Learning Chinese through comprehensible input

There are two ways you can get Chinese into your brain: through your ears by listening and through your eyes by reading. You can only say and write things you already know, after all.

For this input to be effective, it needs to be comprehensible. When you are able to connect spoken or written forms of words with their meaning in context, you gradually build your own mental version of Mandarin.

As you engage more with the language, your brain will gradually figure out how things work based on the input.

Tune in to the Hacking Chinese Podcast to listen to the related episode (#264).

Available on Apple Podcasts, Spotify, YouTube and many other platforms!

For example, if a beginner sees a teacher point to themselves, saying wǒ, and then point to the student and say nǐ, the student can connect how these words sound with what they mean.

Similarly, if a student reads a text about 中国, they can figure out what the word means by context, perhaps because there is a map of China with a flag on the page.

Comprehension through context, not translation

Input in the form of spoken or written language can be made comprehensible in many ways, including pictures, body language and contextual clues.

Routinely relying on direct explanation or translation using another language is out. While telling you that wǒ means “I; me” leads to comprehension, it does so through English, and the goal is to understand Mandarin.

So, learning Chinese through comprehensible input means that you engage with as much language you can make sense of as possible. Everything else will take care of itself. Over time, it will naturally convert to other areas, including speaking and writing.

If a teacher uses a comprehension-based method (there are many), they will do their utmost to help you with this, but they will typically not translate much, explain all the grammar or encourage you to drill sentence patterns you have just encountered.

There are many approaches centred on comprehensible input

Before we continue, it is worth highlighting the fact that there are many different methods for learning and teaching Chinese which have comprehensible input at their core:

• Some want you to understand every single word, others think the gist is enough.
• Some say that producing language is only useful if it gives you more input, others say that practising speaking and writing directly can be helpful.
• Some focus on input only, while others think that interaction in the target language is valuable in itself, beyond the extra input it provides.
• Some prefer to stick to only the target language, arguing that using any other language, such as through translation, does not help and can even be harmful. Others don’t have a problem with translation as a stepping-stone to comprehension.
• Some claim that studying the language explicitly (such as grammar rules or pronunciation) has no effect on your learning. In this view, explicit knowledge can’t transform into implicit knowledge, which is what you need when you produce language spontaneously. Others think that explicit learning can build implicit knowledge under certain conditions.
• Some think that the language has to be authentic (not produced specifically for language learning), but others think comprehensibility is more important.

My goal in this article is not to argue in favour of a specific method, but rather to give you an idea of what comprehensible input is and what it can offer you as a student. You can then pick and choose which parts to implement in your own learning.

Comprehensible input through listening and reading

The most important point is that input is extremely important. Even if you do not buy the argument that speaking will emerge naturally when you have listened enough, the listening bit is still crucial.

No one who researches second-language acquisition (SLA) seriously doubts that input is of critical importance. As I said in the introduction, that is the only way your brain can get data about how Mandarin is used. When you speak and write, you can, by definition, only use language you already know.

Stephen Krashen’s input hypothesis

Comprehensible input is closely associated with a set of hypotheses by linguist Stephen Krashen, originally published in the 70s, but later developed further both by Krashen himself and other researchers.

The first hypothesis is the input hypothesis, which states that we make progress in a language by engaging with input that is more advanced than our current level, but not so advanced that we don’t understand.

This is reflected in the formula i+1:

• i represents language you already know
• +1 represents the next step in your learning process

Ideally, then you should listen to and read Chinese that is on your level, but yet contains things you don’t know. That way, you can comprehend and acquire the unfamiliar by relying on the familiar.

Update: In earlier versions of this article, I said that “i” stood for “interlanguage”, which is not right. Diane Neubauer actually emailed Stephen Krashen about this, and it turns out “i” does not stand for anything; it is just an arbitrarily chosen letter.

Are you learning or acquiring Chinese? What’s the difference?

Before moving on, it is worth mentioning one of Krashen’s other hypotheses: the acquisition-learning hypothesis. To put it briefly, he claims that acquisition and learning are two distinct processes.

• Acquisition is subconscious and is the result of comprehensible input
• Learning is conscious and deliberate; what most people would call studying
• Language development is only dependent on acquisition, not learning

It should be noted that most people, including most SLA researchers, do not accept this dichotomy and use “learning” to include “acquisition” as well.

