The Abacus, Mental Maths, and the Chinese Mathematical Tradition

My grandmother could do things with numbers that I have never been able to replicate.

She had worked as a cashier in a Chengdu department store in the 1970s, when the abacus was still standard equipment at every retail counter in China. She could look at a column of figures — prices, quantities, totals — and produce the sum before I had written the first number down. Not approximately. Exactly.

She learned on the suanpan — the Chinese abacus, its two-tiered frame of wooden beads — and somewhere in the course of that learning, the abacus became internal. The physical instrument was eventually unnecessary. The mental model of the beads, the columns, the manipulation of place value through a tactile logic — this had been internalised into a system of mental arithmetic that she used for the rest of her life.

I think about my grandmother when I consider the question of whether abacus learning has a place in contemporary education. Because the question is not really about the physical abacus. It is about what the abacus, at its best, does to the mathematical mind.

A brief history of suanpan in Chinese education

The suanpan has been used in China for at least a thousand years, and its formal incorporation into school curricula dates to the Republican period. In mainland China, abacus instruction was a standard element of primary school mathematics through much of the twentieth century, and while it was deprioritised during certain periods of curriculum reform, it has never entirely disappeared.

In 2013, UNESCO added the Chinese abacus and abacus-based calculation to its list of Intangible Cultural Heritage — a recognition that the tradition carries cultural as well as practical significance. Several provinces in mainland China, including Zhejiang and Jiangsu, have in recent years renewed their emphasis on abacus instruction, framing it as both a heritage practice and a cognitive training tool.

The specific claim made for abacus learning is not simply computational speed. The claim — supported by a growing body of research — is that sustained abacus training produces changes in how the brain represents and processes numbers. Students who have learned abacus calculation appear to develop what researchers call a "mental abacus": an internal spatial-numerical representation that supports rapid and accurate mental arithmetic independently of the physical tool.

What the research actually says

The research on abacus learning and its cognitive effects is more nuanced than the enthusiastic claims sometimes made for it.

Several well-controlled studies, primarily from Japan and China, have found that sustained abacus training (typically two or more years of regular instruction) is associated with improvements in mental arithmetic speed and accuracy, improvements in visuospatial working memory, and in some studies, enhanced performance on numerical tasks that go beyond what direct training would predict.

These findings are genuinely interesting. The visuospatial working memory result in particular suggests that the abacus is doing something more than teaching a calculation procedure — it appears to train a spatial-numerical representational capacity that has broader applications.

The caveats: most of this research has been conducted in contexts where abacus training was embedded in a coherent pedagogical system with trained instructors and regular practice. The results may not transfer to abacus courses that are shorter, less intensive, or taught by instructors without deep expertise. The comparison groups matter enormously — some studies compare abacus-trained students to students with less mathematics instruction overall, which is not a fair test. And as with all educational research, generalising from research populations to individual children requires caution.

What I observe in practice

In my nine years teaching at an international school in Hong Kong, I have had students who have attended abacus classes — typically weekend programs offered by private providers. My observations, which are informal and not systematic:

Students who have maintained abacus training for two or more years and who have reached the stage where they are using mental abacus rather than the physical instrument do show notably rapid mental arithmetic. Some of them have a quality of numerical confidence — a comfort with number manipulation — that I find striking.

Students who have attended abacus classes for one term, or who abandoned the practice, show less consistent benefit. The training effects appear to require sustained practice to consolidate.

Whether this investment of time — typically one to two hours per week, outside of already demanding school schedules — is appropriate depends heavily on the individual child. A child who finds the abacus engaging and the mental challenge rewarding may benefit significantly. A child who resents the additional instruction time, or who is already stretched across multiple extracurricular commitments, may not.

Hong Kong's mixed-curriculum context

Hong Kong sits at an interesting intersection. Local school curricula do not typically include abacus instruction. International curricula, including the IB framework used in my school, do not include it either. It exists, if at all, in the private extracurricular market.

This means that for Hong Kong families considering abacus instruction, it is an addition to an existing curriculum rather than an integrated element. The research benefits are most convincingly demonstrated for integrated, sustained programmes. Piecemeal addition may deliver less.

I think the honest answer to "should my Hong Kong child learn the abacus" is: if they are interested, if the instruction is good, and if the time investment doesn't crowd out other things they need — including free play and rest — then yes, probably. The tradition has real cognitive content. My grandmother is the proof I carry with me.

But I also think the cognitive benefits of sustained abacus training — improved numerical fluency, spatial representation of quantity, confidence with mental manipulation — can be built through other means. The abacus is one path to numerical fluency. Good number sense instruction, sufficient practice with mental arithmetic, mathematical games and puzzles, and the consolidation emphasis I described in my piece on mainland Chinese curricula — these are also paths.

What my grandmother had, more than the abacus, was a mind that had spent years in productive, engaged relationship with numbers. That relationship is the thing worth cultivating, by whatever means suits the child.