I, Kenji Hakuta, declare as follows:

1. I am a Professor of Education at Stanford University. I have been in my present position since 1989. Prior to that, I was Professor of Education at the University of California, Santa Cruz, and before that, Assistant and Associate Professor of Psychology at Yale University. My first research study was published in 1973, and I have published numerous articles and books on the topic of language. I specialize in research on the development of LEP children. I conduct research with LEP student populations, and I am directly familiar with a substantial proportion of the scholarly and applied research in this area.

2. In 1994, a planning meeting was convened which included nine experts in language development, cognitive development, bilingual education, immigrant education, minority child development, and educational evaluation to determine whether there was a sufficient knowledge base to inform the development of a research agenda on the education of English language learners. In response to the suggestions resulting from that meeting, the National Research Council established a committee which was charged with developing such an agenda. I was chair of the committee. That committee, through the work of its members, commissioned papers, and hearings, reviewed virtually the entire universe of research bearing on the education of English learners, sometimes known as "limited English proficient" students. The culmination of this project was a book entitled Improving Schools for Language Minority Children - a Research Agenda, published in 1997 by the National Research Council. The National Research Council is an independent organization chartered by Congress to provide advice to the federal government on scientific and technical matters. Its committees are appointed by the governing board to ensure balance of expertise, and its findings and recommendations are extensively reviewed internally and through a well-established scientific peer review system governed by a Report Review Committee consisting of members of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine.

3. In California, as in other states, students are expected to be exposed to and acquire knowledge about a range of subjects. This is true at all grade levels. The adopted course of study "beginning in grade 1 and continuing to grade 6" in California includes mathematics, social studies, science, visual and performing arts, health and other studies that a school board might prescribe. (See, Cal. Education Code § 51210.) At upper grade levels, a range of courses is required and offered. To receive a high school diploma in California requires completion of the following courses, among others: two courses in mathematics, two courses in science, and three courses in social studies, including U.S. history, world history, American government, civics and economics. (Cal. Education Code § 51225.3.)

4. I am familiar with Proposition 227, and have been asked by counsel to express my expert opinion concerning its efficacy in providing curricular access to school children. I will use the term "curricular access" in this Declaration to refer to access to the subjects referenced in Paragraph 3, above.

5. Proposition 227 prescribes a one-year course defined alternatively as "sheltered English immersion" or "structured English immersion." These are terms which, as written, have no meaning in the educational literature. There are, however, sheltered English approaches to working with limited English proficient children, as well as some literature concerning structured immersion. It is assumed that the authors may have borrowed the terms used in the Proposition from these sources.

6. The central goal of each of these approaches is to teach children English as a second language. While either English or the child's second language is the principal method of instructing the child, virtually all of the programs which have been evaluated also use the child's primary language to assist with the instruction of the child. Such use may precede introduction of instruction in English, or may be a planned adjunct to it. It would be my conclusion, based on a review of the research literature, that there are virtually no programs that have as a central thesis the essential withdrawal of all instruction in a child's own language. While programs do exist which use very little of the child's first language, it is my observation that these are either designed to instruct the child solely in English acquisition skills, or are necessitated by logistical causes that have little to do with academic merit.

7. While "sheltered instruction" or "structured immersion" approaches often seek to provide some curricular access to students, there is no research which would suggest, much less establish, that either approach can or does provide the same level of curricular access as is provided to English-speaking children instructed through their own language. The literature that does exist is restricted to the early elementary grades and addresses English language development. To the extent that there are research findings concerning these approaches, it is that they may, under optimal circumstances, be as helpful (though not necessarily more helpful) in teaching children English as some of the alternatives that rely upon initial development of a child's first language. There is, however, absolutely no credible research that either of these approaches can provide children with meaningful access to the subject-matter curriculum which is required for graduation, as referenced in Paragraph 3 above. While it may be that under certain circumstances (e.g., with children who have a substantial but incomplete grasp of English, or with children who have a strong literacy base in their own language) that an ideal sheltered approach can be useful, it would be sheer speculation to draw the conclusion that such an approach can provide the same level of knowledge as is made available to English-speaking children who are provided access through their first language. In any event, Proposition 227 applies to all children irrespective of their level of English, age, native literacy, or other relevant factors.

8. A substantial body of basic research in cognitive science supports the view that academic knowledge does not develop efficiently when a student does not understand the language of instruction. Recent research on the ways in which understanding in academic disciplines develops shows that different areas of knowledge, such as history, science and math, have different structures of knowledge, and their development depends deeply on language. The NRC review concluded thus: "The research reviewed here makes clear that language interactions, questioning, expert explanations, discussions of alternative solutions, formulations of reasons for conclusions, contribute to the development of understanding of complex subject matter. Serious practical and ethical questions arise if these optimal methods for content area instruction are inaccessible to second-language speakers, who are thus excluded from participation in the best teaching practices." (Improving Schools for Language Minority Children, at p. 73.)

