Christian Saunders gives an enlightening speech on how L2 teaching works and what myths there are about language teaching and learning.
Surprising truths about language learning | English teaching masterclass
https://www.youtube.com/watch?v=fPP887v8EWw
HIGHLIGHTS:
Immersion influences the speed of language progress.
Immersion is taking a language in often and analytically - not necessarily living in the speaking country.
Immersion is mentality.
The accent is fixed by 11 months old.
From this age a person loses ability to comprehend sounds of other languages but his/her own.
Native speakers are not more successful teachers. A research(*) shows no correlation between language proficiency and teaching self-efficacy.
L2 teaching efficiency shows more correlation with instructional strategies than classroom management and students' engagement.
However, a teacher's language proficiency determines his confidence while using a language which leads to his / her choice of teaching techniques.
Non-native speakers as teachers tend (as their students say) to give a lot of homework, correct errors, prepare learners for the exams, assess progress realistically and rely on the course book. Which is good.
Native speakers as teachers tend to focus on speaking skills, improvise, provide information about the culture, to be interested in the opinions, apply group work regularly. Which is much better in many ways.
This means a teacher should better allow himself/ herself make mistakes in conversations in order to use actual speaking in class.
Mixing L1 and L2 in the classroom is a good thing: it creates extra connections, it give cultural background. A human brain is a machine for making connections.
Business Language does not exist. It is a commercial product of publishing houses. What exists is business culture.
FSI (*) (their wisdom in teaching languages is indisputable) : from teaching the textbook to teaching a student to learn, from sentence-based pattern drills to communicative activities, to authentic materials.
Memorizing words: to build a context with the purpose words is more effective than to memorize them intentionally in a list. (I do not agree: OV. Memorizing through using (functional memorizing) is no doubt deeper and more functional and this is how it should be done, but it is slower and still needs component-focused support. Christian was talking about speed which is irrelevant in this case.).
Mixing lexical items which are closely connected in meaning (like synonyms) together in one learning section leads to their conflict in mind (interference).
There is no need for a too big vocabulary to function in English.
NSM Semantic primes (Natural Semantic Metalanguage) (I will attach the chart).
If you have a good foundation with simple building blocks, you can do a lot. Fluency isn't about "bigness", fluency is about repetition of "smallness".
Language and thought are not the same thing. (Interesting example of people with aphasia whose speech parts of the brain do not function, but they can do maths, navigate through city, think about other person's thoughts etc.)
Forcing the students to think makes them intake language.
(*)Connecting language proficiency to teaching ability: A meta-analysis.
(*) the Foreign Service Institute