“I think in English and then translate to Chinese when speaking.”
“When someone speaks to me, I have to translate before it makes sense.”
These are two sides of the same problem. When speaking, you know what you want to say, formulate it in English and then translate it into Chinese. When listening, what you hear has to pass through English before making sense.
This is a problem because translation is slow.
There are situations where this is not so serious, such as when writing or reading and you have plenty of time, but it can be a problem even then.
When speaking, constantly translating is too slow for fluent speech. When listening, mental translation is also associated with worse listening performance (Vandergrift et al., 2006; In’nami et al., 2023).
So how can you stop translating and start thinking in Chinese?
Tune in to the Hacking Chinese Podcast to listen to the related episode (#321).
Available on Apple Podcasts, Spotify, YouTube and many other platforms!
Removing English entirely is the wrong goal
Trying to remove English from the process entirely is the wrong goal, however. Research on bilingual speech has found that even highly proficient bilinguals activate both languages when speaking only one of them (Jacobs et al., 2016).
Instead, you should try to stop English from being a necessary intermediate step. Words in the “wrong” language popping up when speaking and listening are not problematic, but feeling that you have to translate everything is.
It’s also important to note that translation in and of itself is not bad. In fact, one of the major benefits of knowing Chinese is that you can translate to and from the language, for example, to help other people understand what something means in context. What we’re talking about here is when English becomes an obstacle to communicating in Chinese.
You can already communicate in Chinese without translating
Being able to speak and listen without translating is not the Holy Grail of language learning; it’s something you probably already do.
If someone says 你好 (nǐ hǎo), “hello”, do you really need to translate that before it makes sense to you? And if you want to thank someone in Chinese, do you first think “thank you”, and only then 谢谢/謝謝 (xièxie), “thank you”?
Your goal is to extend this ability so it covers more and more of what you want to produce or understand. This is not easy, of course, but the fact that you already have this ability for parts of the language means that it can be done!
Four obstacles to stop translating and start thinking in Chinese
I think there are four main obstacles:
- You don’t know enough Chinese yet.
- The Chinese you do know isn’t automatic enough.
- You get too little meaning-focused practice.
- You’re not able to tolerate ambiguity and approximation.
These obstacles overlap, as do the remedies that I will propose for each. It’s also likely that you still translate for more than one reason, so check all four obstacles before deciding what actions to take!
Here’s an overview of the whole article:
- Obstacle 1: You don’t know enough Chinese
- Obstacle 2: The Chinese you do know isn’t automatic enough
- Obstacle 3: You get too little meaning-focused practice
- Your goal should be to communicate, not express something already formulated in English
- Information gaps make sure you’re focusing on meaning
- Language-focused learning can be useful as well
- Many classroom and textbook activities focus exclusively on form
- Analyse and balance your Chinese learning with Paul Nation’s four strands
- Obstacle 4: You have too little tolerance for ambiguity and approximation
- How to stop translating and start thinking in Chinese
- References and further reading
Obstacle 1: You don’t know enough Chinese
This is obvious, but often overlooked: if you don’t know the words and grammar needed to say something in Chinese, you can’t think it in Chinese either.
Your inner monologue in your native language is likely far more sophisticated than your Chinese can carry, so trying to just convert your thoughts to Chinese is not realistic until you reach a very advanced level indeed.
Practise how to use the Chinese you have to communicate
So, part of the answer to how to stop translating is simply to improve your Chinese in general, even though I realise this is not very helpful. The good news is that improving what you can do with the Chinese you already have can be at least as useful.
The goal when speaking (and thinking) is not to actually produce something equivalent to what you would have said (or thought) in English, but rather to communicate some kind of meaning.
Therefore, any activity where you practise getting meaning across with the limited linguistic resources you have is worth looking into. My favourite by far is playing word-guessing games where you are supposed to explain a word without using the word in question. This is excellent practice for talking around words you don’t actually know how to say!
Create and expand smaller areas where you think in Chinese
As mentioned already, you already think in Chinese at least some of the time. It’s not hard to expand this to new areas either, even as a beginner.
As soon as you’ve learnt numbers and dates, or words for exercise, cooking and other limited everyday domains, you can use them to think in Chinese in limited contexts.
