Theories

Theories on what Works in Teaching Language to Beginners

It seems plausible that seeing written material in a target language too soon will harm pronunciation, since learners will start to pronounce letters as they would in their native language, and will have a harder time learning target language pronunciation. If written material is delayed until learners have solid pronunciation, they can then read it with target language pronunciation. However I have not found research on this.

It has been said that learning any 200-400 words in the target language allows the brain to organize the sounds and then learn other words more quickly, even if the 200-400 are not the most common words in the language. This is why most courses, including the lessons here, do not simply teach the 200-400 most common words. Again I have not found research on this.

I have taught beginning French, but used standard textbooks, starting with the first lessons. The students learned more or less well, but their accents suffered.

en.wikipedia.org/wiki/Second_language_acquisition
Wikipedia gives an overview of theories on how learners learn a second language. One line of research is that hearing understandable content in the target language is the key to learning it. Another line of research is that trying to speak (or write) the language is important, as soon as students have a bit of the language to practice. When they notice where they have difficulties they gain motivation to go back and learn more.

ielanguages.com/linguist.html
Wagner helpfully classifies approaches to learning second languages:

• "Grammar-translation: the student memorizes words, inflected words, and syntactic rules and uses them to translate from native to target language and vice versa; most commonly used method in schools because it does not require teacher to be fluent; however, least effective method of teaching
• Direct method: the native language is not used at all in the classroom, and the student must learn the new language without formal instruction; based on theories of first language acquisition
• Audio-lingual: heavy use of dialogs and audio, based on the assumption that language learning is acquired mainly through imitation, repetition, and reinforcement; influenced by psychology
• Natural Approach: emphasis on vocabulary and not grammar; focus on meaning, not form; use of authentic materials instead of textbook"

The approach suggested on this site is "Audio-lingual" since it uses audio, imitation, and repetition. It is also a "Direct Method" in the sense that only the target language is used, so the mouth and brain become used to that language. Its advantage over more common Audio-lingual and Direct methods is that from the very beginning learners know the meaning of words, even with no teaching in their native language and no personal interaction with teachers.

After learning sounds in the target language from native speakers, by radio or MP3, learners can then seek out written grammar lessons or local teachers, to clarify the rules of the new language.

fourhourworkweek.com/blog/2007/11/07/how-to-learn-but-not-master-any-language-in-1-hour-plus-a-favor/
Ferriss suggests exploring grammar by asking a bilingual person to (a) translate a dozen sentences from the learner's language to the target language and (b) explain the translations, in order to see the order of words, use of pronouns, word endings, etc. in the target language. He uses sentences like:

1. The apple is red.
2. It is John’s apple.
3. We give him the apple.

The goal is to see how different the target language is, before deciding to learn it. Ferriss suggests learning the most common words in a language first, then using any material with high personal interest:
fourhourworkweek.com/blog/2009/01/20/learning-language/