TL;DR (applies to both replies): I hate memorising and regurgitating lists of grammatical structures and vocabulary, especially when no context is provided. However, I believe that formal grammatical knowledge, especially knowledge of technical vocabulary used by experts to describe how languages work (which is what Refold would call ‘advanced grammar’ that ‘there’s no point studying’), allows one to greatly accelerate language learning and to access knowledge that most other learners, and even natives, rarely have. This allows one to ‘stand on the shoulders of giants’ by directly acquiring the knowledge that experts have collated regarding usage, essentially bypassing many hours of immersion that would otherwise be necessary, allowing one to reach and even surpass native level fluency much more rapidly. I speak from experience with other languages, though I have yet to reach near-native fluency for Japanese. I’ve provided a few examples below. To draw an analogy, having such knowledge in one’s mind is like being a musician who can play by ear and improvise music in any style: such a musician is not only good at observation and imitation (aka the most important skills for successful immersion), but also understands the framework behind music and what sorts of chords and rhythms might recreate similar effects to those desired, allowing him or her to rapidly narrow any search for musical inspiration down to only a few chords and other musical elements. Similarly, advanced grammatical knowledge makes mining for usage knowledge much more efficient.
PS: Studying this way doesn’t mean I stop immersing myself. I just immerse myself differently – sometimes in things that are plain fun for me, like science articles and anime, and other times in studies of grammar and usage, which I read in the target language. That way, I learn even more technical vocabulary and get used to technical language analysis even as I acquire more knowledge about how I should use the language. (I don’t do this actively, by the way – I’ve already acquired quite a bit of technical knowledge over time, so I just use it whenever I have a question and read whatever I get until my question is answered.)
I agree that grammar is something that only gets formalised based on actual usage patterns. If we were to draw an analogy with sports or music, I guess we could say that grammar is like a rulebook or musical theory: it tells you what you’re supposed to do and what you’re supposed to avoid, but it can’t teach you to play well. However,…
…the question is what this point is. I agree with this idea in principle, and it truly is the ideal: reaching the point where you’re independent enough to acquire new language skills based on observation and imitation alone. However, in certain languages, there are things that are more easily acquired through formal study, or at least with the help of references meant for formal study. If there were no demand for such resources, why would dictionaries exist?
French conjugation as an example of when immersion is insufficient, and grammar is helpful
For example, in French, conjugation mistakes are very common, even among native speakers, with a major reason being the fact that many forms sound the same while being spelt differently. However, in theory, native speakers, being immersed in the language all their lives and having received the education necessary to reach the point of independent learning, should be in the best position to avoid these mistakes and improve themselves. Why doesn’t that happen? I’d say it’s because of the difficulty of learning through observation alone. Sure, it’s true that I personally managed to get a good feel for the French simple past (which is growing increasingly rare in common usage) through immersion while reading a book. But that’s because 1. I was interested in those forms (many native speakers are not, especially since the form is quite useless on a personal level) 2. I already had a conjugation framework in my mind from a textbook, dictionary or reference source, meaning I knew what to look out for, and seeing what I was hoping for reinforced what I had learnt.
Advanced Japanese expressions as another example of the difficulty of immersion, particularly when such expressions are relatively rare
In the case of Japanese, while it’s true that many ‘grammar points’ are really just idiomatic expressions, the problem is that there are a lot of them. Worse, the ones that are considered marks of advanced Japanese ability (e.g. the ones recommended for N1 study) are frankly quite rare in common usage. Ideally, they should be learnt in context through immersion (which is what I personally prefer to do), but it’s fairly likely that they won’t be encountered enough to be remembered. More importantly, even if they are encountered, they won’t be encountered often enough to allow the learner to infer that they are expressions, and not just random series of words that happen to work well in an isolated case. For example, I came across 今泣いた烏がもう笑った among other dictionary examples, and I had no idea it was a saying that referred to a child’s mood changing rapidly. I understood its literal meaning though. A set phrase like this was already hard enough to detect. What about random little phrases like を余儀なくされる？I know this phrase because I came across it once and found it intriguing, but I still need Google suggestions to correct me because I get the order of the words wrong. It’s an N1 grammar point.
I guess my point at the end of the day is this: not even the best textbooks and classes in the world will prepare you for everything, and there’s a lot you’ll have to pick up on your own even at the most advanced levels. However, believing that one can rapidly reach the point where immersion alone will suffice, even with the help of a dictionary, is really quite idealistic. I wasn’t able to comfortably and efficiently use a monolingual dictionary for French until I hit the equivalent of a C1 level on the CEFR (which would probably be high N2-low N1 in Japanese), and even then, I continued to read articles on grammar and usage to advance more quickly.