For example, start counting everything in Chinese: scoops of coffee, push-ups, steps up the stairs at work, and so on. Depending on your level, you can start doing this for all sorts of activities you engage in often. I wrote more about verbalising your everyday life here: The cheapest and most convenient way to improve your spoken Chinese.
The cheapest and most convenient way to improve your spoken Chinese
If you can’t say it in Chinese, say something else
Simplifying your language and focusing on what you already know only takes you so far, though. What if, in class, you’re asked what you think about environmental protection? Or if someone asks you what the cultural differences are between China and your home country?
Saying 环保很好/環保很好 (huánbǎo hěn hǎo), “Environmental protection is good”, or 这个问题有点复杂/這個問題有點複雜 (zhè ge wèntí yǒu diǎn fùzá), “That’s a bit of a complicated question”, might be the only things your Chinese allows you to say that are also true.
One solution here is to stop trying to say what you would have said in English, thereby avoiding the translation question entirely. In essence, just role-play and pretend your standpoint or opinion is something other than it is. This avoids frustration and allows you to practise using the Chinese you know.
Your Chinese proficiency puts a hard limit on listening comprehension too
For listening, the same proficiency limit applies. If you listen to Chinese way above your head, avoiding mental translation won’t help much. You need comprehensible input and more listening.
Obstacle 2: The Chinese you do know isn’t automatic enough
Imagine I show you a picture of a ball. An advanced learner might simply say 球 (qiú), “ball”. A beginner who technically knows the word might instead go through several steps: “That’s a ball. How do I say ‘ball’ in Chinese? Right, 球.”
Knowing a word on a flashcard or a test is not the same as knowing it well enough to process or retrieve it quickly in real language use. On a flashcard or test, the intermediate translation step won’t stop you, but in actual conversations, it will.
Merely knowing a word is not enough; you need to know it fast
The point here is that the more often you connect sound and meaning (when you listen) or meaning and sound (when you speak), the faster the process becomes. When you’ve heard or said a word hundreds of times in different contexts, it’s highly unlikely that you will need to pass through English before understanding it or saying it yourself.
Here, quantity is a problem. The way many people study means that they only see words a few times in a textbook or a dozen times in a flashcard app. This just isn’t enough, and processing a word in the same context over and over is not the same as doing it in varied contexts.
The solution here is very large amounts of meaning-focused input and output (see the next obstacle). The more times you successfully process a word, the faster that processing becomes.
Words you hear often, for example because your teacher, friend or co-worker says them often, quickly become second nature. When you have told twenty people why you started learning Chinese, you’ll be pretty good at it.
You need to expand this to all areas of the language where you want to operate fluently. When you’ve said something fifty times in different contexts, you won’t have to rely on translation to say it.
Automatic processing is crucial for successful listening comprehension
Listening comprehension is extra sensitive to slow processing. When you speak, you can pause before saying the right word, and when you read and write, you’re in even more control of time, but when you listen, you often can’t influence how quickly the other person speaks.
If you spend three seconds processing a tricky word, the other person might have said an entire sentence. If you process the language more slowly than the rate of speech, you will fall behind and get lost almost immediately.
I wrote more about automated processing in listening comprehension here: Beyond tīng bu dǒng, part 4: Learning to process spoken Mandarin quickly and effortlessly.
Beyond tīng bu dǒng, part 4: Learning to process spoken Mandarin quickly and effortlessly
Obstacle 3: You get too little meaning-focused practice
The first two obstacles we have discussed are about what Chinese is available to you and how fast you are at accessing it. This third obstacle is about what kind of activities you engage in when practising.
Quantity is important, both for speaking and listening, but it also matters how you practise! Many common classroom, textbook and course activities reinforce translation as a necessary mental step.
In essence, you should engage in activities where you care much more about what you say (the message) than how you say it.
Meaning-focused input works the same way, but in reverse. When listening or reading, the goal is to understand the message, not to account for every word or produce an English version of it. The language also needs to be mostly familiar so you can focus on what is being said.
Your goal should be to communicate, not express something already formulated in English
For speaking, this also changes what counts as success. In most courses and almost all tests, success means producing the right forms (correct word choice, correct grammar, correct pronunciation, and so on).
