Historically, one of the most important ways for immigrant children to achieve economic, social, and political access to the American dream has been through education. Besides offering these students the opportunity to carry their studies to a level not possible in most countries, American public schools have given them what may be the most important opportunity of all — the chance to learn the language of the country. Now this critical avenue to success is being narrowed. The modern immigrant has been betrayed by a confederation of power-seeking politicians, unprincipled educators, and unwitting Americans.
Though no single person or group is totally responsible for this betrayal, much of the culpability can be assigned to the leaders in the bilingual education movement. There are many sincere people involved — parents, teachers, administrators, legislators — who, however noble their intentions, do not understand that bilingual instruction can only retard the learning of English. My criticism is directed at those who know how bilingual programs affect the learning of English and still defend the theory despite its inefficacy in practice.
The myth of bilingual education has been built on ignorance about how immigrants learn a new language. The premise that inspires almost all justification for bilingual programs — that because non-English-speaking students find it difficult at first to understand concepts in the new language, they should receive subject matter instruction in the home language — breaks down upon analysis. Yet this premise is almost universally accepted, by opponents as well as proponents.
The judiciary in particular has been duped by the argument that instruction must be given in the home language so that students do not fall behind in their studies while they learn English. The question of when, where, and how they are to learn English if they are using another language is always left unanswered, if it is ever asked. Judge William Wayne Justice, a federal judge in Texas who has been elevated to the ranks of liberal sainthood because of his several landmark decisions on integration and bilingual education, expressed this credo very simply in a recent decision that ordered bilingual education to be extended to all grades of Texas schools.
Unless they receive instruction in a language they can understand pending a time they are able to make the transition to all English classrooms, hundreds of thousands of Mexican-American children will remain educationally crippled for life, denied the equal opportunity most Americans take for granted. n1
Implicit in his statement is the assumption that somehow these students would learn English as they received instruction in the home language.
n1 “Federal Judge Possibly the Most Hated Man in Texas,” Los Angeles Times, Apr. 12, 1981, Part V, p. 2.
Another star of the federal bench, Judge Paul Egly, who presided over the Los Angeles school desegregation case until he finally resigned in a pique, recently complained about California’s “inhuman” treatment of Hispanic students. Egly blamed the state’s bilingual education law, which he said had caused this type of education to be “ineffective in Los Angeles because of a lack of qualified [bilingual] teachers.” n2 Like his Texas colleague, Egly assumes that immigrant students learn English by means of translations provided by bilingual teachers.
n2 “Desegregation Efforts Called Symbolic,” Los Angeles Times, Mar. 2, 1981, Part I, p. 2.
But this sublime credulity about bilingual education has not been restricted to the judiciary; it has captured the liberal mind of the media. The Los Angeles Times, for example, published an editorial last year that was a masterpiece of confused and contradictory thinking. n3 Though the Times conceded that “children can be taught English by hearing English,” it concluded that because they “are often immersed at home and in the street in Spanish only, the bilingual approach often works better.” That is, in order for children to learn English, it is better for them to hear less English than to hear more English.
n3 “Language: Politics and Survival,” Los Angeles Times, Feb. 5, 1981, Part II, p. 6.
Critics of bilingual instruction have done no better than their liberal opponents in demonstrating their comprehension of the issue. They often equate bilingual education with bilingualism, for example, and from this supposed equivalency stems their contention that bilingual education is bad because it promotes bilingualism, which they believe to be socially divisive. Social division may indeed be a consequence of bilingual instruction, but not because such education contributes to bilingualism — because it does not. Instead, bilingual programs often retard English language learning and thus inhibit the communication that is so necessary to social understanding and harmony.
Bilingualism is not socially undesirable. Rather, it can be of positive value not only to the individual but also to society. If immigrants (a cover term in this article for legal immigrants, illegal aliens, and refugees) learn to function well in English, they will add English to their first language and by definition become bilingual. Unless we believe that the ability to communicate in English creates social division rather than unity, immigrant bilingualism will benefit our society. The key point of dispute, then, has to do with whether students can learn English more efficiently by instruction in two languages or in special English programs.
The allegation that bilingualism is socially divisive is the weakest argument against bilingual education, yet this was the ineffectual stance of California’s former superintendent of public instruction, Max Rafferty, who not long ago called it the hoax of the eighties. n4 He avowed that immigrant children should learn English so that the country would not be “balkanized” but then demonstrated his lack of understanding of how they were to learn the language by attacking the phrase “English as a second language,” a term used for at least 15 years to signify a variety of education programs and methods that have one thing in common: They teach English by using special instruction in English.
n4 Max Rafferty, “Bilingual Education: Hoax of the ’80s,” American Legion Magazine, Mar. 1981, p. 15.
Ironically, critics of bilingual education miss the very point that would invalidate the case for bilingual education — that English cannot be learned through instruction in Spanish, Chinese, or any other language. For several reasons bilingual instruction retards the learning of English. First, the amount of time that the learner has to interact with speakers of the new language is critical; in bilingual programs that time is considerably less than in other language programs. More important, because of the heavy emphasis on translation in bilingual programs, many students become inhibited about responding in English when they know the teacher understands their native language. Only a few highly motivated, gifted students can surmount this obstacle. Most immigrant students in bilingual programs become habituated to responding in their native language and find it difficult to reach the point at which they no longer translate into English but instead think in English.
