The Immersion Delusion

I’ve written a blog in which I outline five pitfalls new learners commonly fall into when learning Mandarin (which apply to other languages too).

Pitfall number four is a school of thought which I refer to as “the immersion delusion.”

“…The recent rise in popularity of online immersive techniques has led some learners to a new pitfall. I call this the immersion delusion. There is an extreme school of thought which claims immersion is not only a vital component, but the only necessary component of achieving fluency in a language…this is a complete distortion of reality.”

Check out the full blog here and let me know your thoughts: The 5 Most Common Chinese Learning Pitfalls & How to Avoid Them – I'm Learning Mandarin


Great read, thank you. I think all language learners could benefit from avoiding these pitfalls.

I’m certainly starting to rethink how I feel about immersion. I used to think that if you immerse long enough you will automatically be able to speak with some level of fluency. I’m finding that (at least for Chinese) this really doesn’t apply. Early output (while paying careful attention to corrections and feedback on your pronunciation) seems essential.

1 Like

Thanks :slight_smile: Pure inputism is an absolute disaster of an idea and one which has caused great misery and strife to language learners all over the globe. It is totally delusional to think that through input alone you will pick up correct tones, intonation, accurate sentence structure, grammar and correct word usage. Yet this is what is implied, if not outright stated, by some prominent influencers.

The majority of overall learning time should be input but if this isn’t supplemented with other activities the results will be unsatisfactory (assuming your goals rise to the level of being minimally comprehensible when you speak the language.)


I’m sick of being told I just have to go to Greece or Cyprus for a while to learn Greek.
A family member once said a person could learn Greek by picking up a word a day and after years they would be fluent. The metaphor is wrong. It isn’t Lego.
Currently I am thinking of it as a tide (coming in) that ebs and flows but slowly covers (learning) the sand after many repetitions (20 + for me for reading ministories). This suits Complrehensive input.
I find large volume listenign in Greek awful. I might hear 1% of words and can’t process them quickly. If a language is overwhelming, more of it may well be demotivating. Plus it’s what people are telling you to do which cuts off some support.
People long-windedly giving you corrections or explanations in an overwhelming language is highly conterproductive; it sucks up time and is socially awkward while not being communication that you can process.
Babies get 10000 words a day and some of it is slow and clear from context.
I’m using LingQ, 3000 words of reading a day, 20 to 40 reading repetitions of ministories and then listening to them. I’m 2700 known words in (250,000 words read) having done other things prior to LingQ. It’s helping. I’m hoping for 3000 words read a day and words known varies but seems to average 10 a day (not recently. It comes in fits and starts).
It all seems a process of managing familiarity and processing of an acknowledged fog.
Thanks for writing the blog post.
Immersion: I go to Cyprus and talk English to my family and shop keepers, say hi in Russian at cafes. Below a certain level and without reorganising your social situation (family agreements), going to the place can be worse than apps. Below a certain level of a language, going to the place doesn’t create all that many useful conversations. Some good context and motivation but there is an element of delusion. Next time I go, I’m afairly sure I’ll learn more from LingQ and that will be more awkward to use without a home PC and macro buttons (@Zoran please have a read+speak button).


I totally agree with your opinion on the immersion delusion. We’ve all met non-native speakers of a language who have been living in the country of their target language for years and yet still make a lot of grammar mistakes. This is a perfect example of why immersion isn’t everything. You need to take the mistakes that you’re making and drill the corrections into your brain using SRS. Whether that’s through word flashcards for mandarin tones or sentence flashcards for grammar.

Immersion is very good for quickly learning the high frequency structures that occur in your input. But the trouble is that the low frequency stuff doesn’t occur often enough for it to be efficient beyond a certain level. You need SRS to modify the frequency of your input, and drill you more on the things you’re getting wrong, and less on the stuff you know off by heart.


It is extremely expensive to hire a tutor every day for having one hour conversation over italik also for a year who is patient enough to provide me feedback in regards to my mistakes. Does it sound realistic to you?

1 Like

You make an excellent point. Immersion is key but on its own will get you only so far. Repetition activities are crucial for drilling grammar structures, rare vocabulary and tones. Lots of output practice is also hugely important.

