At LingQ we set the target for "known words" high. We will be introducing tests to measure whether the words that are claimed as "known words" are, in fact, known. However, we will make sure that we test the learner only against the words that he/she claims to know. The important thing is to have a vocabulary level of 8,000 or 12,500, which have been "earned" through listening and reading. If there are still many lower frequency words that the learner has not encountered in listening and reading often enough to know them, that is not a problem.
Remember that the native speaker might know 50,000 or more, and the learner cannot match that, but can focus on contexts which are relevant to him or her. There will always be holes.
2) What does knowing a word mean?
To me, knowing a word, just like knowing people, means recognition. There is such a large potential range of understanding of a word, its scope, how it is used with other words, when it is used most appropriately etc., that there is no clear point at which we can say that a learner has achieved total mastery of the word. Once we have recognition of a word, we are on our way to grasping more and more of the word, and this process might include forgetting it and relearning it. Hopefully we wil understand it when we meet it again and build on that.
I doubt that there is only a 20% difference between active and passive vocabulary in a non -native speaker. I think the difference is much larger. BE quotes a source which describes the vocabulary knowledge that Japanese students have of English as being "large, shallow and useless". This is unnecessarily harsh. The non-native speaker has had more limited exposure to the words he/she has learned and therefore his/her grasp of these words is necessarily shal ower. This is not unique to Japanese learners. Only continued exposure can gradual y deepen this understanding.
We will get better at using these words through use. We can build up our potential (passive) vocabulary, but ultimately to get good at using them we have to use them. As long as we have no need to use them we can happily continue building up our potential usable vocabulary, and our understanding of the scope of meaning and usage patterns of these words through meaningful input.
3) How do we best learn words?
Most people learning languages have limited opportunities to use the language. That would certainly be the case for Japanese learners. For that reason, although not only for that reason, I think the correct strategy in learning words is to focus on building up one's passive vocabulary.
This is also easier to do where there are not a lot of native speakers around. This means a great deal of emphasis on input, meaningful input. It means reading and listening to a lot content that is of interest and at an appropriate level of difficulty.
In my view, the first goal in language learning has to be a defensive one, to understand what is said and written in the language. The native speaker of English knows anywhere from 30-50,000 or more words. Even a 14 year old knows 14,000. With the native speaker, the difference between active and passive vocabulary is not as great as with the learner, so we have no idea which words the native speaker is going to use in communicating with us, oral y or in writing. We need to prepare ourselves in order to understand. There is no escaping lots of reading and listening.
We al want to learn to use words. We want more active vocabulary. However, I believe that having a large passive vocabulary is very important, and it can be 2 or 3 times or even ten times larger than the active vocabulary, it does not matter. Passive vocabulary will convert to active vocabulary over time.
I would guard against any attempt to be perfect in another language. Even people with a high level of fluency and a large vocabulary will stumble on the simplest words or structures.
This is not a problem.
BE says that not consolidating one's grasp of the first 2,000 words is like building on a foundation of sand, and that this is a bad thing. He points out that children use a very small number of words. However, children spend up to two years just listening, without speaking.
Learning vocabulary is not like building on sandy foundations, but rather like climbing up a sand dune. That is why you do not want to stand still , but rather need to keep going if you want to get to your destination.
I was asked in an email about learning idioms in a language. Maybe I am different from most people, but I do not bother with them. To me they are kind of the dessert of language learning. They come at the end of a meal. Once you have fil ed up on the solid fare, the key words and phrases that are used in a variety of situations, then you can add a few idioms to spice things up. And by that time you wil be able to pick them up natural y anyway.
I find that many learners have an obsession with slang and idioms, as if trotting out some very col oquial expression is going to upgrade how they sound in a language. In fact it is the opposite. Idioms are difficult to use correctly. They can easily sound very strange coming from someone who obviously does not have a good sense of the language.
Yes, occasionally you hear idioms or slang and do not understand. But, in my experience, there are always situations where you do not understand. It is best to focus on the most important words, what they mean and how they are used. Certainly, it is best for a non-native speaker to stick to standard speech.
Some teachers even try to teach the "real language." Then you hear non-native speakers saying things like "I gonna" "You wanna" etc. I just cringe. You will learn al the idioms you need just by exposing yourself to the language.
Nation defines comprehensible input as content with only one in fifty unknown words. That is 2%! I think this number is too low to be realistic.
First of al , when you start a new language, and for a long time thereafter, it is difficult to find any content, and in particular interesting content, that meets this definition of comprehensible. For a long time most content wil have a much higher number of unknown words, unless you read painful y childish content, which is unlikely to motivate a learner.
Second, the main job in language learning, in my view, is to acquire words and phrases. If you have to read fifty words, for every new word you learn, you wil have to read an awful lot in order to acquire a fluency level vocabulary, which I consider to be over ten thousand words.
I prefer the term meaningful input, where the interest of the learner is taken into consideration. I am not interested in reading children's stories with a lot of easy words. I prefer to struggle through the authentic version of some novel, conversation or news report, as soon as I am able, even though there may be 40 or 50% unknown words. Having the chance to listen, as wel as read, helps. Having access to an online dictionary helps. Using LingQ helps.
Using LingQ for Russian I have seen the percentage of new words come down from 40-50% to a level of roughly 20% for the novels I am learning. Some podcasts are now down to 10%.
Using LingQ, that level is quite comfortable. Other LingQ members have told me the same thing.
So I consider that an unknown word rate of around 10-20% is a reasonable balance between ensuring that the content is "comprehensible‖, on the one hand, and that new words are being acquired, on the other hand. It is often more important that the content be of interes t to the learner. This assumes that it is a learning assisted form of input, such as reading and listening at LingQ. Reading a book, unassisted, is another matter.