Recent theorizing, inspired by work in bilingual education, presents a different view, one in which the first language can accelerate second language acquisition. This happens in two ways: First, and most relevant for this discussion, education in the first language supplies background knowledge, which can help make input in the second language more comprehensible. Second, providing literacy in the first language is a short cut to second language literacy: If we in fact learn to read by understanding what is on the page (a version of the Comprehension Hypothesis; Smith, 1994), it is easier to learn to read in a language one understands. And once one learns to read in any language, it is much easier to learn to read in another; developing reading ability in one language is a short-cut to developing reading ability in any other language, even when the writing systems are different.

Some of the evidence supporting this position comes from evaluations of bilingual programs, programs in which minority language children are provided with subject matter knowledge and literacy in the first language, along with plenty of comprehensible input in the second language. Students in these programs acquire at least as much of the second language as those in non- bilingual programs in which all instruction is in the majority language, and usually acquire more. This position also explains the success of children who arrive as immigrants in another country while of school age, and who do well in school, succeeding academically and acquiring the majority language well. These children, in every case, had a good education in their own country before emigrating. They had "de facto" bilingual education: Subject matter instruction and literacy development in the primary language (Krashen, 1996; 1999a).

SOME SUGGESTIONS

If the hypotheses presented above are correct:

We can modify our goals

The idea of comprehensible input allows us to adopt more reasonable goals for classroom language instruction. All we need do in language pedagogy is to help students reach the intermediate level, defined as the level at which students have acquired enough of the language to continue to acquire on their own. In other words, they can then obtain comprehensible input outside the language classroom. This means that the goal of the language class is not to produce error-free performance and complete understanding of all rules of grammar. (Note that a pure "skill-building" view of language acquisition, one that presumes that direct instruction, exercises and drills are the true and only path, predicts that use of the language outside the classroom is a waste of time.)

Steven Sternfeld (personal communication) has pointed out that in a sense it doesn't matter how much of the second language students acquire in a given pedagogical program. What matters is whether they have the motivation and tools to continue to acquire after the course has ended. Thus, students need to know how language is acquired. A knowledge of language acquisition theory also helps justify methodology. The methodology proposed here differs from the "traditional" approach, and some students may need reassurance that the method is based on a coherent theory and that it works. A review of what is known about language acquisition can be done initially in the primary language, and eventually done in more detail as sheltered subject matter. S. Y. Lee (1998) provides evidence that providing adult EFL students with a knowledge of theory has a positive effect, when presented before a sustained silent reading program began.

4