Well, dear Celia, I �d say that those arguments of "natural learning", based on Rousseau�s thoughts on education, though rather appealing and much in fashion some 2/3 decades ago,  proved to be disastrous in Education. Grammar can essentially be defined as a set of rules that regulate spoken and written language that we really must follow if others who use the same language are to have an opportunity to appropriately understand what we �re saying. So, I wouldn�t give much credit to such �innovations�. It is a fact that rules don�t provide much fluency in the �real world�, but knowing them makes it easier to break them, which can be both fun and creative. Have a wonderful day.

I agree with Urpillay.  The area of second language acquisition has been the subject of extensive research.  A second language can be learned and the extent of this learning will vary from one learner to another.  Yori�s classification of learner variables provides a detailed list of factors for exploring SLA.  Proposed in 1976 these variables remain applicable today.  He proposed six main factors: age, cognition, native language, input, affective domain and educational background.  He further classified each of these resulting in many areas of inquiry.  Each has an impact on SLA; the extent of the impact will vary among learners.  Ideally, an effective teacher will be aware of these factors, how they interrelate, their impact on individual learners, and how they can assist learners to apply their learning beyond that of the classroom.  There is no single way of teaching or learning.