Reflections on learning Chinese 2: how I memorised characters

[This is the second post in a series that began here. This post assumes you’ve read the first one, which contains a summary of how written Chinese words are formed.]

In my last post I talked about the methods that Chinese people use to learn Chinese characters: namely, rote learning and regular testing. I mentioned that while this approach is clearly successful for many East Asian children, I did not use that same method myself.

The main reason is personal experience. I grew up attending a weekly hour-long Chinese class, which was done in a very traditional Chinese style with weekly dictation & spelling tests. However, because I only attended once a week, the frequency of rote learning and testing was far less than a local would do. The night before class, I would copy out the brushstrokes of a character and until I could repeat it consistently without looking. Yet over the longer term, I regularly forgot what I had crammed and became discouraged by my complete lack of anything resembling literacy, despite years of these weekly classes.

Firstly, note that this is not a criticism of the methods per se. The main reason for failure was my infrequency in testing myself. Nonetheless, as I restarted learning Chinese as an adult, I knew personally how much work the local method entailed, and how mindless that work was. I was reluctant to use it again without first seeing if there was a better way.

Then I came across this book in my local library, which changed everything.



How not to forget the meaning and writing of Chinese characters‘. Given my past experience, what a compelling subtitle! This book, or the ‘Heisig method’ (its more common name) was my first real introduction into the science of memory – a topic that is even more crucial to me now that I am a teacher.

What Heisig did:

  1. Take the 3000 most frequent Chinese characters.
  2. Group them together in a very carefully sequenced order (the focus of my next post).
  3. Assign a unique and meaningful term to each radical and each character – usually based on one of the actual meanings of each character.
  4. Finally, craft stories which link the meaning of a character with the meaning of each of its constituent radicals/characters.

This post will focus on the 4th aspect of the method – crafting stories to aid memorisation. Here’s a great example of it in the book:


At the start of the book Heisig provides detailed stories and images, such as the above, for each character. As the book progresses, the stories lose the detail until Heisig just presents the constituent radicals/characters (the italicised terms below) and leaves the story-making to you:


That’s the method.

Now for my experience of it: it was transformative. Here’s why:

  1. It made Chinese characters ‘sticky‘ in my memory for the first time in my life. Instead of just seeing characters as a collection of marks and brushstrokes, seeing them as images & stories gave me powerful memory hooks. As a result, my speed at learning and reproducing characters was far quicker than I had ever experienced before.
  2. It made learning Chinese characters enjoyable for the first time in my life. Instead of repeatedly writing marks to tattoo them onto my memory, story-based memorisation opened up a vivid world of imagination and creativity.
  3. It made learning Chinese characters successful for the first time in my life – in large part down to reasons 1 and 2. The characters stuck; I enjoyed the process; and so I was successful in becoming (relatively) literate.

Now, I mention all this because it’s given me a different perspective on various debates on pedagogy that I’ve encountered as a teacher.

Understanding vs. memorisation?

There’s an interesting and understandable backlash to Heisig (most often from people who have never tried it): it does not present a true and deep understanding of the components of Chinese characters. This is indisputably true; without going into too much detail, the majority of Chinese characters have both a historical specified semantic and phonetic component to them: one part gives you a clue as to its meaning, another part gives you a clue as to its pronounciation. In this light, it’s true that Heisig butchers the true understanding of a Chinese character, since he breaks down characters into rarely-etymologically-accurate semantic components.

Yet I found Heisig a brilliant method. More than that – in enabling me to memorize vast quantities of characters, Heisig enabled me to understand and use the true understanding of Chinese characters, which only works when you have a lot of characters already memorised (so you can identify the semantic and phonetic components and make use of them). So my deeper understanding followed and depended upon my initial memorisation. The latter enabled the former, rather than being in conflict with it.

This ties into a common debate on understanding vs. memory, where the former is often presented as the ultimate goal while the latter is neglected – a debate summarized and challenged in this fantastic post from Kris Boulton. (There’s a great example there too of using stories to memorise the quadratic formula).

There’s much more that can be said here, but my main conclusion is just this: understanding isn’t a cure-all method, and it certainly isn’t in conflict with memorising lots of facts; indeed, understanding can often follow fluency and memory. This is particularly true of foundational concepts like alphabets, Chinese characters, times tables – things that have been referred to as knowledge frameworks.

