Scott Wood argues that 'It seems that the examples I (Scott) gave support the learning of these structures by examples and not by explanation, by induction and not by deduction.'

It's crucial here to differentiate between the acquisition of passive and active, and easy and complex grammatical competence. The teaching approach that Scott has described appears related to 'Total Physical Response' which can no doubt enable learners to recognise easy grammatical-lexical differences. To my knowledge, however, there are no studies which demonstrate that it can enable students either to acquire passive complex knowledge (of the present perfect, for example), or active accurate easy grammatical competence.

As to Johanna's point on her successful inductive learning of German when immersed in the language, there are notes of caution to be sounded here in terms of arguments in favour of promoting inductive learning in the classroom.

1. It is feasible that Johanna is a gifted langage learner and as such should not be used as an argument for inductive learning.

2. I do not know Johanna's background but I'd guess that as she had been exposed to substantial grammatical instruction before her inductive learning experiences, she cannot be cited as an example of natural inductive grammar acquisition.

3. Even dismissing (1) and (2) as valid, one cannot use success during total cultural immersion as an argument for classroom inductive learning.

By the way, though I am an advocate of the central role for deductive classroom learning, I'm also an advocate of using inductive exercises as a way of leading into deductive learning.