Sunday, November 15, 2009

Grammar-check theory

Leverett, T. (2009, Nov.) Grammar-check theory. Unpublished manuscript. Available

I have to admit that what I've been writing is somewhat rudimentary, but it has occurred to me that it's important to document my thinking here, because I think we're moving into a world where the grammar that a person produces is actually a complex result of several forces: the automatic functions of one's grammar-checker, with the responses to the grammar-checker layered on top of that; in addition, the grammar that any given learner carries around is a result of interplay with grammar-checkers; this exchange has basically provided some learning, if only by a rudimentary system of stimulus-response activity.

It started out with the spell-check, frankly, and spell-check is probably the best way to understand it. The student tries to spell "fate" as in "fate of the crocodile" but spells "fait" which the spell-check then corrects as "faith". The student not only believes that his word, fate, is now properly spelled, but also walks out of the situation a little closer to mistaking "fate" and "faith" as separate words. The spell-check thus simultaneously alters his production, and teaches him something wrong. This is not to say it can't be undone by a teacher or by other experience. It's to say that what is happening is real and has real effect, no matter how miniscule.

If the effect is to be understood in its entirety, it can explain more important behaviors in the esl/efl classroom. For example, students get quizzical looks when teachers explain the passive. This is because grammar-check has been ironing out passives for years now, or to be more specific, not allowing students to make them, without marking them with a green line. It is intended to remind the native writer that passive is to be avoided (as pedantic, or cold, stylistically outdated) if possible, but in effect implies to the student that it is wrong or to be avoided for more basic right/wrong reasons. The student with his/her quizzical look is in fact reflecting a combination of factors; he/she has been taught, perhaps, in several classrooms, that passive is acceptable; yet he/she has been slapped, so to speak, by a green line, whenever he/she has tried to create one. It occurs to him/her, perhaps only for a moment, that he/she doesn't know whom to believe.

The theory is recognition of simple fact: that the beliefs the student brings to the situation are a combination of factors, both classroom teaching and technological stimulus-response. In the same way, any given writing product is a similar combination. It may have no SCGC (spell-check grammar-check combination) applied to it whatsoever, because it is handwritten and spontaneous, or because those have been disabled on any given machine. Or it may have whatever is provided as a default (this would probably account for over 95% of papers done in our lab, given what I've observed); but, these could be divided in terms of their users' awareness of, and active interplay with, the programs that are there. Some students literally address every green line that they encounter; they address each one, one at a time, while writing, and struggle with it until it disappears. Others wait until a paper is done, but do the same, essentially. Still others struggle with it a little, but take a more proactive approach and decide for themselves, sometimes, to ignore it. My point is that there are degrees to which each student responds to what is there, and their final product, naturally, reflects this.

Finally, a small percentage, which I couldn't even venture to guess, reflects active intervention with the system that is there. Some students, aware of how they can change or alter the settings beneath their grammar-checker, actively change or alter them, at least at home, in the privacy of their own den. Others are aware of the programs that are available on the free market, and crunch what they write through them. Again, this may happen only at home, or may happen on any machine that has the web open and available for free exchange with a word program. It can be assumed that the student will believe that the teacher won't approve of this, until it is proved to him/her otherwise; in other words, if you, the teacher, have said nothing about grammar-check programs, don't expect students to volunteer to you that they use a better one than the one your computer now offers as a default. Assume that they do it when you aren't watching. Don't assume that all do it every time, but assume that you don't really know what happens when you aren't watching, unless you have successfully disabled everything. And even then, watch out for the pin drives!

Not that this is all bad. I still say that the teacher should be open with the students about it, in any case, because all of these innovations are part of the student's world and ultimately part of what will create everything that student will write. Why not at least be able to discuss what happens, and why? Even the teacher who disables every SCGC should at least be able to tell students why, what grammar-checkers do, and why disabling them is the best option. Teaching a person how to live without a car is entirely different from teaching how to drive- which better prepares a person to live in the modern world? I can't say that I know.

Having not taken a stand on the essential question in this debate, I can say with certainty, at least, that rapid improvement in the nature and quality of grammar-checkers will not totally alter the landscape, or the theory as I've laid it out. It will simply make it harder to figure out what we are seeing.