An interesting side note: there are possible parallels with the phonics debate. I can’t ever imagine something like the Heisig method being implemented wholescale in Chinese schools, even if it’s a more efficient method. That’s partly because their rote learning and testing works, but also because it flies against true Chinese language etymology (a big deal for a written language that’s been around for millenia). In a very similar way, phonics seems to be effective but is widely opposed, with many objecting to how it seems to distort and oversimplify the English language.

Efficient memorisation can be fun

That debate between memory/fluency and understanding often sets itself up as a battle between two extremes: the ‘memory-focused’ classroom  where procedures are endlessly drilled via rote learning with little focus on anything else, vs. the ‘understanding-focused’ classroom where fun and creativity are valued, so pupils talk, discuss, ask questions and investigate for themselves.

Both are caricatures, but this reflection set me thinking: when I teach for memory, I do usually resort to some form of rote learning and repetition. Now, as I focused in my last post, rote memorisation works. Yet as Heisig shows, a focus on memory doesn’t intrinsically preclude fun, creativity or imagination. Heisig’s story method is great fun, while being rooted in well-documented memory techniques. Regarding teaching methods, I’m reminded of Bruno Reddy’s TTRockstars and rolling numbers; Kris Boulton’s aforementioned quadratic equation story; the numerous strategies Bodil mentions here. Whereas ‘Learn this formula – I’ll test you on it tomorrow’ is good;,’Let’s rap/sing/imagine this story together, then see who can repeat it flawlessly tomorrow!’ might be even better. Heisig first taught me that.

More generally, this fact is another direct challenge to views that requiring students to remember things is a ‘disastrous’ part of an old-fashioned education. I think one big reason for this is the unspoken assumption that memorisation is dry, dull, and uninspiring. But Heisig (and the many methods I linked to above) blows apart this straw man. Memorising can be really good fun. The favourite part of many of my school’s KS3 classes (from their own mouths) is TTRockstars: learning and testing their times tables, set to cheesy rock music.

… but success is an even better motivator than fun

All that said, the main reason I extol the Heisig method is that it worked far better than anything I’d tried before. Yes, it is fun dreaming up stories of people sticking thumbs into tinned food only to find out that it’s spoilt (that’s my story for the character for ‘rotten’). But I would regularly abandon my mental stories and change them because they weren’t helping the character stick in my memory. Making the story was fun only if it was effective; when it wasn’t, the story quickly became a source of frustration instead. I’m on board with Greg Ashman’s posts on this very topic. The fun and creativity must be subservient to aiding learning, understanding and memorisation – not vice versa. Fun and creativity that doesn’t help learning quickly ceases to be fun as I/pupils realize how little learning has stuck.


Test, test, test.

I’ve written a lot about the benefits of using stories over rote brushstroke repetition. But don’t mistake me: efficient memorisation techniques must still be combined with continual low-stakes testing and revisiting previously learnt characters. My stories were only half of the strategy. Despite the evocative scenes in my head, I wouldn’t have gotten anywhere without daily practice, recall and testing. Even when using the most powerful & fun memorisation techniques, learning must be consistently revisited – i.e. tested – if it is to be memorised.

This is such an important and neglected part of pedagogy. When I first started teaching, the bulk of my focus was upon making pupils ‘get it’ in each lesson, so I worked hard on explanations and assessment for learning. And so I got to the point in my teaching where each lesson, pupils would ‘understand’ and would perform well on tasks. But when it came to assessments, it became clear that the bulk of what had been learnt was now forgotten. The lesson is very similar to my experience with Chinese characters. Even when using the most effective pedagogical techniques within lessons, learning must be consistently revisited – i.e. tested – if it is to be memorised. But memory over time doesn’t seem to be a big focus in pedagogy and ITT, compared to marking, AfL, and differentiation. As Bodil puts it, ‘a lesson is the wrong unit of time… nothing is ever learned in one lesson’. As Joe Kirby puts it, curricula and assessment aren’t designed with memory in mind – but they really, really should be.

In my next post in this series, I’ll look at another aspect of the Heisig method: his ingenious and unique ordering of Chinese characters (his ‘curriculum’, even) designed exactly with memory in mind.



One thought on “Reflections on learning Chinese 2: how I memorised characters

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s