1 Like

Good luck with your journey. It’s taken almost 3-years and I’m only starting to output! I could have halved the amount of time if I could just find content decent for learning.

Keep doing what you’re doing.

1 Like

Good points!

I totally agree at this point that immersion alone won’t get a person to outputting.

I’ve read & listened to so much content at this point, but can’t speak since I’ve never practiced.

Sometimes I wonder if it’s because of how the chinese characters work, there is somewhat a phonetic pattern, but honestly you can read without subvocalizing if you forget the pronunciation (or like when Chinese learners start Japanese, the brain auto subvocalizes familiar kanji in Mandarin). I’m kinda curious if massive reading is more effective for speaking for languages like Spanish, German etc.

Luckily, little to no output doesn’t hamper reading comprehension (a thing I was worried about since I only care about reading). SRS, vocab / hanzi drilling are almost required to get up to speed quickly though…I’ve recently figured out how to easily import books into Anki with Chinese Text Analyzer and the Morphman add on to mine sentences easier.


How much is “so much content”? I have a feeling that immersion only produces output if you take it deadly seriously. I suspect most people who are doing “immersion” are actually spending more time/day in their native language than their target language. I doubt that would produce output, at least not a high level of output.

The more I understand about how we learn our native language, the more I realise just how much sustained volume of input we were getting. Trouble is, unless you have next to zero responsibilities, it’s bascially impossible to replicate that.

Therefore, the “end” result isn’t even close to what it is for a native speaker. If we’re spending more time using another langauge than the language we’re learning, it’s no wonder we don’t get the results we want.

I genuinely believe that the brain needs to be tricked into thinking this new language is the only option. The brain is lazy, it won’t get to work on a new language with the kind of intensity required if you’re feeding it another language for the majority of the day. It needs to believe this is a survival issue to fully commit, IMO.

In short, I don’t think it helps that people do 2-3 hours/day of “immersion” and then deduce that it doesn’t work.


Don’t get me wrong, I’m currently doing the “mass input only” approach and seeing where it gets me. But it shouldn’t used as a blanket recommendation to everyone. Immersion is probably the most important activity for language learning, but so is practicing whatever it is that you want to get better at.

I’ve read >10,000 pages, watched near 1500 episodes of Cdramas, done about 200k anki reps and listened to probably at least a thousand hour of passive listening within 500 days of starting…but I can’t magically speak, because I’ve never practiced speaking. I do however suspect that if I did a 1/2 hr to an hr of shadowing practice daily that would vault me to a conversational level quickly.

Although maybe it is just the insane difficulty of Chinese…maybe I do need to read 50k pages (my plan) and listen to 10,000 hours of audio over many years to be able to speak effortlessly?


That’s 2 hours/day of passive listening, which is like 1/5 (or less) of that of a native speaker has had after 500 days. Not many natives can “magically speak” after 100 days either.

10k pages sounds like a lot, and it IS a lot, but there’s a lot, then there’s a LOT. I feel like we’re still grossly underestimating just how much input a native speaker gets. It’s absolutely MASSIVE.

Native children are usually pretty much conversatioallly fluent by the time they’re 5 or 6 years old, but that’s a HUGE amount of input that dwarfs what you, I, and pretty much every second language learner has done over a similar time period.

Again, I’d estimate that even if we go hardcore and do like 4-5 hours/day, every single day, we’ve still only had about 1/3 of the input they have, and that’s in a less than ideal environment, spending a large proportion of that time immersed in a different language to the one we’re actually trying to learn.


I will admit I haven’t been doing as much listening practice as I should yet, as I’m not quite convinced passive / incomprehensible audio is anywhere near as effective as reading / SRS for acquiring vocabulary. (Chinese has a LOT of similar sounding words)

My theory has been that I’d read an insane amount until I’m near fluent, then start doing more and more listening until I can listen to native podcasts and audio books without needing a script, then try speaking to see how well the mass input only method works.

