A realistic roadmap for a six-year-old to learn — and love — English

Every parent asks the same question: what age is best to start English?

It is almost always the wrong question. The right one is: over the next six years, how many hours will my child spend with English — and how many of those will be hours the child actually understands and does not resent?

The second question is far less exciting. But it has an answer, and the answer can be planned.

Start with the arithmetic

Here is the sum almost no language school will do with you.

Two lessons a week, ninety minutes each, forty weeks a year. That is 120 hours a year. In a class of fifteen, the time the child spends actually speaking is perhaps a few minutes per lesson.

Six years of that, from six to twelve, comes to roughly 700 hours. It sounds like a lot. But Council of Europe guided-learning estimates put a single early CEFR level at something like 180–200 hours, and later levels cost more. Seven hundred good hours land a child somewhere around B1. In practice it is usually less, because most of those hours are spent sitting quietly while someone else answers.

Now compare a different sum. Twenty minutes a day — one cartoon, one picture book, one song — over six years is 730 hours. The same total, with no fees, no traffic, and the child in a state they chose.

If you take one thing from this article, take this: a workable roadmap is not a matter of finding a better school. It is a matter of making English something that happens daily at home. The class is the supplement, not the plan.

Six is not late — and no golden window is closing

The commonest fear is of having missed the critical period. That fear is fed by a misreading of the critical-period literature.

Johnson and Newport’s founding study (1989) looked at immigrants living fully immersed in English, and even there the grammar scores declined gradually with age of arrival rather than falling off a cliff at any particular birthday.

The largest study ever run on the question — Hartshorne, Tenenbaum and Pinker (2018), with data from roughly two-thirds of a million people — concluded that the ability to acquire grammar holds up until around seventeen or eighteen before it drops. Not six. Not ten.

And for children learning a foreign language at school — a few hours a week rather than immersion — the finding runs the other way from folklore. Carmen Muñoz’s BAF project in Barcelona tracked cohorts starting at different ages and found that later starters progressed faster per hour of instruction. Older children have more working memory, better literacy, and the ability to notice a pattern when they meet one.

So: no, six is not late. But do not let anyone sell you the idea that starting at three is a sound investment. At a few hours a week, it very largely is not.

The one thing young children are genuinely better at

There is an exception, and it is real.

Sound. Flege and colleagues (1999) found that degree of foreign accent correlates tightly with age of first exposure, and that the correlation begins to bite much earlier than it does for other areas. A young child’s ear still registers phonetic contrasts the mother tongue does not use — and English uses a good many that other languages do not: final consonants, consonant clusters, the difference between sheep and ship, where the stress falls.

That gives the roadmap a very concrete priority: for the first two years, the ear comes first and print comes second. If you can only do one thing well between six and eight, make it a lot of English reaching the child’s ear, in real voices, every day.

One caveat matters. Patricia Kuhl’s work (2003) showed that infants do not learn new speech sounds from recordings — the learning machinery needs a live human to switch on. By six this has loosened considerably, and out-of-school exposure studies by Kuppens (2010) and Lindgren and Muñoz (2013) show that screen media does real work at this age. But the principle survives in weaker form: a screen with an adult beside it, talking about what is happening, beats a screen alone.

Ages six to seven: ear first

The goal of these two years is not a word count. It is that English becomes a familiar, comfortable, meaningful noise.

What to do:

Watch cartoons with predictable content. Twenty minutes a day. Choose shows where the pictures carry most of the meaning — Peppa Pig, Bluey, Numberblocks. Webb and Rodgers (2009) found children’s television has low vocabulary demands and high repetition, which is exactly what this stage needs. Let them rewatch freely: the third viewing is when the language actually goes in.

Read picture books aloud. Every night. Even when the child does not follow all of it. Point at pictures, ask in the home language if you must, and do not translate line by line.

Sing. Songs load whole phrases into memory in a way exercises cannot, and a six-year-old will sing without embarrassment.

What not to do:

Do not teach grammar. A six-year-old’s working memory, per Gathercole and colleagues (2004), is roughly half an adult’s. Explaining the present simple to a six-year-old is pouring water into a full glass.

Do not force speech. Many children go through a silent period lasting months, in which comprehension climbs and production does not. Pushing during that window produces anxiety, and anxiety closes the shutters.

Do not start with the alphabet and handwriting. Print comes at the next stage, and it comes far faster if the ear got there first.

Ages seven to nine: decoding and the first real book

Now print enters — and this is where the evidence in this article is strongest.

Systematic phonics is one of the few educational interventions confirmed by large-scale meta-analysis. The National Reading Panel report (2000) and Ehri and colleagues’ analysis (2001) both concluded that teaching letter–sound relationships systematically clearly outperforms teaching them incidentally or not at all, with the largest effects in younger and struggling readers.

What to do:

Do phonics properly. Fifteen minutes a day, following a sequenced programme, for about a year. This is the only stretch of the whole roadmap where I would recommend a rigid syllabus.

Move to graded readers the moment decoding works. The rule comes from Nation (2006): comfortable reading requires knowing about 98% of the words on the page — no more than one unknown word in fifty. If a child stops more than twice a page, the book is too hard. The child is not.

