Monday, February 15, 2010

Myths and misconceptions about learning a second language

An interesting paper about best practices when teaching a second language to children.

One point that struck me regarding English speaking kids who are put into French immersion in Quebec:
English-speaking children in late immersion programs (in which the second language is introduced in grades seven or eight) have been found to perform just as well or better on tests of French language proficiency as children who began their immersion experience in kindergarten or grade one.
Combined with this report, it makes you wonder about the wisdom of forcing immigrant children into immediate Catalan-only immersion:
The first predictor of long-term school success is cognitively complex on-grade-level academic instruction through students’ first language for as long as possible (at least through Grade 5 or 6) and cognitively complex on-grade-level academic instruction through the second language for part of the school day, in each succeeding grade throughout students’ schooling. Here, we define students’ first language as the language in which the child was nursed as an infant.
I'm happy to know that my children will be able to blame our move for any upcoming academic failures.


Jeremy Holland said...

They'll blame you regardless when they're teenagers, although there seems to be less adolescent angst and parental conflict here.

santcugat said...

I noticed that teenagers are a bit closer to their parents here. Perhaps they know that they will be living at home for the foreseeable future.

Rab said...

I'm sure you are familiar with the concept of data mining. You can find any academic paper to support your thesis, and someone else will find some other academic paper to refute it. In social "science", academic research is more often than not coloured by cultural bias and preconceived notions.

These two papers, from 1992 and 1997 (!), do not deal with the idiosyncrasies of the Catalan system, where 1-2-1 support is available to pupils, or with the fact that social usage of Catalan language is not the same as in Quebec or in the US.

The first paper (1992).
I am afraid you partake in "quote selection", which is very easy to do. As the author admits:
"The research does not always show an advantage to children who begin at an older age, but differences in performance are by no means as great as relative amount of classroom exposure would lead one to expect."
Since the research does not show any clear evidence either way, and no confidence interval or any kind of statistical result (T-stat, etc) is provided, it appears that what the researcher found was that some kids who started immersion later performed as well or even better than some kids who started immersion earlier. Causality is tenuous to say the least and the author fails to find or explore alternative explanations. Could it be that it is not the age at which immersion starts but the social context of the family that frames their children’s learning experience? That would seem more intuitive than a non-proven, non-evidenced causality statement.

But the author continues: "Pronunciation is one aspect of language learning where the younger is-better hypothesis may have validity. A number of studies have found that the younger one begins to learn a second language, the more native-like the accent one develops in that language".
And I could add: what better way to achieve a cohesive, fully integrated society where people cannot tell who is a "native" and who is a second-generation immigrant because they both speak like natives? The alternative, where some citizens are "native-speakers" but some other citizens are less fluent seems to me a less desirable social outcome.

The author also states: "Language minority children in American schools need to master English as quickly as possible while at the same time learning subject-matter content. This suggests that in the American context early exposure to English is called for."

Let me rewrite this:
Language minority children in Spanish schools need to master Spanish as quickly as possible while at the same time learning subject-matter content. This suggests that in the Spanish context early exposure to Spanish is called for.
Do we agree with the new statement?

Let's try again:
Language minority children in Catalan schools need to master Catalan as quickly as possible while at the same time learning subject-matter content. This suggests that in the Catalan context early exposure to Catalan is called for.
And what do we think now?

For the second paper (1997): it is not often I read an academic paper making a virtue of non-randomised samples. For 'sample restriction' and 'blocking' one could read 'sample fitting' or 'sample tinkering'?
And the paper focuses exclusively on the learning of English for speakers of other languages living in the US. A completely different sociological situation faced by learners of Catalan living in Catalonia.

Perhaps if instead of looking for pseudo-academic justifications for your own cultural and presumably politically-motivated bias, you embraced the indigenous language and culture of your adopted country, your children would not be handicapped by your own misconceptions?

Recommended reading:

Link 1
Link 2


santcugat said...

Our kids are doing fine and are now totally fluent in all three languages. We are very happy with the school they go to (Catalan is the vehicular language and subjects are taught 40% Catalan, 30% Spanish, 30% English and the teachers are generally native speakers). Unfortunately, this kind of schooling is out of the reach of the vast majority of the population.

The point I was trying to make was that the goal of cultural assimilation is sometimes in conflict with academic excellence. There is no doubt that if assimilation is the overriding goal, early immersion is the way to do it.

However, there are many, many studies about the advantages of mother-tongue education in the early years.

Rab said...

Agreed that the ideal thing is to study in your mother tongue. Nobody would argue with that. And my mum makes the best paella too.

But if someone's mother tongue and the indigenous language of a region/country/nation/state are different, why should the state fund that service?

santcugat said...

As long as immigrants pay as much in taxes as the indigenous people, I'd think they have as much right for state funding as anyone else. (assuming there is enough demand in the local community)

Perhaps I should start complaining about my fiscal deficit when comparing the amount of money our family pays in taxes and the amount we receive back in services (exceeding at least 30% of our gross domestic product).

Jeremy Holland said...

santcugat said...

Ok Ok! I confess!

Rab said...

Sorry, but I don't get it.
Are you saying that tax-paying immigrants are entitled to change government policy if they are sufficient number?
Where does it stop? First educational and language policy, then a separate health system, maybe a separate government and law? It is a ridiculous argument, of the kind that could not be made in any other place in the world. If Catalonia was an independent state in the EU, we would not be having this silly conversation. Like nobody seriously questions Danish/Dutch/French/Italian governments on their educational policy.

I have a better idea: What about immmigrants respecting the culture and language of their adopted country and if they don't like it they can bugger off? It is a much easier life for everybody concerned.