I’m having an absolute blast doing the input only method, because I love Chinese media and don’t care about speaking, I just warn people that have a priority / urge to speak to go ahead and practice, especially shadowing as mouth & tongue positions are different and need to be exercised (not to mention matching the correct tones). I can hear Mandarin in my head, I just can’t spit it out without it sounding absolutely terrible. (And I have been going crazy hardcore, I’m probably at at least 4000 hours at this point, I am constantly immersing in some form of Chinese all day, except when I dabble in other languages or my part time job / socializing.)


But you’re warning them of this based on limited input. What you’ve described is still a volume issue, IMO. I mean, can you understand Mandarin as well as a native 6 year-old? Have you acquired the grammar like they have? My guess is a resounding no. IMO, the overall contributing factor to the difference is the sheer volume of comprehenisble input they’ve had Vs what you have so far had.

In a way, you’re right, because it’s near impossible to match a native for volume of input and quality of immersion. So unless one can do like 8+ hours/day for 5-6 years (for a language like Mandarin), then a high level of natural output is unlikely to emerge. But that doesn’t mean, therefore, that ‘immersion’ doesn’t produce output, it simply means that ‘low volume/poor quality’ immersion doesn’t produce output.


I have never practised speaking English. After cca 2 years I am better speaker then friend of mine who practised a lot. Just out of curiosity I started practise speaking Russian which I learn for shorter time. The result is that my Russian is way more horrible than English even if it is very similar to my native language (Czech).


Imho, immersion learning is just a fad.

Reading about immersion learning in Internet forums and listening to gurus on Youtube, has left a rather negative impression on me. There seems to be a lot of fetishizing over big numbers, hero worship, and ostracizing of non-believers. I haven’t seen much of value, tbh.

What the proponents of immersion learning fail to prove is, that it is the immersion that makes the difference, and not the 10.000+ hours.

It also remains completely unclear how immersing for x hours results in one’s speaking. Where is the logical connection?
Let’s get this straight, you painfully avoid doing an activity for years, then, on the day of a blue moon, you wake up and start speaking fluently? That just doesn’t add up.

It’s probably just me, but I don’t get it.
Therefore, I’ll treat immersion learning just like Dogecoin, Anki and other schemes. I’ll just sit this one out and let everybody else get rich. Too much belief, too many gurus.
Don’t mind me, while I wait for the next big thing…

1 Like

I am not sure if it is a fad but there is living proof of Japanese learners who could speak fluent Japanese by following the immersion route.


My opinion, based on interviewing a number of high level learners of east asian languages is that immersion-only learning simply doesn’t exist.

At least I’ve seen no evidence for it. No adult I’ve ever come across actually learned like a baby. All the most extreme AjAATers supplemented hours of immersion with daily drilling of whole sentence flashcards. The results of this hybrid approach were they became fluent but didn’t acquire pitch-accent - a key component of Japanese pronunciation.

Extreme AjAAT approaches are almost certainly quite inefficient. They are effective precisely because the people using them spend all of their waking hours (and in some cases sleeping hours) engaging with the language and supplement them with SRS.

Also, in dogmatically refusing to ever use alternative techniques - such as study grammar or rely on feedback and corrections from native speakers - they are closing the door on techniques which could make their learning much more efficient. Most people don’t have the luxury of being able to ignore questions of efficiency in favour of spending all their time learning the language because most people have responsibilities. The message from extreme Ajaaters as consistently been: if you have other resonsibilities you can’t learn the language.

But in recent times new cases have come to light of incredibly high achieving learners who didn’t dogmatically follow immersion-only approaches. One example is the YouTube Aussieman who promotes immersion combined with early output. Another example is Will Hart, a Mandarin learner I interviewed on my podcast who learned Mandarin while studying on a medicine degree and reached fluency within a year: Interviewing This Master of Oral Chinese Made Me Rethink Everything I Believed About Language Learning – I'm Learning Mandarin

The immersion delusion has therefore reached a dead end.


@TofuMeow Thanks for commenting. Very interesting to hear your views on this as somebody who has taken an immersion heavy approach. Keep us updated on your progress. And the door is open if you ever want to discuss this on my podcast in future.

1 Like

Great idea! I’ve been been following TofuMeow’s progress for more than a year, and there is no doubt in my mind that she’s a language learner extraordinaire, certainly the most successful Chinese learner in recent years here on LingQ and an inspiration to many.

1 Like