Duncan Parrish is an educational technologist and coordinator of Arti Languages, a project using mobile games and cognitive science to enhance how we immerse in Chinese. In this guest post, he introduces the concept of “micro-fluency” and argues that it can help us set achievable and useful language goals.
On the Trans-Mongolian train from Ulaanbaatar to China one December, I was enthusiastically asked by a Mongolian tour organiser if he could practice his English. His vocabulary was modest, but he made it count, telling funny stories, filled in here and there with Russian and Chinese.
We moved to the buffet car, whose only other customer was a hardy Swiss tourist (hardiness being a prerequisite for Mongolian winters). A hearty discussion in French soon began.
Hang on, I thought, I probably know more words of French than my Mongolian friend, but I’ve never had a conversation in French with such confidence and wit. And then they switched to German…
You don’t always need many thousands of words to enjoy yourself (and impress others) in a language. But what you do need is “micro-fluency”.
What is “Micro Fluency”?
Micro-fluency is another word for automaticity - the ability to use the vocab and grammar you have quickly at any given level, with no translation and little conscious effort. The famous linguist Stephen Krashen calls this “Acquisition” of a language, as opposed to mere theoretical book “Learning”.
This automaticity comes from automatic brain systems that process language: without them, we can’t process speech fast enough to keep up with the speed of a normal conversation. Knowledge of vocab and grammar are not enough. When speaking and listening, these brain systems operate like “Next-Word Prediction” models, similar to AI Language Models or text prediction on our phones, figuring out which word might come next.
You can tell something like this is happening when you speak your native language: try it! Your conscious brain doesn’t choose the precise words you say - that job is taken care of by automatic (predictive-like) processes. “You” just “monitor” this speech, “listening in” as the words appear as if by magic.
In contrast, when these automatic networks aren’t yet in place for a new language, speaking and listening are difficult, conscious processes, as you try to remember words and apply new rules.
You can develop some automaticity at any vocab level, so you could be “micro-fluent” at 500 words or “macro-fluent” at 10,000. And there’s much more to building up these brain systems than just prediction: we also need to handle grammar, learn to parse new sounds and detect other information, including Chinese tones.
How can “Micro-Fluency” help us?
To learn a language, we need a set of goals that are achievable and on the road to fluency (itself an ever-changing destination of course). Language-learning courses and apps - which ought to be honest about how we’re really doing - often let us down here.
We could use language proficiency exams or vocab lists like HSK to chart our progress, but these are arguably not a good way to spend our time.
What if instead, we structured our language learning routines so that we achieve “micro-fluency” at 500 words, then 1000 words, then 2000 words, and so on? That way we know we’re making real and useful progress, not just faking our way through HSK exams.
Some learners might want to speak “micro-fluently” at each stage, others may want to hold off the speaking stage and just want to achieve listening “micro-fluency”: comfortably being able to understand new content graded to that level.
But how would we know we’ve reached “micro-fluency” at any given level? Can we measure it?
Measuring “Micro-Fluency”
Language Acquisition researchers have techniques to measure what they term “processing speed and automaticity”. Some use direct measures of brain activity using electrodes, but many methods don’t require this and could be developed for apps and online tests. They often detect the presence of a “prediction” network by doing something unpredictable.
If you are micro-fluent in basic English and someone read out the following monologue, you would immediately sense there was something “weird” when you got to the third sentence: “Good evening, please sit down here. Welcome to our beautiful restaurant, I hope you like it. Would you like to eat my shoes? Or perhaps try some of this wine?”
If you have automaticity, your predictive brain network will quickly tell you something unpredictable has happened - in this case, a surprising “error” in meaning (a so-called “pragmatic violation”), which would show the brain lit up with surprise and confusion if you had electrodes attached.
How would you do in Chinese? “晚上好，请坐。欢迎光临我们美丽的餐厅，希望您喜欢。您想吃我的鞋子吗？尝尝这酒吗？”
“Micro-fluent” listeners can easily press a button when something seems “off” in this simple narrative, making this technique a form of instant listening comprehension - with no need to recall or guess answers to questions.
If you’re not micro-fluent, it’s harder to catch these “mistakes” as they happen - suggesting you could be working on your listening, building up the brain networks you need for automaticity. It’s much harder to “game” or “get around” these tests than with typical listening assessments like HSK exams.
Of course, you can simply judge your micro-fluency by listening to some new graded material that you think should be at your level. If you struggle to understand it, and looking back on it, you knew all the vocab, you’re probably not “micro-fluent” at that level yet. So what can you do about it? Lots!
How do I improve my “Micro-Fluency”?
“Immersion” is the centre of any successful language learning programme. “Extensive” listening and reading (particularly listening) of a wide variety of interesting content is what trains your predictive systems and exposes other brain systems to grammar, phonology and so on.
“Intensive” listening and reading - which means repeatedly listening to the same material over and over again, has also been an important technique for many successful language learners, training these brain systems in a different way, through repetition.
Conversational practice with a friend, tutor or just some guy on a train is as important for developing your listening as it is for speaking. For speaking, sentence mining is very important, as the editor of this Blog, Mischa Wilmers, has documented.
Why should you aim for “Micro-Fluency”?
Micro-fluency (at any particular level) is just a goal - but it’s an achievable goal with real purpose that you can achieve in months not years. Rather than scare yourself with how distant the prospect of fluency seems, make real use of what you already have. It’s often better to have “micro-fluency” at a few hundred words than a poor command of thousands, as my Mongolian acquaintance had long discovered.
If tests for micro-fluency had determined my language grades at school, I wouldn’t have scored nearly so high. But accurate micro-fluency tests could have helped my teachers focus more on what matters in language learning and set me up better for languages in the long run. And perhaps wider use of tests would show many learners that their preferred app or study method was letting them down too - or showing them they are on the right track.
If you’re studying a language in a conventional way such as Duolingo, and you’re past the 500-word mark, you may still never have listened to graded material like podcasts for pleasure (I recommend TeaTime Chinese) and you should correct that now! Asking the simple question “How can I most enjoy my Chinese or make myself more confident?” is never a bad idea.
And of course, as you become micro-fluent at any particular level, it’s time to explore again. The point of thinking in terms of micro-fluency is not to stay stuck in your ways at a comfortable level, but to enable you to make progress with a spring in your step, setting out for a new horizon that you can soon reach and enjoy.
Arti Languages are developing a new way to immerse in Chinese and improve your automaticity, called Artificial Immersion. You can watch their 2-minute explainer below, and download their initial experimental game for iOS or Android on their website.
*Subscribe to the I’m Learning Mandarin blog below for regular podcasts, blogposts and updates from our community pinged straight to your inbox