I don’t want to be a wet blanket, and I know that your point, @JesperHH, was that immersion is an essential part of language learning, and not that grammar and vocabulary are bad. Nonetheless, as much as I think a lot of Refold’s ideas about immersion are very good (the use of Anki aside, because I dislike flashcards, personally, even if I acknowledge that they help many people), the idea that ‘a little basic grammar and vocabulary to get you started’ is enough is, to me, very wrong.
Grammatical knowledge allows us to use better/more respectable language when communicating with native speakers
Sure, grammar is an explanation of the chaos of language, as Refold puts it, but in our age of standardised education, grammar is also the new norm. Grammar teaches us what is correct according to central/widely recognised authorities, which is often a result of a consensus among the more prominent and persuasive writers and grammarians in all of history. Not knowing at least the most common rules makes us seem less educated or scrupulous about ‘good language use’ to native speakers.
What grammar allows us to do that immersion cannot, at least within a relatively short time frame
More importantly, while I recognise that it’s possible to pick things up through immersion, including advanced grammar (my friend studying in Japan does that), it’s not only slow, but also imprecise. Unless you have massive exposure to a particular structure, you cannot properly acquire how it is used. But what if it’s an expression that is extremely common, but only among distinguished writers? How can you access such knowledge through immersion alone as a non-native speaker who doesn’t know who those writers are? My answer: through grammar and other frameworks provided by reference sources. If you truly want to go far, and fast, you’ll need extra tools, and one of those tools is grammar and grammatical terms. Even if you’re like me, and you hate having to memorise grammar for grammar’s sake with no context, knowing the basics of advanced grammar or technical grammar will allow you to reach information and handle ideas that people using immersion alone will take much, much longer to acquire. Grammar isn’t simply ‘a crutch’ for basic comprehension, as Refold puts it. It can just as well be a booster rocket or a nitro supply, if you know how to use it well.
I think what’s truly harmful/problematic is the idea that learning what’s in the textbook is ‘enough’. Learning all of the grammar in Genki won’t be harmful in the long term, but thinking that learning it through Genki is enough might be, because there are probably use cases that aren’t covered by Genki or Genki’s explanation is too rigid. That’s where immersion comes in. Even so, should you just drop grammar at some point, assuming the rest will come with immersion? I suppose you can, but I wouldn’t recommend it.
French past tenses (of which there are five in the indicative mood) as an example of what technical grammatical knowledge allows
I’ll admit that I probably have a bias towards ‘structured learning’ and ‘getting a grounding’ because even for English (my native language), I was taught grammar as soon as I was deemed of age to understand it. My entire cohort in primary school started at the age of seven, studying things like when to use ‘will’ or ‘shall’ (which I didn’t understand at all at the time, but which I figured out later). Sure, it may seem like that proves that explicit grammar instruction isn’t useful, but my love for technical terms, of which I frankly know not the source, is something that made my life much easier when I learnt other languages. Knowing technical language means that I can access the analyses of experts, understand what dictionaries are referring to when they break down structures and parts of speech, and precisely express what I still don’t understand. For example, other people might attempt to learn, through immersion, what the five main forms of the past tense in the indicative mood in French are used for. The problem is that some of those tenses are much too rare now, so unless one is an avid reader with plenty of time, one would have great difficulty acquiring the relevant nuances, and be unable to use them oneself. I simply read articles and books by grammarians who had already analysed the subject, meaning I acquired the same skills with clear guidelines for not even 10% of the effort someone learning through immersion alone would have needed.
The ば conditional in Japanese as another example
It’s the same thing in Japanese: because I know the technical terms for different verb forms, different sorts of keigo and different parts of speech, I can get precise answers to my questions very rapidly. For example, I was thoroughly incapable of figuring out what ば meant and how it differed from たら or と as a conditional form because in order to understand things in the process of immersion without a teacher, I had to give myself an approximation, meaning I treated them all as some sort of ‘if’. I knew how to form each conditional, but I just couldn’t remember which had been used in what sort of context, and I didn’t have enough resources to build my own corpus of examples which would allow me to compare them and understand. What was the solution? Formal grammar knowledge. I checked explanations in English, like those on Duolingo. They were insufficient. Luckily, technical grammatical terms in Japanese aren’t completely foreign to me, so I looked for studies in Japanese and read them with a little dictionary help when I was lost. I found out, as a result, how Japanese people actually use the structures, because those studies relied on both literature reviews and surveys of today’s native speakers. Learning these nuances through immersion would have required corpus building and painstaking analysis, but thanks to grammar, I improved my understanding without needing to do so and while avoiding unnecessary confusion.
Grammar isn’t just a crutch: it’s an accelerator, and practically an unfair advantage, even at the highest levels of language use.
