Skip to main content



Learning languages for (and through) life

Most of us have discovered that repetition is the first law of learning — while the first law of forgetting is not using something we learned.

Follow us on Instagram and Tiktok, and join our Telegram channel for the latest updates.
Follow us on TikTok and Instagram, and join our Telegram channel for the latest updates.

Most of us have discovered that repetition is the first law of learning — while the first law of forgetting is not using something we learned.

Exam cramming is a classic example, where for 48 hours, you remember all you crammed; and two weeks later, you remember very little.

When does learning by repetition start? By the time infants are seven or eight months of age, they are able to distinguish words from a continuous stream of sounds. Most likely, the learning process is driven by repetition — in this case, hearing frequently repeated sounds from parents, care-givers and playmates.

Infants probably associate these sounds with concepts: Milk, want, food, no, Mummy, so on. They learn to recognise patterns of sounds and somehow develop a sentence structure that, over time, is grammatically correct.

Something interesting happens when one grows up in a multi-language community — words from different languages are mixed in a conversation (as referenced by Luke Lu in a recent commentary in this newspaper, “Should we change the way we teach languages?”)

In Singapore, most children, whether Chinese, Indian or Malay, understand certain frequently used terms they hear during the course of each day and develop an intuitive concept of synonyms. Heard in the food court: Eat (English) and makan (Malay); while heard on the MRT or in a lift: Can (English), ke yi (Mandarin).

Almost everywhere you walk within Singapore, you will hear multi-language sentences built from English, Mandarin and Malay synonyms. When children arrive at school and university, the learning language shifts from informal frequently-heard sounds and patterns of sounds to a more formal approach based on memorising vocabulary, verb conjugations and syntax.


In the United States, most children hear only English and do not develop the concept of language synonyms.

As an American, I struggled with Spanish in high school. One class hour per day plus self-study was insufficient for me to remember vocabulary and verb conjugations.

At university, I faced German and French. I managed to barely pass, again because I am a poor learner with respect to memorisation.

Consequently in the US, the label of “bright” or “talented” is linked with those who can communicate in multiple languages. What a surprise to find in Singapore that virtually everyone must be bright and talented because they are multilingual.

While working in Russia, I found I could ignore conversations I overheard — or I could try to understand. In a roundabout way, I rediscovered the frequently-heard-sound learning strategy of infants. Here in Singapore, I also began to pick up a few Mandarin or Malay words using the same strategy.

Listening to conversations as sounds is like the eight-month-old infant hearing a continuous stream of sounds and having to discern which sounds were for communication. The context of informal learning, for me, is much more effective than flash cards that literally translate a word between two languages.

But repetitive exposure is limited by the conversation. So how can one increase exposure to frequently heard sounds, whether Mandarin, Malay or Tamil?

Duke-NUS Graduate Medical School Dean K Ranga Krishnan has written in this newspaper about the role Google Search plays in augmenting memory: We use Google Search to quickly find an infrequently used or never-learned concept. I use Google Translate to augment my exposure to frequently used words. For many languages, Google Translate, though not perfect, provides both written and spoken translation.


Realising that synonyms reflect a path to language acquisition, learning enough of another language to survive can be fun. I have two examples:

I like doughnuts. For an English speaker, the Mandarin for doughnut is great fun: Sweet circle (in pinyin, tian tian quan). One morning at Duke-NUS’ Narambi Cafe, as a Mandarin “infant”, I constructed a sentence asking for two doughnuts: “Tian tian quan er.” (Er being the number two.) The nice Chinese lady smiled and laughed and said: “Tian tian quan liang ge.”

Later I asked someone from my team what liang ge means and was told it means “two pieces”. So now I can say: “Tian tian quan liang ge, ke yi ma?”. If they are available, then she replies, “Ke yi.” If not, she might say: “Wu fen zhong.” (Five minutes)

Slowly, I realised that my Mandarin-speaking colleagues could be a great source of morning lessons complete with repetition. We have fun.


My second example illustrates the utility of listening for frequently occurring sounds while not understanding much of the message.

In 1991, while I was working in Moscow at the All Union Center for Clinical and Experimental Cardiology, the Communist Party Congress was taking place. President Mikhael Gorbachev gave a State of the Union message. The next day at the lab, several asked what I thought of his speech. I said it was rather disappointing.

My colleagues responded, how could I say this when my grasp of Russian was that of a child’s? I replied that the Russian verbs gave the real message in Gorbachev’s speech — a large portion of it employed verbs in the past tense (ending in l, la, lo, li. Future tense is preceded by budu, budish, so on).

By listening out for the tenses, I guessed that about 60 per cent of the speech was past tense, 20 per cent present and 20 per cent future. A 60:20 ratio of past to future tense was not encouraging in my view. My colleagues were surprised by my “histogram” or frequency approach to assessing the speech but agreed that my assessment was correct.

To sum up, with Google Search serving as a memory extension for accessing infrequently used concepts, I use Google Translate to increase the frequency at which I can master frequently used terms outside a conversation and, at the same time, develop my personal synonym dictionary.

What a wonderful era to live in where learning can be tailored to our personal needs instead of the average needs of a classroom of students. With this in mind, could we adapt formal language instruction to include this synonym approach augmented by Google Translate?


C Frank Starmer is Associate Dean of Learning Technologies and Professor, Cardiovascular & Metabolic Disorders Programme at Duke-NUS Graduate Medical School. Education reform is one of his passions, and he has co-authored pieces on restoring the joy of learning and team-based learning.

Related topics


Read more of the latest in




Stay in the know. Anytime. Anywhere.

Subscribe to get daily news updates, insights and must reads delivered straight to your inbox.

By clicking subscribe, I agree for my personal data to be used to send me TODAY newsletters, promotional offers and for research and analysis.