Right, so, the bit of my brain that has been immersed in science is shouting “quackery!” at all the anecdote and hypothesis being asserted as firmly established facts. It may be completely correct—it isn’t; there are obvious problems—but it isn’t proven, and to write as if it is is misleading, however well intentioned.

You can learn to ride a bike with one hand tied behind your back. You can point to numerous children who have done exactly that and say “See?” But to then suggest that everyone absolutely should learn to do it that way, and that using both hands first is somehow harmful to learning cycling—since it leads to a less-refined sense of balance or something—is to draw a dubious conclusion from verifiable observations.

Now, I still think a lot of the ideas presented are useful, particularly for less-confident learners. For example, for folks easily put off by making any mistakes, avoiding feedback mechanisms with an extended familiarisation period seems like it would be quite beneficial. Demoralisation is incredibly destructive to learning, so methods that help maintain motivation should be encouraged even if they’re not ideal. Accepting that mistakes are inevitable and that confronting them should not be a cause for demoralisation might be better, but insisting on that “just git gud” approach is useless if it ends up being off-putting: the alternative of no progress at all is clearly the worse outcome.

And I write this as someone following an alternative approach to language learning, one that is also deprioritising speech, though for different reasons, with less dogmatism, and with less initial emphasis on listening as well. My personal prioritisation is reading, writing, listening, speech. I’m following the comprehensible input approach of taking small steps into unfamiliar territory, but since my target language is Japanese, I see the writing system as more outside my prior experiences than the sounds and, therefore, more in need of initial familiarisation. And, since reading and writing are also less time-sensitive than listening and speech, my reasoning is that they’re the easier starting point.

However, I do not subscribe to the thinking-is-bad, just-absorb-it approach, so while I am prioritising reception over production, I’m not purposefully avoiding either writing or speech. In fact, I’m finding that trying to put my thoughts into words is actually very helpful in building a sense of my progress. (A running count of hours or words seems far too abstract.) And while working through the kanji (essentially a 2000+ item vocabulary list) is getting more of my attention than explanations of grammar, I’ve found that conscious analysis of the language useful as a framework for learning. It’s certainly helped having a place for new ideas already marked out before encountering them. That said, the more vocabulary I learn, the more easily the example sentences demonstrate the grammar.

I did try the approach of starting with books for babies and building up from there, but since I do not have the experiences and wants of a baby/child, I found them (aside from a few funny exceptions) generally quite tedious. However, by picking up some understanding of grammar and knowledge of vocabulary, I’ve been able to turn other sources into comprehensible input, sources that I do actually find engaging, which I agree is key.

So, while I certainly wouldn’t discourage anyone from using (very useful) immersive approaches, and agree that learning a language is not the same as studying linguistics, I think the stubborn, absolute insistence on delaying certain aspects of learning is misguided, and promoting the notion that you can passively absorb a language because that’s what children do—ignoring the plentiful interactions and feedback that kids actually encounter—and because he thinks that’s what worked for him individually, strikes me as less than ideal.

But, like I said earlier, if this approach works for you personally, and other approaches do not, then what does it matter whether it is the most efficient? Effectiveness should be the more important consideration.

An older discussion from the same forum:

Pablo from Dreaming Spanish: Language can’t be explained

At least they’re not using Duolingo

I’ve had massive success with Dreaming Spanish.

Is it the most efficient method? Dunno. But it is one of the easiest to STICK WITH as a solo learner. Maybe there’s better methods but if you burn out on them they’re not useful at all. But everyone is different.

I stopped strictly tracking at around 1000 hours (and over a million words read from books) but I can understand pretty much any Spanish language media. Some things will still go over my head but I get them from context.

My speaking lags behind but that’s mostly down to my lack of practice lol. Binging YouTube/shows is much easier than actually socializing. But I can definitely make it through convos with a bit of stumbling. And it seems to improve much faster than the comprehension process.

Feel free to ask anything if you’d like.