I am interested in the claim that technological support such as spell-check and grammar-check actually makes people
lose skills. I have always tended to believe this but actually have no concrete proof, and I find it a complicated question. Someone once pointed out to me that when the printing press was invented, humans lost their ability to memorize huge works (such as
The Iliad or
The Odyssey) which they had previously memorized in their entirety; now, such a huge and complete memory simply wasn't necessary, so we lost it. It could also be said that with calculators we lost the skill to do huge sums, etc., and that with any kind of technological support there will be some kind of corresponding loss in what most people are able to do on their own, when measuring the population.
So, we look around, and we say, spelling sure has gone to hell, since the advent of spell-check. Actually, I've been known to say that, but I'm not sure it's true, and I'm not sure other people would say the same thing. For one thing, there is a lot more informal writing these days, so it's much more common that one is reading blogs, online fora, chat, all kinds of things that encourage a general quickening of conversation, and loosening of need for correctness. Also, technological support has crept up; it used to be, that when typing on a blog or online forum, you didn't have the support of spell-check, but now you're more likely to; also, spell check has gotten better, so, now it's able to fix errors that were routine several years ago.
An example of this was the their/they're/there problem. It seemed for a while that everybody and their brother was making this mistake; it exploded. I thought, for a while, that within the general population there were just a lot fewer people who knew the difference. But that's only partly true. Thirty years ago, only the most literate of us ever made it to print, now everybody can be published with a single stroke, so we are seeing a
far more accurate picture of what the average person knows. But fast-forward just a few more years, and now the machines have become able (virtually) to fix this problem, and now again I see much fewer their/they're/there errors.
Let's do a longitudinal study of people's
actual knowledge of the difference between their/they're/their. As I said, thirty years ago, only a percentage of people knew the difference, but they were the ones that always got published, and there was always an editor to catch them if they for some reason missed on their story's way to the press. The only stuff that was published was stuff that had been vetted by editors in newsrooms and magazine offices, and you mostly saw all correct usage in this area. The
actual knowledge of the actual population was much more dismal. If you were to make a simple test people would have flunked it in droves, just as they would now.
So the next question: does the advent of the machine-support, giving modern writers the impression that the machine will take care of it (even when it obviously can't always do so) actually make them
less likely to learn the proper rules? It would be almost as if the impression that it is being done properly in so many areas, gives the impression that grammar in general doesn't need to be learned; that it would be a lot of effort just to learn the few rules that aren't covered properly by the machine. Thus if we were to test a cross-section of the population's knowledge of grammatical rules, they would perhaps tell us that they hadn't bothered to learn their/they're/there because they had assumed that the machine would take care of it for them, as it did all other mistakes, like its/it's (perhaps the most stubborn of them), or
cut the mustard, my favorite. The machine's grammar-check systems are getting better, as we speak; they are building on previous knowledge which is quite extensive, and they are more and more able to approximate what the writer wants to say, and provide it. The question is whether people
know less as a result.
In mathematics the concern over the use of calculators was justified; teachers found that if they gave kids calculators too early, the kids were unable to develop the sense of when a calculation
looked wrong, which was always a kind of check on one's work; the loss of this skill made kids look bad and made them come up with numbers that were grossly unlikely. Correspondingly their advice to math teachers was, give them the calculators, but not until you've taught them some basic rules of how to make simple multiplications and divisions; so, they will always be able to check their work and see if the calculator is on the right track or not. I'm not sure if this same reasoning can be applied to spell-check/grammar-check; for one thing, we are not talking about complex calculations here. You either know the right word, or you spend a lot of difficult frustrating time trying to figure out which one
is the right word (spell-check), or, you get a red line and some incomprehensible advice that gives you choices, all of which look equally mystifying (to the non-native speaker)- though, to the native speaker, some clearly
sound better than others. In the case of their/they're/there or hear/here, sound is of no help. The failure to know the difference is bound to peek out at some point or another, as you struggle to finish a piece of work within a time limit of, say, two hours too much.
