A six-year-old who speaks Korean at home and English at school is asked to add 13 and 5. She says the numbers to herself quietly in Korean — sip-sam, thirteen, literally ten-three — and the structure of the word tells her there is a ten and a three sitting inside it. Her English-monolingual classmate hears "thirteen", which sounds like a single lump with no ten visible in it at all. Both children get to the right answer. One had a head start she did not know she was using.
This is one of the more concrete ways that language and math intersect, and bilingual parents run into several others. Which language should the times tables live in? Is it a problem when a child starts a problem in one language and finishes in another? And why does a child who can compute perfectly well suddenly stall when the same arithmetic is wrapped in a paragraph?
Number words are not all built the same
English number words between ten and twenty are irregular. "Eleven" and "twelve" give no clue to their structure. "Thirteen" through "nineteen" reverse the order — the ones digit is spoken first, then the ten. French adds another wrinkle in the seventies through nineties: soixante-dix is sixty-ten, quatre-vingt-dix-sept is four-twenties-ten-seven. Danish is stranger still. German reverses two-digit numbers throughout: einundzwanzig, one-and-twenty, for 21.
Mandarin, Cantonese, Korean, Japanese and several other East Asian languages use a fully transparent system after ten. Eleven is ten-one. Twenty-three is two-ten-three. Nothing is memorised beyond the digits one to ten and the word for ten.
| Number | English | Mandarin (literal) | German (literal) |
|---|---|---|---|
| 11 | eleven | ten-one | eleven (elf) |
| 13 | thirteen | ten-three | three-ten |
| 21 | twenty-one | two-ten-one | one-and-twenty |
| 34 | thirty-four | three-ten-four | four-and-thirty |
The observed pattern is that children learning to count in transparent systems tend to reach higher counting ranges earlier and grasp place value with less explicit instruction. It is a modest, early advantage rather than a permanent one — the difference narrows as children get older and the arithmetic gets less about naming numbers. But for a four- to seven-year-old it is real, and it is available to you if one of your languages happens to have it.
Practically: if your home language has transparent number words, use them when you are teaching place value, even if your child's school work is in English. Say the number in both. "Twenty-three. Two tens and a three — èr shí sān." You are not teaching two systems. You are using one to explain the other.
If your home language reverses digits, as German does, be aware that a child writing 34 while hearing vierunddreißig has to hold the four, wait, then write it second. Digit-reversal errors in early written work are extremely common in German-speaking children and are usually a transcription issue, not a math one. Say the number, have the child repeat it, then have her say which digit goes in the tens place before writing anything.
Which language should the math facts live in?
Retrieved arithmetic facts — 7 × 8 = 56, the ones that come back without calculation — tend to get stored with the sounds of the language they were drilled in. Adults who learned their tables in one language and later moved to another often report doing a quick internal switch to recite them. That is not a defect; it is just how rote verbal material is stored.
The consequence for parents is this: whichever language you drill in is the one the facts will come back in fastest, and that switch costs a small amount of time under pressure. So the question is where the child will need speed.
- If school and testing are in the community language, drill the facts in that language. A one-second retrieval delay is invisible at home and expensive in a timed test.
- If the child is young and the home language is much stronger, build the concept in the home language first — what multiplication means, why 4 × 6 and 6 × 4 match — then move the drilling to the school language once the idea is solid.
- Drilling in both is fine but slower. Expect roughly double the practice to reach the same fluency in each. Some families do it deliberately so the child can do mental arithmetic with grandparents. That is a reasonable trade to make on purpose, and a frustrating one to stumble into.
One thing worth avoiding is switching the drilling language halfway through a set. A child who has half the six-times table in Tagalog and half in English has two partial sets rather than one whole one, and the gap shows up as hesitation on exactly the facts that changed hands.
Code-switching mid-problem is usually fine
Parents often worry when a child reads a problem in English, mutters the counting in Urdu, and writes the answer in English. It looks like confusion. It is generally not.
Bilingual children routinely pull whichever language handles a sub-task best. Counting and fact retrieval often stay in the earlier or more practised language while reading and explanation happen in the school language. That is efficient. Interrupting it — insisting the whole thing happen in one language — usually slows the child down and adds a layer of self-consciousness on top of the math.
There are two situations where it is worth stepping in. The first is when the child cannot explain her reasoning in the language she will be assessed in. She may understand perfectly and still lose marks for a method she can only describe at home. Practise the explanation separately from the calculation: let her solve it however she likes, then ask her to walk you through it in the school language.
The second is vocabulary. Math terms are not always transparent cognates, and false friends do real damage. A child who knows billón in Spanish may be surprised by English "billion" — in older Spanish usage billón is a million million. Decimal points and commas swap roles across much of Europe and Latin America, so 1.500 can mean fifteen hundred or one and a half depending on where it was written. Terms like "borrow", "carry", "take away", "quotient" and "factor" need to be explicitly taught in both languages rather than assumed.
Word problems are a reading test wearing a math costume
This is the single biggest source of misleading results for bilingual children. A child computes 45 − 28 correctly every time on a worksheet, then gets the same subtraction wrong when it arrives as: "Priya had 45 stickers. She gave some to her brother and had 28 left. How many did she give away?"
The math has not changed. What changed is that she now has to parse tense, pronoun reference, and a phrasing where "gave away" signals subtraction but the unknown sits in the middle rather than at the end. Two or three unfamiliar words in a sentence is enough to break comprehension of the whole thing.
A useful diagnostic takes two minutes. When a word problem goes wrong, ask the child to tell you what is happening in the story — in any language she likes. If she can retell it accurately and still cannot set up the calculation, it is a math gap. If she cannot retell it, it was a reading problem all along and drilling more subtraction will not touch it.
Things that help:
- Translate the problem, don't simplify it. Say the same problem in the home language. If she solves it instantly, you have your answer about where the difficulty sits.
- Build a vocabulary list from actual errors. Words like "each", "altogether", "twice as many", "remaining", "share equally" and "difference" carry mathematical weight and are easy to skim past. Collect the ones that actually tripped her up rather than working from a generic list.
- Watch the cultural furniture. Problems about quarters and dimes, cricket scores, or a school bake sale assume knowledge a child may simply not have. Not knowing that a dime is ten cents makes the problem unsolvable and tells you nothing about her arithmetic.
- Separate the two skills during practice. Some sessions on computation with no words at all. Some on reading problems where she only has to say which operation is needed, without computing anything.
Whatever practice tools you use — a workbook, a school platform, something like Kid Genius World — the useful feature here is seeing which specific items went wrong rather than a score for the set. Six correct out of ten tells you nothing. Knowing that all four errors were on multi-step word problems while the bare computation was flawless tells you exactly what tomorrow's fifteen minutes should be.
What this looks like across a week
A workable pattern for a seven- or eight-year-old in a bilingual household: three short sessions of fact practice in the school language, kept to five or six minutes each; one session of word problems where you read them aloud together and she is allowed to think out loud in whichever language she wants; and ongoing everyday number talk in the home language — prices at the shop, portions at dinner, minutes until bedtime.
That last part carries more weight than it looks like it should. A child who hears numbers used casually in both languages builds vocabulary in both without ever sitting down to a lesson. The formal practice is where accuracy gets built. The kitchen conversation is where the words stop feeling like school words.