This is what focus on form looks like.
When you focus on meaning, success is when the other person understands what you say. If that’s your goal, preserving something originally conceived in English is not very interesting. You have something you want to say (an idea, an opinion, information of some kind). Use the Chinese you have to convey it. There is no canonical English phrase to translate.
Translating also makes the task harder than it needs to be. When translating something from English into Chinese, you are no longer just trying to express an idea; you are also trying to preserve the wording and structure of the English original. Chinese and English often organise the same information differently, with different word order, grammar and choices about what needs to be said explicitly.
Starting from the meaning instead of English gives you much more freedom to express it in a way that actually works in Chinese. This is actually true when your task is to translate something as well: identify the underlying meaning and express that instead of translating!
Information gaps make sure you’re focusing on meaning
That definition of success also tells you what useful practice looks like. An actual conversation, telling somebody a story, playing a board game or sharing an experience all give you a reason to communicate something.
If you know something the other person does not, there is an information gap, and when Chinese is what you use to bridge that gap, the language becomes a means of genuine communication, not an object of interest in and of itself.
The word-guessing game from obstacle 1 should be mentioned here for the same reason. There is no predetermined sentence you need to produce. If the other person guesses the word, you have succeeded. The focus is on meaning.
Real communication: What it is, why you want it and how to get it
Language-focused learning can be useful as well
Meaning-focused practice as described above is essential, but it doesn’t completely replace deliberately studying the language.
For example, flashcards are an efficient way to establish basic form-meaning connections. Still, there is no compelling reason to avoid your first language when learning the basic meaning of words.
At first, it might seem like using Chinese-only definitions would help to avoid translation later, but keep in mind that flashcards should make up a tiny proportion of your exposure to and use of words. Most of the processing that eventually makes the word fast and direct should happen elsewhere, through listening, reading and actually using Chinese.
Flashcards are a stepping stone to more listening, speaking, reading and writing; nothing more. In fact, I strongly recommend that you don’t use monolingual flashcards for learning Chinese. There is no good Chinese learner dictionary, and your definitions will end up being more difficult than the words you’re trying to use.
Pictures can help, especially for concrete words, but they might still not be worth it because of the extra hassle involved in creating and managing such flashcards. And as already mentioned, most processing should happen outside of flashcards entirely, so exactly what you put on them or how they are formatted is of less importance.
Use flashcards for what they are good at, then get the meaning-focused input and output elsewhere: graded readers, understandable listening, conversations, stories, games and other activities where the point is actually to understand or communicate something.
Many classroom and textbook activities focus exclusively on form
One of the problems with the way Chinese is often taught in textbooks and classrooms is that the focus is almost exclusively on form. You are supposed to translate something to or from English, or use a new word or grammar pattern correctly, rather than say something interesting.
This type of practice teaches you to focus on form. If most of your practice involves translation and thinking about exactly how to word things to get them right, you train yourself to rely on translation to or from English.
Analyse and balance your Chinese learning with Paul Nation’s four strands
This distinction between meaning-focused and language-focused learning is part of Paul Nation’s four strands model. Since I mention this regularly, I won’t go into details in this article, but you can read all about it here: Analyse and balance your Chinese learning with Paul Nation’s four strands.
Analyse and balance your Chinese learning with Paul Nation’s four strands
Obstacle 4: You have too little tolerance for ambiguity and approximation
Tolerance of ambiguity is about how well you can deal with things you don’t know. For example, many learners’ listening comprehension can be completely derailed by one unknown word.
What did that word mean? Have I heard it before? What’s the English equivalent?
And bang, you missed two sentences and have no clue what the speaker is saying anymore.
Tolerance of ambiguity is associated with higher language proficiency
Tolerance of ambiguity in second language acquisition has been studied for a long time (see e.g. Wei et al., 2022). In a study specifically of learners of Chinese in Taiwan, higher ambiguity tolerance predicted higher Chinese proficiency and greater use of Chinese-oriented learning strategies, with less reliance on the first language (Chu et al., 2015).
When speaking, low tolerance of ambiguity, or perhaps approximation, shows up as an unwillingness to accept a less precise solution. You know what you would say in English and keep searching until you find the words that express exactly what you want to say.