9. What undergirds this conclusion is the research-based opinion that unless a child receives substantial assistance in a language he understands, he will not attain full grasp of a school's curriculum. Indeed, it is my opinion that the reason that most sheltered classes attempt to provide meaningful degrees of native language assistance is in response to this understanding, and the limits of sheltered programs in providing full access to most children. It is my expert opinion that the approach utilized under Proposition 227 will leave virtually all children with significant educational deficits which will impede their ability to understand curricular offerings.

10. There can be no doubt that, when implemented properly, bilingual education works. For example, in 1980 the California Department of Education launched a project known as "Case Studies in Bilingual Education." The project selected five K-6 schools where there were high concentrations of English learners who were poor and achieving below district norms. The treatment for the project consisted of the following: intensive native language development, which not only taught students to read in their native language but also cultivated thinking skills in this language. Students were not transitioned into mainstream, all-English classrooms until they had developed cognitive academic language proficiency in English.

Unfortunately, the project goals were not all realized as funding was cut two years prematurely. Nevertheless, the program treatment did demonstrate measurable success. By 1986-87, after five years of the project, the median scores for the 3,500 children in these five schools had significantly surpassed district norms in English, reading, writing, and mathematics. At least 60% of students in grades 2 through 6 who began the program in kindergarten were performing at or above California norms in math. In 1984-85, 52% of 2nd graders at these schools had achieved English proficiency, as had 72% of 3rd graders and 91% of 4th graders. The Case Studies treatment represents a successful model of bilingual education implementation in which there was intensive staff development and careful monitoring of classroom practices, and teachers trained to teach language minority students.

To the extent bilingual programs have been under-effective in certain instances, I am of the opinion that any ineffectiveness cannot be attributed to a lack of soundness in the theory of providing English learners with liberal amounts of native language instruction. These implementation problems appear related instead to a shortage of trained teachers, as well as to other more diffuse social and institutional problems that plague many districts in low-income socio-economic areas. For districts trying to improve their educational programs for English learners, the NRC report concluded that school districts should focus on developing and implementing various components that research on school and classroom effectiveness have found to be successful.

11. It is certainly true that there are instances in which a bilingual education program failed to provide an adequate education to some of its pupils. Anyone who argues that merely providing a child with understandable instruction will achieve success is blind to both the research and reality. Understandable instruction is a necessary precondition to education access but alone is insufficient to achieve success.

The findings from the body of research on "effective schools" relied on by the NRC report has identified the following components among others as those most likely to engender effective schools: a supportive school-wide climate, school leadership, some use of native language and culture in the curriculum, a balanced curriculum that incorporates both native basic and higher order skills, opportunities for student-directed activities, extensive staff development, systematic student assessment, staff development, and parent involvement.

12. It might be asked whether the lack of equal curricular access which I have described is balanced by an increased opportunity to learn English in a classroom of the sort described in Proposition 227. In fact it is far from clear that an all-English approach offers any greater opportunity in learning English than a properly-implemented program that utilizes the child's native language. The NRC report suggests most of the reliable evidence indicates that long-term English fluency is attained by first teaching a child to read in a language he or she speaks fluently. The primary evidence underlying this comes from an earlier NRC panel of statisticians chaired by Professor Stephen Fienberg of Carnegie Mellon University, which conducted two large-scale studies to look at the relative effectiveness of bilingual and English immersion strategies. Specifically, this panel found that "the immersion strategy students tend to score eight to nine points lower than early-exit students on the reading subtest, and the difference is statistically significant at the .05 level" (pp. 77-78). The early-exit programs mentioned are those that use the students' primary languages, while the immersion programs are those using only English. The NRC report also found that teaching LEP students to read first in English could result in negative consequences when certain risk factors are present. These risk factors are often present among children who come from homes in which the parents are poorly educated, and in schools with a preponderance of students from low socioeconomic backgrounds.

In sum, while there are a few studies that show a slight advantage to learning English in an English immersion program, the most respected studies show the opposite. Those studies that do show success in using such a method tend to be in populations different from the immigrant population served in California programs. Further, those programs that report success in teaching students a second language all report success only after several years in a program. No evidence exists that such a program can be successful in teaching English in one year.

13. A final point should be made. It is a central thesis of good teaching that each child and each situation require the teacher to be able to adjust a program to fit the needs of the child. Indeed, this is one reason why the English immersion programs that are successful in teaching English utilize bilingual teachers. Proposition 227, by prescribing one sort of program, denies the school and teacher the flexibility needed to shape an appropriate program for children.

On the basis of the foregoing, I conclude that the program envisioned by Proposition 227 will (a) deny children meaningful access to the curriculum and leave such children substantially behind their English-speaking classmates, and (b) fail to provide limited English proficient children the English language skills needed to participate in an English-only classroom.

I declare under penalty of perjury under the laws of the State of California that the foregoing is true and correct.

Executed this day of May, 1998, at _________________________, California.

Dr. Kenji Hakuta

Supplemental Declaration of Kenji Hakuta