When I spoke at the FLASH seminar at UT Dallas in January, Anne van Kleeck asked if it mattered whether clinicians let bilingual children know they were bilingual. I’m not really sure what the answer is, and it’s one of those questions that I continued to think about after I got back to Austin.
I know that previously, I said that at the group level, the language we start in when testing bilinguals doesn’t really seem to matter. And the children we test do know we’re bilingual (at least by the time we switch languages at least). It would be pretty difficult to switch people during a screening that only takes 15-20 to complete in both languages.
On the other hand, in our studies when we do more in-depth testing in two languages we try to do at least only one language in any one visit. And, we complete all the testing in one language before we start in the next language. So, in our Diagnostic Markers of Language Impairment study we started in Spanish with everyone. We started with standardized testing, moved on to NWR, narratives, then non-verbal cognitive testing. So, the non-verbal testing was the middle of the “sandwich.” After that, we switched to English testing. We did this so that we couple provide a context that would signal English or Spanish to the child and we stayed in that language, sometimes reminding the child what the target language was.
I think that at the individual level, when we test this way we want to know the limits of their L1 and their L2. Using one language at a time can be a way to do that. But do the children know we are bilingual? I think they probably do. It’s one reason we started in Spanish– we wanted to give children the opportunity to use their Spanish even with the many pressures (intentional or not) to speak English. Also, we couldn’t pretend not to speak English because it would be a hard sell in an American public school that the adult coming to work with them only spoke Spanish.
So, have I answered the question better? Probably not. I’m not certain we have an answer, we have clinical observations over a large number of children. Yet when looking at individual performance for the purpose of making clinical decisions we do need to make the effort to provide a context in which we can observe language performance in each language.