The other day I was in a conversation and couldn’t for the life of me retrieve the Gàidhlig word for “question”. All I could think of was freagairt, which is “answer”. I had to ask what it was. It’s ceist, of course. Duh. That’s a word I “know”, or “am meant to know.”
But the real question is never ‘do you know this “word/phrase/structure/chunk of language”?’ It’s always, ‘can you comprehend this chunk of language right at this instant, or produce this chunk in a way that effects communication?’
Which means the strongly binary model most of us inherit of language learning, which includes “Teacher taught word X, therefore student learnt word X” (wrong not just for languages, but for instruction in general), and “You memorised word X, therefore you know word X in all circumstances” or even “you once got X right on a multiple choice question, therefore you can actively recall X for communication production”, and so on – these are just wrong.
‘Knowing’ is a lot fuzzier. It’s a huge range of contextualised, circumstantial, bits and pieces that determine whether communication is going to take place in any particular instance, and how well a message is going to go from producer to receiver.
Which is why, at the end of the day, “vocab testing” is mere approximation. It’s testing, “can you on particular occasion X, recall particular word Y (actively? passively?) in particular context/decontext Z which may or may not bear much relation to any genuine language encounter?”
It’s also why we should basically ‘lighten up’ on students. “I taught you this” has no real place in a language teacher’s teaching vocabulary (except maybe as a joke?). Students don’t really need to feel shame/guilt/frustration at not knowing a chunk of language in that moment, they just need the minimum amount of help to make the utterance comprehensible, so they can get on with getting meaning and so acquiring language. And the next time they encounter, or need, “chunk X”, it will hopefully come a little easier. Or the next time. Or the time after that. Or however many times.
Often I will recognise a greek word in the context of a completely unfamiliar sentence that I would not have known at all if that word had it popped up on a flash card. (I know this because often I will come across a sentence, read it, and then realise one of the words is one of those words I can _never_ remember in my flashcard deck.)
I am beginning to think that the usage of words in context needs to be appreciated a lot more, it should be a greater part of the learning process, aligned with it being a greater part of the testing process.
In a sense, all ‘vocabulary’ learning is contextual. The context of a flashcard is rather dull though – but even the elements like medium (physical/digital/audio), setting (where you physically do your flashcards), emotion (how you feel when you encounter words), can all be part of the context in which you ‘learn’ a word, and influence its recall.
But, yes, vocab recall is (a) generally stronger when encountering a word in an utterance, and (b) stronger passively rather than actively (we generally can comprehend at a much higher competency than articulate).
All of which should influence both teaching and testing.