I am no exception, as can be seen from the title of this article: Learning Chinese through comprehensible input. I bring up the distinction here because it’s very important to understand what Krashen is saying (and what he isn’t).

Yes, but…

There are, in fact, many people who do not agree with several points brought up so far. Let us have a look at some possible objections. I will also provide my own take on each, based on my reading of the research and my experience learning and teaching Chinese and other languages.

…what about explicit learning and studying?

How much of an effect, and for what aspects of learning, does explicit studying have? If you look at how most people learn Chinese, both inside and outside classrooms, it’s clear that most students and teachers believe that it has a big effect.

Both students and teachers typically believe that explicitly studying things like grammar and pronunciation will help them develop language ability. Why would it not?

It might surprise you that this is not the general consensus among SLA researchers. While few say that explicit studying has no effect whatsoever, most still think that the effect is limited. We simply do not learn grammar by studying grammar rules.

As some argue (see VanPatten, 2014a, for example), the rules for grammar our brain relies on when producing language do not even remotely resemble the rules you see in your textbook, so there is not even a theoretical ground for believing that studying them will help your language development.

This does not mean that there is no room for focusing on grammar, but it needs to be done in a different way than the traditional one where the teacher tells you how something works (present), you do some exercises and drills (practice), and you then use the language more freely (production). For an alternative, see VanPatten (2014b), about input processing.

Based on my own fairly extensive reading of SLA literature, most researchers support a limited effect of explicit learning on language development.

The idea is that teaching or learning something explicitly can help you notice it in subsequent input, making that input more effective for learning.

According to Schmidt’s noticing hypothesis (Schmidt, 1990), this kind of conscious noticing is necessary for language acquisition to occur: we don’t learn what we don’t notice.

In usage-based approaches, however, noticing is just one factor among others, such as frequency, salience, and prior experience. Attention is sometimes needed so students can notice “blocked, overshadowed, or otherwise non–salient aspect of the language form” (see Ellis and Wulff, 2014).

While attention still plays an important role, these approaches allow for more learning to happen implicitly.

In both cases, though, input remains essential. Without meaningful exposure to language in use, explicit learning alone does not lead to lasting acquisition.

My take on explicit learning of Chinese

Students and teachers vastly overestimate the importance of learning about the language. You should strictly limit how much time you spend learning about Chinese in your native language.

Still, when you encounter issues, including with grammar and pronunciation, studying the point you are struggling with explicitly may be helpful, but only if you’re developmentally ready for it. Ideally, this would be embedded in a meaningful context, too, not studied in complete isolation.

For example, I am pretty sure that I would never have fixed my problem with the third tone without somebody explicitly pointing it out to me. This does not mean that tones are not mainly learnt through input, just that directing attention can be crucial too. This is a good example of what Ellis and Wulff (2014) meant by attention and noticing being necessary in some cases.

…what about output and interaction?

For most people, the most surprising part of the approach outlined so far is that speaking and writing are viewed as having little or no effect on your acquisition. In a way, this is obvious. As mentioned, the only way you can feed data to your brain is through listening and reading.

Still, the idea here is not just that you learn completely new things through input, but that enough comprehensible input also leads to the ability to speak and write in a new language, much more so than directly practising speaking and writing would.

This is counter-intuitive for many and goes directly against the advice offered by most polyglots and other people with lots of experience learning languages, who often argue for speaking from day one.

That, however, is at least partly a reaction to institutional language learning, which is focused on passive knowledge and theory. If you have learnt Spanish in school for six years but cannot use the language, it is not because you did not practise speaking from day one; it is because you learnt things about Spanish through English. You don’t have to be a CI-purist to think that’s bad.

My take on output and interaction when learning Chinese

I agree that input is much more important than output, even if your goal is to improve your speaking and writing. That is why my article 20 tips and tricks to improve your Chinese writing ability starts with encouraging you to read more before even thinking about improving your writing ability.

Still, to me it seems obvious that there is a skill component to speaking and writing that can and should be directly practised. This does not mean that I think it’s a good idea to drill words and phrases you just encountered in your textbook, but that directly targeting fluency and automaticity is useful once you have acquired the necessary language.

