Someone on a listserv thought perhaps Bill VanPatten had mentioned the problem with correcting an error by simply repeating the structure in its correct form, that problem being that the learner often did not ’notice’ the correct form.
Another poster mentioned an article by Lyster and Ranta on this very topic, but I did not see that book mentioned in VanPatten’s recent (2003) From Input To Output. However, he does address the issue so I thought I would quote him.
p. 85
Do Learners Develop Bad Habits If They Aren’t Corrected?
The concpet of bad habits stems from the days of behaviorist approaches to language acquisition. Learners came to the L2 with L1 habits that needed to be suppressed. Activities were to be constructed so that errors would not be induced. If errors did arise, they were to be corrected immediately and appropriate practice engaged. The idea was to avoid the development of bad habits.
In SLA theory, there exists a construct called fossilization. This term refers to arrested development, the process by which a learner plateaus or stops learning. What is different from the behaviorist idea of bad habits is that fossilization – though not desirable, and certainly not anyone’s goal – is taken to be something that just happens and that may be unavoidable for some learners. Not everyone fossilizes on the same rule nor do entire systems fossilize in the same way. And from the presentations in Chapter 4, you might guess that fossilization in the development of speech procedures is different from the fossilization iof the implicit system or any rules it contains. Several famous cases of fossilization have been presented in the SLA literature. All cases have involved L2 learners in more or less natural environments, some having received instruction and some not. Attempts at remediation had no effect.
What is clear from the …to be continued.