Turn English subtitles on. While decoding is being learned, subtitles turn screen time into sound-to-print practice.

What not to do: do not make everything a read-aloud. Reading aloud tests decoding; it does not deliver pleasure. Fuse the two and the child will come to dislike both.

Ages nine to eleven: volume

This stage has one keyword: more.

Elley’s classic work (1991) on book flood programmes in Fiji, Singapore and South Africa showed that simply putting good books into children’s hands in quantity, with time to read them, outperformed conventional instruction — and that the gap widened year on year.

The number worth remembering: a child who reads twenty minutes a day meets roughly 1.8 million words a year. A child who reads one minute a day meets about eight thousand. After four years, those two children no longer live in the same linguistic world.

What to do:

Favour quantity over difficulty. Ten easy books beat one hard one, every time.

Give real choice. This is not indulgence. Under Ryan and Deci’s self-determination theory (2000), autonomy is one of three basic psychological needs feeding intrinsic motivation. A child who picks the book reads substantially more than a child who is handed one.

Permit abandonment. Closing a boring book is not a failure. It is a reading skill.

Permit rereading. Returning to a book whose plot is known is the cheapest way to move vocabulary from recognised to usable.

Eleven and up: speaking, writing, accuracy

Only now do the things that look like studying English start to earn their place. Working memory is sufficient, the ability to think about language as a system has arrived, and the listening-and-reading base is thick enough for grammar to stick to something.

Cummins (1979) drew the distinction between conversational English and academic English — the second takes years longer to build, even for children living in an English-speaking country. This is the stage that builds it.

What to do: begin explicit grammar, but keep it brief. Begin regular writing, even three sentences a day. Find the child a real reason to speak English to a real person — friends in an online game, a club, a distant relative. Dörnyei (2009) argues that durable motivation comes from a picture of the person one will be once one knows the language, not from marks.

Why the sticker chart ruins it

This is the least welcome section, and the one with the firmest evidence behind it.

The meta-analysis by Deci, Koestner and Ryan (1999) pooled 128 experiments and found a result consistent to the point of discomfort: tangible rewards contingent on task completion reliably undermine intrinsic motivation. The effect is strongest in children, and strongest for activities the child already found interesting.

The mechanism is easy to see. When you pay for something, you say implicitly that it is work. A child who reads for stickers learns that reading is a thing one gets paid for — and when the stickers stop, so does the reading.

What to do instead:

Praise the process, not the outcome. You got through that one quickly beats you’re good at English.

Let English visibly open something the child wants. A game with no translation. A channel about dinosaurs. An origami tutorial. The best reward for understanding English is more English to understand.

Do not turn dinner into an assessment. What words did you learn today? asked for three years teaches a child that English is the thing they get questioned about.

What a week looks like

Here is the minimum viable version for a busy family, with no requirement that the parents speak good English:

Daily, 20–30 minutes: one episode or one book. The child chooses. No quiz afterwards.

Four times a week, 15 minutes: phonics (ages 7–9), reading aloud together (ages 6–7), or silent reading (from nine).

Once or twice a week: a class, if it is affordable — mainly for someone to correct pronunciation and someone to talk to.

Nothing at all at weekends, if that is what keeps this sustainable for six years.

That totals 4–5 hours a week, or 180–220 hours a year, or nearly 1,200 hours by age twelve. Half again as much as the school-only plan, with most of the hours pleasant.

Where the money goes to die

Schools whose main selling point is a native teacher. An untrained native speaker, in a class of twenty, for ninety minutes a week, does not outperform a good local teacher. What a child needs from native voices is volume of listening, and volume of listening is free.

Certificate exams taken too early. Starters, Movers and Flyers are useful as gentle milestones. They become a burden when a whole year gets bent around the exam, and when the results get compared between parents.

Over-gamified apps. Streaks and badges are extrinsic rewards — precisely what Deci and colleagues warn about. They build the habit of opening an app, not the habit of using a language.

Extra classes to keep up. If a child is struggling, it is nearly always because the listening and reading base is too thin. More grammar hours will not fix that.

How to tell it is working

Do not measure in words memorised. Measure by these four signs:

One: the child puts something on in English unprompted. This is the best single indicator, and it arrives before any of the others.

Two: the child guesses rather than asks. Inferring meaning from context is the foundational skill of all later reading.

Three: the child uses a phrase nobody taught them. That means language is arriving through input rather than instruction — which is the right road.

Four: the child is not afraid. If a ten-year-old flushes when asked to say one sentence in English, something in the last six years has gone wrong, however good the marks look.

What is actually being built

Six years is a long stretch, and none of the arithmetic above survives a child who hates the process.

That is why love is not decoration in the title. It is an engineering constraint. Seven hundred hours only accumulate if most of them are hours the child does not resist. Every choice in this roadmap — letting them pick, letting them quit a book, letting them rewatch, not quizzing afterwards — serves one purpose: keeping the number going up.

What you are building over these six years is not vocabulary. Vocabulary comes later, and comes fast, if the base is sound.

What you are building is a twelve-year-old who thinks English is unremarkable — not a subject, not a test, just another way of finding things out. From there, most of the rest runs itself.