Except you probably won’t, and you will spend most of your time thinking about how to translate an English sentence into Chinese as precisely as possible.
Practise being okay with ambiguity and approximation
Sometimes precision matters, but ordinary conversation usually gives you more freedom. Nobody cares what you would have said in English or whether you know exactly what every word someone said to you meant.
Paraphrase, simplify, say something slightly different and keep going. Or, as mentioned earlier, role-play and make something up!
You know the deal by now.
And yes, word-guessing games work really well here, too.
Fluency development should include an element of time pressure
Time pressure makes it easier to focus on meaning and accept ambiguity and approximation. If you have twenty seconds to communicate something, you don’t have time to phrase something in English and then think about how to translate it into Chinese.
This is one of the reasons word-guessing games work so well: you simply don’t have time to think too much, which forces you to rely on the Chinese you have, no matter how imprecise it is.
How to stop translating and start thinking in Chinese
If you translate too much and want to start thinking in Chinese, simply telling yourself to do that won’t work.
Instead, try to figure out what the problem is: missing Chinese, slow retrieval, too little meaning-focused use, or an unwillingness to accept ambiguity and approximation. Or a combination of some or all of them.
Then practise in a way that targets your weaknesses.
The goal is not to entirely eliminate English from your mind when using Chinese; it’s just to remove it as a necessary step.
This will happen naturally as your Chinese improves, but you can also do a lot about it right now:
- Get better and faster at what you already know
- Make sure you get enough meaning-focused input and output
- Train yourself to tolerate ambiguity and approximation
And remember, if you don’t need to translate 你好, this means you can already think in Chinese. The task is to make more and more of the Chinese language accessible in this way.
References and further reading
Chu, W.-H., Lin, D.-Y., Chen, T.-Y., Tsai, P.-S., & Wang, C.-H. (2015). The relationships between ambiguity tolerance, learning strategies, and learning Chinese as a second language. System, 49, 1–16. doi:10.1016/j.system.2014.10.015
Ely, C. M. (1989). Tolerance of ambiguity and use of second language strategies. Foreign Language Annals, 22(5), 437–445. doi:10.1111/j.1944-9720.1989.tb02766.x
García-Gámez, A. B., & Macizo, P. (2022). Lexical and semantic training to acquire words in a foreign language: An electrophysiological study. Bilingualism: Language and Cognition, 25(5), 768–785. doi:10.1017/S1366728921000456
In’nami, Y., Cheung, M. W.-L., Koizumi, R., & Wallace, M. P. (2023). Examining second language listening and metacognitive awareness: A meta-analytic structural equation modeling approach. Language Learning, 73(3), 759–798. doi:10.1111/lang.12548
Jacobs, A., Fricke, M., & Kroll, J. F. (2016). Cross-language activation begins during speech planning and extends into second language speech. Language Learning, 66(2), 324–353. doi:10.1111/lang.12148
Kim, H. S., Lee, J. H., & Lee, H. (2024). The relative effects of L1 and L2 glosses on L2 learning: A meta-analysis. Language Teaching Research, 28(1), 7–28. doi:10.1177/1362168820981394
Vandergrift, L., Goh, C. C. M., Mareschal, C. J., & Tafaghodtari, M. H. (2006). The Metacognitive Awareness Listening Questionnaire: Development and validation. Language Learning, 56(3), 431–462. doi:10.1111/j.1467-9922.2006.00373.x
Webb, S., Yanagisawa, A., & Uchihara, T. (2020). How effective are intentional vocabulary-learning activities? A meta-analysis. The Modern Language Journal, 104(4), 715–738. doi:10.1111/modl.12671
Wei, R., Kang, Y., & Wang, S. (2022). The relationship between tolerance of ambiguity and multilingualism revisited. System, 107, 102798. doi:10.1016/j.system.2022.102798
Yanagisawa, A., Webb, S., & Uchihara, T. (2020). How do different forms of glossing contribute to L2 vocabulary learning from reading? A meta-regression analysis. Studies in Second Language Acquisition, 42(2), 411–438. doi:10.1017/S0272263119000688
