My child answers in the other language

You have five ways to respond when he answers in the other language, and they sit on a line from pushing him toward your language to following him into his. King & Mackey report that the two one-language responses were linked to stronger skills in the language being pushed. This comes from their account of Elizabeth Lanza's research, not from the original study.

What is going on

Lanza studied two Norwegian-American families, and King & Mackey report that she identified five main strategies parents use when a child answers in "the other" language. They range from simply pretending not to understand the child to code-switching yourself. The app tags each one by where it sits.

1. Pretend not to understand (one language). You ask him to say it again: "I don't understand, can you say that again?" The most one-language of the five.
2. Make a guess, in the language you want (one language). He points and says a word in the other language, and you guess aloud in yours: "Do you want the ball?" Usually a yes/no question, so gentler than the first.
3. Repeat back in the other language (bilingual). You say what he meant, in the other language, with no question and no correction. He is not asked to change what he said, he hears it echoed back.
4. Move on (bilingual). He uses either language and you continue the conversation naturally, without steering. No preference shown either way.
5. Switch with him (bilingual). He uses the other language, and you follow and reply in that language too, for a sentence or more. The most fully bilingual of the five.

The mechanism King & Mackey describe is about which language the child has to produce. The first two ask for more of your language. The last three leave him free to stay in his. The categories are the app's own labels for King & Mackey's account, so use them as a map of what each reply asks of him, not as a score for you.

What to do tonight

Start with your goal. King & Mackey report that children's competency levels were higher when parents used more one-language strategies, especially when one language is in a weaker position. So if you are deliberately building the weaker language, try the first two. If you are not pushing either way, any of the last three is consistent with that.

• If pretending not to understand feels harsh, start with the guess. The app's note calls it gentler.
• Moving on is the easiest in the moment, but the app's note says it does not build the weaker language the way the one-language strategies do.
• Pick one response and use it through the evening. This is a suggestion for keeping your own reply steady, not a finding.

Keep it from becoming a fight. King & Mackey's own advice, in their chapter on a child who resists, is that you cannot make a child speak a certain language any more than you can make a strong-willed four-year-old wear a certain outfit. They suggest making the language a live choice the child wants, not a mandate.

One family's report, not a finding: Lucinda, stepmother of Matilda (4), says that when Matilda went through a phase of answering back in English, they kept responding in Portuguese without making a big deal of it, and the phase passed on its own. King & Mackey collect this kind of story as an illustration of what has worked for some families, not as a controlled study.

What is still unknown

No source read for this site explains why a child who understands one language still answers in the other, how long that lasts, or how long it takes before he answers in the language you want. Lanza's finding comes from two families and reaches us through King & Mackey, so we cannot say how it holds for yours. We have not read Lanza's own study.