error analysis and contrastive hypothesis?

[azamouri]

I would like to know the difference between error analysis and contrastive analysis also which of them is more useful?

[lolwhites]

"Useful" in what sense? And if it's to help you write your book, I should warn you that regular posters on this forum take a dim view of people joining in order to ask us to help them write their books and theses. We take the view that writers should do their own research.

[metal56]

I would like to know the difference between error analysis and contrastive analysis also which of them is more useful?

How much have you read on each of those?

[azamouri]

hi again,

please forget about the book.
is this forum done to discuss ideas and problems?
If yes then give your opinions about this subject.

And if you think that my questions are useless then tell me and I leave this forum.

Concerning contrastive analysis and error analysis:

which of them is better for a trainer to improve his teaching skills?

according to what learned, I found that error analysis is better, do you agree?

[lolwhites]

As I understand it, the assumption behind Contrastive Analysis was that all learner errors are due to L1 interference, therefore you can predict them by analysing the differences between L1 and L2. In practice, however, CA both over-predicts (predicts errors that don't occur) and under-predicts (fails to predict errors that do occur). That doesn't mean to say that L1 interference doesn't happen, but it suggests that CA's pedagogical value is limited.

[Lotus]

according to what learned, I found that error analysis is better, do you agree?

I think you should go with your instincts on this one. Lolwhites has given you a sound analysis. CA demands that the instructor be competent in both languages and able to do an in depth comparison and analysis of each on many levels. You will waste an awful lot of time working on problems that will very likely never occur, and then have to deal with the ones that do come up.

I understand that as a theory, CA has it's supporters, but are there actually any real, live teachers who use it? Anyone out there?

[Stephen Jones]

both over-predicts (predicts errors that don't occur) and under-predicts (fails to predict errors that do occur).

It's all a question of how clever you are at predicting.

Let's take a simple example. It's very common for Spanish students to make errors like
*The table it's in the corner
instead of
The table's in the corner.

This appears not to come from L1 as Spanish does not do this, but it is in fact an example of L1 interference.
What we have is not
L1>>>L2
but
L1>>>>Student's internal grammar of L2>>>>L2

The student hears phrases like It's a chair, It's a table which in Spanish, which does allow null subjects, would beEs una silla and Es una mesa.
The student accordingly forms an internal grammar of English in which it's is the equivalent of es and esta, and it is this defective internal grammar which causes the mistake.

[lolwhites]

Sure, Stephen, but what about sentences like *I no speak English? If your student is Spanish you can put it down to L1 interference, but it's quite a common error even when the student's L1 doesn't form negatives by simply sticking no in front of the verb. It seems to be a common "transitional" form that students of many language backgrounds use i.e. it's a developmental error rather than a transfer error.

[Stephen Jones]

The point is that auxiliary 'do' is most unusual (it's actually a Celtic feature). So the fact that their L1 is different from English means they will have to improvise a form until they have internalized the correct one.

[lolwhites]

Does that count as L1 interference though? I always thought "L1 interference" meant inappropriately using an L1 structure in L2, not improvising any old thing because L1 lacks an equivalent construction. More of an "absence of L1 interference".

[Stephen Jones]

The point is that L2 is different from L1. All comparative linguistics suggests is that there will be problems when this is the case and not when it isn't.

You'll find a Frenchman making many more mistakes with the progressive than a Spaniard.

Arab speakers have problems with p's and b's but not with l's and r's as Japanese speakers do.

[lolwhites]

Steven - I agree with what you're saying but I think we may be talking at cross purposes. For me, L1 interference, or transfer, means inappropriately applying a feature of L1 in the production of L2 as opposed to making false analogies. If you're saying that problems will arise where there are difference, then I totally agree; after all, anyone who teaches monolingual classes and knows the students' L1 is at an advantage. The problem with CA, as it was explained when I first studied AL, is that it goes too far in trying to predict exactly what errors learners will make.

For example, it's quite common for students who've studied used to for past habits to start saying *uses to for present habits. In the case of a Spanish speaking student you can say "Aha! That's because they're thinking of the verb soler" (though you'd be well advised to ask the student if that's really the case), but that may well not be the case for students whose L1 isn't Spanish. CA might predict that Spanish-speakers will make kind of error, but what about everyone else? I think if you ask, students who make that mistake are most likely to say "I just assumed you could say it in the present because you can say it in the past", or even "Why can't you use it in the present when you can use it in the past?"