More things that can be pushed to subconsciousness the better you get. You cannot constantly map your native tongue to a target language consciously and improve.

Look at any activity, chess, music, physical sports, at top levels most of it is subconscious effort.

Imagine having to constantly look at your keyboard, consciously searching for the letters to communicate your thoughts.

How would you expect to communicate well if your conscious process includes the menial search for the individual letters? How would you be a good chess player if you didn't have an intuitive feeling for most board positions but had to evaluate every move for a long time?

The things you learned well, were things that you successfully committed to your subconsciousness. Sometimes you plateau and think there's no room to improve, but by inspecting the conscious effort, you can find things that need to be practiced to make them subconscious.

I have studied a lot of the theory behind second language acquisition. Indeed, acquiring a language, even a second language, is a subconscious process. Or rather, learning a second language is a process that usually happens consciously and subconsciously in parallel. There have been debates whether conscious knowledge turns into subconscious one (skill automatization) or whether there are multiple learning systems that learn independently (the so-called no-interface position https://en.wikipedia.org/wiki/Interface_position ). The debate is not settled.

However, that doesn't mean that the process of acquiring the language isn't effortful! Even the most hardcore no-interface positionists accept that the subconscious learning system uses linguistic input that is _processed_ for its meaning and _comprehended_ (called "intake" by some experts) as its source material. When you don't have automatized knowledge yet, comprehending is very effortful and cannot happen without focused processing. The theoretical debate is more about whether that processing requires "metacognitive" processes and declarative memory and automatizing those, or is just getting intake (input comprehended by one means or others, without regarding any "rules" or "deduction" or "explicit/system 2 thinking") enough. But all in all, it takes effort anyway, in the sense that you still need to focus, and you might get frustrated of not understanding enough, struggling to follow.

Also, there are some studies that claim that children don't pick up languages as effortlessly than is generally thought. Quickly? Yes. Often with good results (in terms of pronunciation and grammar)? Yes. But effortlessly? Not so much!

Language classes do work, as long as you participate.

Children may appear to be picking up languages effortlessly, but consider how much time they spend to do so: take 2 years of a language, and you will be in advance of a 2 year-old native speaker.

> our brains have to process language at speeds too fast for the conscious mind to keep up

good point for oral communication, but literacy doesn't have to be done in real time

While I agree that adult capabilities for language learning are often underselled, you need to appreciate that a 2 year-old native speaker hasn't got the capabilities to learn the language to the full extent. Yet a 2-year old is likely to have a better phoneme representation – often much better – in their native language than an adult who has studied 2 years in language classes.

At a wedding I met a photographer from the far east who had learnt perfect English in his late teens with (to my native ears) a perfect northern accent. So it is possible.

I suspect that even a tin-eared adult learner could master those phonemes with enough obsessive repetition, both listening and speaking. I'm reminded of Liam Neeson's character in 'Taken' spending a transatlantic journey listening over and over again to the voice of his daughter's kidnapper just so he could recognise him later on.

That level of obsession requires motivation which I think is highly under-appreciated where learning is concerned.

We often think of small children as possessing magic learning brains without considering how highly motivated they must be. The ability to speak means they can more easily make their basic preferences and problems known to parents and so on. It's hard to imagine what it was like for us back in early childhood.

However, children do have one natural advantage: if you learnt a language when young then it must be harder to forget it, due to Ribot's Law:

https://www.youtube.com/watch?v=bYPTzw3iN5Q

Yes, it's possible. What I was trying to say, is that children get almost invariably past a certain level of goodness, whereas with adults, the results vary a lot. And I agree that motivation has a lot to do with it, as well as that children are very motivated.

I agree that for most adult learners, aiming for accent-free is likely to disappoint. However, my personal figure of merit is getting to the point where people are happy conversing with you in their L1/your L2, rather than wishing to swap to their L2/your L1, and any reasonably motivated adult ought to find that attainable.

  no effort: your L1, their L2
  easy mode: your L1, their L1
  aim for:   your L2, their L1
  hard mode: your L2, their L2

(indeed, in my experience —past an acceptable baseline— people judge L2 speech far more by register and rhetoric than by phoneme production)

> children are very motivated

That, and much less afraid that they "might make a mistake"

I suspect that environment has a lot more to do with that than innate capability.

A two-year-old lives in a full immersion environment where the only language they can communicate in is the one they are learning and where a large portion of the people they interact with are going out of their way to help them practice. They also have no responsibilities other than learning language and a few other skills.

Put an adult in an environment like that and they will make extremely rapid progress.