.com/showthread.php?p...2#pid64802 and http://forum.koohii.com/showthread.php?tid=3751

o_O I know some people love to pull a language apart like a science, labeling different parts of a sentence the name of noun, verb, adjective, adverb, quantity modifier, particle, pronoun marker etc etc, but whenever I learned a language, I would do it by memorizing entire conversations or at least, entire phrases, and then by replacing the phrases with phrases that were similar to it.

(For example, if I hear the sentences "She's too big to fit in that dress" and "I'm going to buy that car", then I think: "She's too big... to buy that car" or "I'm going ... to fit in that dress" It doesn't always make sense, but it's practice in breaking apart the language by fragments that can be interchangeable).

But yeah, some people learn language just fine by studying nouns and stuff. It's a good way to explain the pattern, but I can't imagine someone actually thinking about where to place the subject relative to the object and action and blah blah blah. That's a lot of thinking to do to say a sentence that would be said and over in 3 seconds or less.

ocircle Wrote:But yeah, some people learn language just fine by studying nouns and stuff. It's a good way to explain the pattern, but I can't imagine someone actually thinking about where to place the subject relative to the object and action and blah blah blah. That's a lot of thinking to do to say a sentence that would be said and over in 3 seconds or less.

When you're first learning a language, those rules help to make you not sound like an idiot when you speak.

But you're right, that -is- a lot of thinking just for a short 'I want a sandwich' sentence. Far too much to sound natural when you're speaking... But do it long enough, while getting good input, and the thinking part goes away.

On the other hand, with your method, you get there, too. Just with some extra embarrassment about how you are saying things.

I tend toward the rules-based method, but I think your way is probably better.

ocircle Wrote:o_O I know some people love to pull a language apart like a science, labeling different parts of a sentence the name of noun, verb, adjective, adverb, quantity modifier, particle, pronoun marker etc etc, but whenever I learned a language, I would do it by memorizing entire conversations or at least, entire phrases, and then by replacing the phrases with phrases that were similar to it.

(For example, if I hear the sentences "She's too big to fit in that dress" and "I'm going to buy that car", then I think: "She's too big... to buy that car" or "I'm going ... to fit in that dress" It doesn't always make sense, but it's practice in breaking apart the language by fragments that can be interchangeable).

But yeah, some people learn language just fine by studying nouns and stuff. It's a good way to explain the pattern, but I can't imagine someone actually thinking about where to place the subject relative to the object and action and blah blah blah. That's a lot of thinking to do to say a sentence that would be said and over in 3 seconds or less.

Well, it's like this paragraph: "The research that led Richard Braddock, Richard Lloyd-Jones and Lowell Schoer to conclude in 1963 that “the teaching of formal grammar has a negligible . . . even a harmful effect on improvement in writing,” is no doubt correct but beside my point; all it proves is that “drilling students on parts of speech” (McLoughlin) doesn’t work. What does work, I have found, is something quite different: drilling students in the forms that enable meaning; and these are not inert taxonomic forms, but forms of thought."

Parts of speech being nouns and whatnot. If you were to quiz me on English parts of speech, I'd fail miserably. Same with any of the sentences I've deconstructed and that have reached maturity in my Anki deck, even the ones from Japanese the Manga Way.

But is there a place for parts of speech? Maybe... For those who go 'meta' on the language and analyze patterns of usage and try to find better ways of describing that usage and need these models and taxonomies for concision amidst these meta processes; perhaps it can be helpful for the timely generation of accessible descriptive grammar guidelines for understanding real world sentences w/o relying on jargon, by those analysts, users, or a combination thereof. (http://www.gabrielatos.com/Grammar-Intuitions.pdf says it better.)

wccrawford Wrote:

ocircle Wrote:But yeah, some people learn language just fine by studying nouns and stuff. It's a good way to explain the pattern, but I can't imagine someone actually thinking about where to place the subject relative to the object and action and blah blah blah. That's a lot of thinking to do to say a sentence that would be said and over in 3 seconds or less.

When you're first learning a language, those rules help to make you not sound like an idiot when you speak.

But you're right, that -is- a lot of thinking just for a short 'I want a sandwich' sentence. Far too much to sound natural when you're speaking... But do it long enough, while getting good input, and the thinking part goes away.

On the other hand, with your method, you get there, too. Just with some extra embarrassment about how you are saying things.

I tend toward the rules-based method, but I think your way is probably better.

Which ever method you choose you're going to sound pretty much like an idiot either way when you first start speaking. It's the ability to accept that and get over it that will see you move in leaps and bounds speech wise. I find a generous amount of alcohol works wonders to avoid the feeling like an idiot syndrom

As for me I tried the rules way and failed miserably. Now I work with sentences and am seeing significant improvements in what I am able to produce. I alos think I sound more natural.

I realize you're responding to one another on a personalized tangent, but just to reiterate, the article portions I quoted above and which resonate with my own preferences are anti-rule + pro-sentence, at least according to my reading. ^_-

Now that I think of it, this goes back to my wanting to start practicing output by remixing things I know well as microblog posts. Or practicing reading/listening by remixing mature cards into new contexts. Anyway. I've done my weekly duty of posting a descriptive grammar-related comment!