One has the impression that Haver feels that initial fluency is not the result of comprehensible input, but is the result of using patterns from the native language, and that correctness is the result of the application of consciously learned grammar rules (see e.g. pp. 36-37): "New English learners cannot rely on their intuition to speak English correctly because they will inadvertently use the patterns of their native language unless taught otherwise" (p. 38). Elsewhere she maintains that students who speak languages in which verb tense is not marked commonly "ignore the entire concept of tense and ... always use the present-tense verb forms if not taught otherwise" (p. 63).

With these statements, Haver ignores the last thirty years of research, research repeatedly confirming that second languages are acquired naturally in the same way first language acquisition occurs, research showing the profound limitations of conscious knowledge in language performance, arguments denying that consciously "learned" language ever becomes subconsciously "acquired" language, research showing the efficacy of comprehensible-input based methods, and data showing that much of first language "interference" gradually disappears as language acquisition proceeds. (Much of this is contained in Krashen and Terrell (1983), which Haver cites. Krashen (2003) updates some of this research. ) Haver is free to disagree with this mountain of research but she is not free to ignore it.

At other times, she inexplicably accepts the reality of language acquisition in the second language acquirer: "Many prepositions can be learned only through frequent use because there is often no logical explanation for why one preposition is used rather than another" (p. 56).

Haver on Reading

Haver enthusiastically endorses the recommendations of Reid Lyon on reading, recommending "systematic phonics programs that have proven invaluable in teaching all children" (p. 42). Haver is apparently unaware of important critiques of Lyon's position. Lyon's view is based on the National Reading Panel's report (National Reading Panel, 2000). Garan (2002) has pointed out that the National Reading Panel (NRP) itself provided only weak support for systematic phonics: Systematic phonics was shown to be more effective than less intensive approaches to phonics only on tests in which children read lists of words in isolation. On tests of reading comprehension given after grade 1, the effect was extremely weak. In addition, Coles (2003) has pointed out that many of the studies "demonstrating" the effect of systematic phonics compare heavy phonics instruction to doing nothing at all, treatments in which children have no exposure to any written language. (It is important to point out that critics of the NRP do not exclude the teaching of phonics; rather, they argue against the over-teaching and over-emphasis of phonics.)

As noted above, Haver recommends a wide array of activities and does not indicate whether they are supported by any evidence. Some are (e.g. TPR) and many (e.g. sentence-combining, dictation, error correction) are not. Haver neglects to mention one activity that has overwhelming support in the research: reading itself. There is massive evidence that self-selected recreational reading has a powerful impact on language development for students at the low intermediate level or higher in first and second language situations (e.g. Krashen, 1993, 2001).

THE ROLE OF THE FIRST LANGUAGE

Haver and the Principles of Bilingual Education

An active critic of bilingual education, and a strong supporter of Arizona's Proposition 203, which dismantled bilingual education in Arizona, Haver appears to at least partially recognize and partially accept both of the classic arguments for bilingual education: Those with a better education in their primary language will have an easier time acquiring English because (1) the background knowledge they obtained in their primary language makes the English they hear and read more comprehensible, and (2) developing literacy in the primary language is a short-cut to English literacy. In support of (1) she notes that in helping students deal with complex problem-solving in mathematics, those who "fortunately" have learned basic concepts in their native language will have less difficulty, and "for those who just cannot understand, it may be necessary to find someone ... to explain the concept(s) in the native language" (p. 50). In support of (2) she states that "some students who are literate in their first language develop literacy quickly and easily" (p. 42). Haver qualifies (2), however, by noting that in some cases "decoding skills" do not transfer adequately. No details are provided for this statement, which is counter to research in this area (e.g. Krashen, 1996, 2002).

3