Monday, November 09, 2009


Funes-Gallanzi, M. (n.d.). English translation tool: New internet "translation" tool brings scientific know-how closer to foreign students. Global Study Magazine. Accessed 11-09.

Describes the frustration of using translation tools in general, but also the use of simplish (, which will translate all articles into a simple version of English. (???)

This brings up an interesting possibility, obviously; this is one that I haven't explored.

grammar checkers revisited

At first, I thought that there were basically two strategies for dealing with grammar checkers as part of the student writing environment: 1) develop the students' skills independent of the grammar checkers, either ignoring them, or treating them as an unnecessary complication, when what matters is what the student knows or can produce correctly without them; or 2) consider them part of the students' environment, in the lab, or wherever word is installed; help students to learn to handle and manage the programs themselves, and teach students how to use them to their advantage.

I have now come to see that even #2 is quite complicated. I was leaning toward #2 myself; they are, after all, an ubiquitous part of our world, along with spell-checker; they have gotten more sophisticated; students can be expected to know where they are, what they are, and how to use them to their full advantage, in any case. But, I can now divide #2 into several possible strategies:

1) deal only with the checker that is the default on whatever machine the student is using, in the lab; let them deal with the rest, at home, out of your site;
2) give students an overview of grammar-checkers with respect to maximizing their potential; show them the best that are free, on the web; teach them to find and use the best of the best, whenever they are available;
3) find out the extent that they have used them in the past, in order to be aware of the twisted view they now have of certain grammatical situations (i.e. passive); lecture them on what has happened to them and what they can expect if they continue their dependence.

This isn't all; I haven't quite thought it out yet. It seems obvious that we teachers are really concerned about the students themselves, their true skills, their success in the future. But there's only so much time in the world. Why should we beat ourselves over the head teaching -s if a machine will simply fix every single one? Should we then only be concerned about the exceptions, and teaching students to spot when the machine isn't covering their back? The world is now full of spell-check absurdities: people using what spell-check has provided, rather than what is right, thus showing off, inadvertantly, both their dependence on the machine, and their lack of real understanding of which word is appropriate. We teachers can thus spot the machine-free, the machine-dependent, and those who are in between. What do you do? Encourage the dependence, or wean them entirely? Teach to the middle, making them aware of the dependence and the problems it brings? I can't answer this question even for spell-check, let alone grammar-check, but I've come to some conclusions based on what I've learned.

One particularly instructive story involves the students who no longer capitalize "I" because "we're used to Word always doing it for us" - not because they don't know it's necessary. It's a matter of how you spend your time, and what you get used to; do you have that green line with you at every turn? Do you always respond to it? Do you keep struggling until it goes away?

Grammar checkers have automatic correction, which means that on many machines it is impossible to make i + 1, for example, because the machine will change the single i to capital, regardless of whether you intend it, until you learn how to disable the function, and the vast majority probably never learns how to disable this, or even alter it in any way.*

In the same way grammar checkers allow a variety of settings, including formal language, etc.; I have no idea (yet) what these do to their language, or whether that is ultimately good or bad for them; it's possible that the vast majority of students, like me, have and use word functions without thinking about it much, and don't alter such functions as a matter of course, and routine politeness to the majority of users of a machine. But a person who chooses to alter the functions of a grammar checker can just as easily undo what they did, if they know what they are doing, and have just a minute or two before and after writing. One problem: how can one predict whether or not one will have those minutes when finished?

Again, I'm talking strategy here. If we know the ideal machine, and feel confident that we can point our students to its correct and prudent use, will they have the luxury of using it in every experience? In the iBT? In the English department's comprehensive writing exam (here, I have no idea if we even have one)?

They will almost certainly have spell-check; every Word does, doesn't it? It would only be an extreme Luddite who would disable the entire program, wouldn't it?

In any case I can now say that the middle road: both teaching to the student and his/her skills, and being aware of the monkey and what it does, and teaching awareness of it, are necessary. There is no way to free oneself of the whole environment, and be completely immersed in either buying into the best of the machines, or rejecting them entirely.

*I have no proof of this, independent of watching my students. I can say that in general they don't even seem to try to alter the functions of the lab computers, but that may be because they long ago figured out, or told each other, that they couldn't; but I don't even know that for a fact. POrhaps they can, and don't. Or, never, in front of the teacher, if it can be avoided.