For example, classroom language can be useful to practise explicitly to increase fluency, automaticity and confidence, but this should be done after you’ve already been exposed to it in class.

Another example of output being useful is that most students who have lived in an immersion environment are very good at explaining how and why they started learning Chinese. This is not because they have listened to many people tell their stories, but because they themselves have told their story so many times that they are considerably more fluent in that area than most others.

The definition of “fluency” that I use here is essentially that it is about how well you know the things you know. How fast can you connect a spoken word with its meaning in context, how quickly can you come up with the right word for “passport” in Chinese when you need it, and how long it takes you to connect written characters to meaning in a text. And so on.

How to become fluent in Chinese

Fluency can and should be practised. The words we use often are easier to recall, not because we hear them all the time, but because recall becomes faster each time.

A good example of a fluency-oriented exercise is so-called 4-3-2 activities, where you talk about something for four minutes, then try to say the same thing in three minutes, and then, finally, you say the same thing in two minutes.

Even without any input or feedback, the third attempt is invariably much more fluent than the first. No acquisition took place, but your fluency still increased. Note, however, that fluency development is supposed to focus on language you already know!

Another example is certain word games that are great for targeting fluency specifically.

So, to summarise: Yes, input is all-important, but targeted output practice is also essential for building fluency and confidence.

…what about vocabulary and flashcards?

As we have seen, in many comprehension-based approaches to learning and teaching languages, studying grammar explicitly is believed to be ineffective, indeed impossible.

Vocabulary is a little bit different, however. Comprehensible input is still the main driver of acquisition, and so the best way to learn words is to encounter them frequently when you read and write. Listening and reading are in themselves a kind of spaced repetition that enables you to expand vocabulary rapidly.

Still, while explicit studying of grammar is sometimes deemed to have zero impact on learning beyond the input itself, many comprehension-based approaches are more open to focusing on vocabulary as a support for comprehension.

While I do not know any comprehension-based teacher who is also a strong advocate of flashcards, this seems to be more because input is just better, not because a supplementary focus on vocabulary would be utterly useless.

My take on vocabulary and flashcards for Chinese learners

As I have outlined in a short series of articles, I believe flashcards are great for learning the basic definitions of both spoken and written words. This greatly accelerates your learning and grants access to more listening and reading materials. Flashcards are, in this sense, a stepping-stone to more input.

The only major exception is handwriting Chinese characters, where I think using flashcards is by far the best method available.

Why spaced repetition software is uniquely well suited to learning Chinese characters

Conclusion: Input is king, but it is not everything

I think that a lot of good has come from Krashen’s input hypothesis. While there are some aspects that I do not agree with, the general emphasis on meaning-focused and comprehensible input over grammar explanations and drills is much needed.

If you are not already listening and reading a lot, you should start by doing so, almost no matter what your ultimate goal is. Listening more will improve your speaking, and reading more will improve your writing.

If you want to be able to use what you have learnt through input sooner rather than later, you also need to practise speaking and writing directly, preferably in interactive contexts that allow you to develop communicative strategies as well.

Your ability to use the language to understand and communicate successfully is not only about having absorbed information through input, it is to a large extent also about how quick, automated and confident you are at retrieving that information and using it in real time.

Fluency, in other words, and fluency comes through practice.

Further Reading

For more about comprehension-based methods for learning and teaching Chinese, check out the series that starts with this article, written by Diane Neubauer.

An introduction to comprehension-based Chinese teaching and learning

Ellis, N. C., & Wulff, S. (2014). Usage-based approaches to SLA. In Theories in second language acquisition (pp. 87-105). Routledge.

Lichtman, K., & VanPatten, B. (2021). Was Krashen right? Forty years later. Foreign Language Annals, 54(2), 283-305.

Schmidt, R. W. (1990). The role of consciousness in second language learning. Applied linguistics, 11(2), 129-158.

VanPatten, B. (2014a). The limits of instruction: 40 years after “interlanguage”. In Interlanguage (pp. 105-126). John Benjamins Publishing Company

VanPatten, B. (2014b). Input processing in adult SLA. In Theories in second language acquisition (pp. 125-146). Routledge.

Editor’s note: This article, originally published in 2016, was rewritten and republished in August 2025.