Chapter 6: Bilingual Education

The Role of First and Second Language Interdependence in Explaining the Outcomes of Bilingual Programs

When you learn to speak as a child, you are using some sort of innate cognitive abilities to understand language, which Noam Chomsky called the General Theory of Language in his revolutionary book from 1957, Syntactic Structures. He proposed that individual grammars of individual languages were based on a cognitive structure used by all languages. For example, all languages (except sign language, of course) use a variety of phonemes, which can be described by universal features, all express various concepts in some sort of semantic structure, words, prefixes, etc., and all have a certain syntactic structure (subjects, objects, verbs, etc.) This is why it is possible for us to learn a strange language by picking up objects to ask for their names and put the names into some sort of comprehensible structure. We still have to learn the specific rules for each language, but the main framework is already in place.

[Linguists] must be concerned with the problem of determining the fundamental underlying properties of successful grammars [of specific languages.] The ultimate outcome of these investigations should be a theory of linguistic structure in which the descriptive devices utilized in particular grammars are presented and studied abstractly, with no specific reference to particular languages.
(Chomsky, p. 11)

Chomsky's book arrived the same year that a symposium on Universals in Linguistic Theory was held at the University of Texas. These books were the "bibles" when I studied linguistics back in the 1960's. I was pleased to see that Chomsky's book has been republished, and that the topic is still of interest. Nowadays, of course, there are bio-neurological theories of language that weren't possible when I studied linguistics, such as the works of Steven Pinker, which I will have to study some day.

So learning a language is learning those specifics about the particular language; learning math is learning math concepts, which are language independent.) We can appreciate a picture drawn by someone no matter what language she was thinking in when she painted the picture; people who speak different languages can plan sports, even on the same team.

Just as a football player has no difficulty learning the rules of basketball or soccer, because they have some general "sports rules" and a pianist can learn to play a violin or flute because he already understands the musical concepts behind playing instruments, then every language that a student learns builds on her cognitive understanding about what the general language rules are that govern all languages. In each case, the person compares and contrasts the sport, or instrument or language with what s/he already knows, to figure out which new parts still have to be learned.

For example, a Spanish speaker discovers very quickly that English has a lot of recognizable vocabulary, and figures out rules so that he can automatically "invent" correct words in English using those rules. He learns by trial and error when these invented words work and when they are incomprehensible, buiilding his cognitive understanding of both languages.

According to Cummins, there are still people who believe in a "Separate underlying proficiency (SUP) model of bilingual proficiency", which apparently believe that you start from scratch when you learn a new language. Cummins is supported by the early theorists in the theory of Common Underlying Proficiency (CUP) model, which assumes that a good part of learning is common, no matter what language you speak, as he shows in this figure:

Figure 6.4 The "Dual Iceberg" Representation of Bilingual Proficiency
(Cummins, p 174)

Cummins reports on numerous studies representing many language combinations that prove this point, that by learning one new language, a student builds a cognitive representation of language learning that aids him in learning new languages. This is why some people can speak enormous numbers of languages - since a large part of each language is common cognitive representation.

The Ramirez Report (1991), the Thomas/Collier Study (1997) and various International Evaluations

Cummins goes into more detail on a four convincing studies, where large numbers of students taught in different types of bilingual programs were studied, both in the U.S. and internationally. All of the studies he describes provide good support to the thesis that the longer students have good training in their L1 as well as the L2, they will become proficient in both, and in some cases more proficient than their monolingual peers, supporting the additive enrichment principle.