An obvious argument is, so we know less grammar, but, we don't have to, since the machine is taking care of so much of it now. This is an interesting argument and has several aspects to it. In my field, ESL writing, an old problem known as S-V matching has virtually gone away. People used to write
he go and
she eat much to our great consternation as we wondered how long it would take for them to acquire the most basic grammatical signal in our language, and we fretted over so-called "fossilization" which was the force which supposedly suppressed their ability to pick up this elemental ability and integrate it into their productive systems. Now, machines change ninety-five percent of these
he eat constructions into
he eats and nobody is any the wiser; we teachers assume they have mastered this
-s thing even though they haven't, and we occasionally see an irregularity (such as:
he who swims in shark-infested waters eat carefully); but, there were always plenty of irregularities. The other day I graded maybe 35 hand-written essays for a writing assessment; there was no technology involved or allowed, and I saw perhaps the same percentage of failure to match S-V (
he eat) as I always used to. I
did not see people learning it better (due to the machine) or massive failure to learn it (due to the machine), or, at least, if there were clear patterns in either direction, I didn't pick up on them.
This may be an inappropriate example simply because it involves
usage more than
knowledge; the vast majority of ESL students at least
know that they should put an
-s on certain verbs and yet still fail to do so. In far more complex situations we struggle with things we don't
know and thus rely on the machine even when we can't be sure that the machine will fix anything in the right direction or pick up on mistakes. The native speaker can rely on an ear that will tell him/her whether something
sounds ok, but often this is inaccurate, and we see lots of bad grammar these days. I guess I could crystallize my questions as follows and say that, if you have any input about the answers, I'd be glad to hear it and am truly open to alternative explanations for the vast wealth of data we are seeing these days.
1. Has the profusion of spell-check technology overall made people's spelling (as it appears on formal documents that they want to reflect correct spelling) better or worse? In other words, does your average person have better or worse writing in formal situations?
2. Has the profusion of spell-check technology improved their actual skill, or made it worse? One could argue this either way: 1) that, by patiently correcting every non-word, it slowly teaches us to stop making them; by guiding us to change wrong to right, it makes us notice, change wrong to right, and ultimately do right; OR, by doing something for us which we should have learned on our own, it allows us to slack and be slack, for a little too long, ultimately working against us.
The above questions dealt with spell-check and the population at large. But I'd like to expand my inquiry to grammar-check technology and to a more specific population, learners of a language or of ESL specifically.
3. Does grammar-check technology have an active role in teaching native speakers (or non-native learners) rules, either correctly or incorrectly, that alter their learning pattern so that people growing up with technology are actually learning in a different way than they did before the technology? If so, is it possible that the technology alters their path in such a way that some things are learned more quickly, while others will now take longer? I am interested in entire populations here. In other words, I'd like to know if
all American writers (who, by and large, all use
Word on their computers) will show a different pattern than they did twenty years ago, with respect to, say, deciding how to construct complex sentences. I suspect that all writing will become more uniform in certain ways as all grammar-check programs come to agreement on certain standards; people will learn to conform to this, but, they may deliberately change other rules or disregard some of these standardizations, even come to resent them. But with the non-native population, some structures will become harder to acquire, and may move to the end of the order, just because acquiring them is a complex process and the machine stunts their growth at every step. I set this out as a hypothesis; I suspect, for example, that making present perfect is a complex process and has been delayed wholesale by the advent of technology, which prevents learners from making the intermediary steps that would show the process of acquiring it (So, on my hand-written tests, for example, I could find
I have eat or
I have eating, which are beginning attempts to make
I have eaten, but when typing, students having typed these forms would find them corrected to
I am eating or
I eat and would settle for a sure thing, thus learning not to try anything new and/or to avoid this tense altogether until they have it mastered a little better). I speculate about what happens here. I don't have enough concrete evidence.
In the big picture: time provides technological developments, and puts them in our path. We open up Word only vaguely aware that we now have a conversation going with both a green line and a red line, which we ultimately must please or at least come to terms with. We grammar fanatics have no problem with this as we are reasonably confident that we know when the lines are misguided. The lines are ubiquitous; they influence every writer at every moment; they do far more damage, and have far more influence, in cases where the writer is really
not confident about any of his/her language, or even intentions, or the rules that guide them. I would maintain that the lines have steady significant influence on
all writing and
all learning systems and that we should not carry on as if we are dealing with the same language we had, say, twenty years ago. It's somewhat like those bumps they put in the road to keep you within your lines as you drive; there is no question that they work, in that they make entire populations drive differently and more consistently and thus make driving safer; this is why they are a good investment and states such as California put millions into providing them even at the expense of a number of other things that the cost precludes. In the case of grammar-correction, steady cause-effect red-line or green-line stimuli
do change our actions, I'm willing to bet on it, and I think that a steady overview of what we write, as a collective society (all pounding out our documents on
Word, would be very revealing. I'd like to be more thorough, however, in finding out
exactly what the results are...