If the bilingual movement has an Achilles heel, it is its record of being unable to produce significant research and objective evaluations in support of the bilingual method. It would seem that in the ten-plus years these programs have been operating, some conclusive evidence could answer the question, are bilingual programs doing a better job of teaching English than the programs they replaced? But researchers have rarely tried to answer that question. Instead, they have studied and evaluated questions related only remotely to language learning, and they have shown little objectivity in doing so.
Nary a Shred
In 1978 Rudolph Troike, then director of the federally funded Center for Applied Linguistics and a bilingual enthusiast, unintentionally focused on the failure of bilingual proponents to offer empirical justification when he pointed out that the Bilingual Education Act of 1968 (Title VII) was passed by Congress “largely as an article of faith, with little research to support it.” n5 In a monograph written for the National Clearinghouse for Bilingual Education, he admitted, “We have very little more of a research base for bilingual programs than we did 10 years ago. . . . Unfortunately, although evaluations should be a prime source of data on program results, the vast majority of them are worthless for this purpose.”
n5 Rudolph C. Troike, Research Evidence for the Effectiveness of Bilingual Education (Rosslyn, Va.: National Clearinghouse for Bilingual Education, 1978), p. 1.
As an example, Dr. Troike cited a survey done by the Center for Applied Linguistics in developing a master plan for the San Francisco schools. The survey found that only 7 of 150 evaluations met even “minimal criteria for acceptability and contained usable information.” n6 Another survey, done by the Northwest Regional Educational Laboratory, resulted in the rejection of all but 3 of 108 evaluations and 12 of 76 research studies. The wonder of all this is that in ten years the federal government alone had, by Dr. Troike’s own admission, appropriated more than half a billion dollars (worth close to a billion now) for bilingual education without a shred of evidence that this instruction had helped anyone except the bureaucrats themselves.
n6 Ibid., p. 3.
In 1975, however, an evaluation was published that would soon be used by partisans as empirical evidence that bilingual education worked. Although Andrew Cohen, a doctoral candidate at Stanford University, was cautious in generalizing about the results of his study of the Redwood City (California) Bilingual Project, he was not reluctant to state, “The preliminary results of this study are highly supportive of bilingual/bicultural schooling,” and on the strength of these admittedly preliminary results he recommended that “bilingual programs should be implemented and continued elsewhere.” n7
n7 Andrew Cohen, A Sociolinguistic Approach to Bilingual Education, (Rowley, Mass.: Newbury House Publishers, 1975), p. 265.
Mr. Cohen’s “A Sociolinguistic Approach to Bilingual Education” has been one of the studies most frequently cited in support of bilingual education. The study covered the 1970-71 and 1971-72 school years and involved four groups of students, two experimental and two control, in two schools in Redwood City. In all, ninety Mexican-Americans (whom Mr. Cohen fails to distinguish from Mexican students, who comprised approximately a third of the students in the experimental groups and almost half in the control groups) took part in the experiment. The experimental groups received bilingual instruction; the control groups received instruction only in English.
The questions Mr. Cohen asked centered on two comparisons. First, after several years of bilingual instruction, were the children in the experimental groups as proficient in English language skills as the control-group children, who had been taught only in English? Second, were the experimental-group students more proficient in Spanish? Yes, on both counts, he concluded. But his conclusions are questionable if one examines the data carefully.
His data showed that for most measures of English language skills the control groups, not the experimental ones, were superior. For example, although he minimized the differences, the data demonstrated that the control groups were clearly superior in the critically important skill of reading. For other language skills the results were more mixed but still favored the control groups. The really surprising results, though, indicated that the students who had no Spanish instruction had done as well as or better than the experimental students in several measures of Spanish language, including reading.
It was not until some years after the Cohen study was published that questions were raised about its value to the cause of bilingual education. Although Rudolph Troike noticed the contradictory results in reading, n8 he still refused to find fault with the bilingual method and instead reported that improper implementation of the method, not the method itself, was responsible for the disappointing results.
n8 Troike, op. cit., pp. 11-12.
Another authority on language teaching, Nicholas Hawkes of the University of York, England, finally penetrated the rhetorical fog that obscured the Redwood City study. In reviewing Mr. Cohen’s report of the study in the International Review of Education in 1979, Hawkes gently posed a devastating question:
Later results from grades 3-5 . . . showed that by 1974 the English-only group were still doing remarkably well in Spanish compared with the [bilingual education] children, yet were well ahead of them in English reading. One is therefore prompted to ask: if Spanish-speaking children who are taught no Spanish can compare so favourably in Spanish with bilingually taught children, while drawing ahead of them in English, is the Spanish medium element in schooling really essential to the realization of the linguistic aims of [bilingual education]? n9
Professor Hawkes had to know that without what he calls the Spanish medium element, bilingual education would no longer be bilingual.
n9 Nicholas Hawkes, review of Andrew Cohen, A Sociolinguistic Approach to Bilingual Education in International Review of Education XXIV (1978), pp. 424-426.
The Cohen study typifies most of the research and evaluation of the last decade: The enthusiastic conclusions seldom follow logically from the data, and few people notice. What is important to the researchers is that bilingual programs grow by constant reinforcement of the myth that such instruction is superior to monolingual, English instruction because it helps maintain the home language without retarding English language development. The question of special-emphasis English programs is carefully avoided, thus establishing in the public mind that there is no alternative to bilingual education.
Despite its inadequacies, the Cohen study has had far more political influence than almost any other research work in the field. An example of the impulse it has given to bilingual education in California can be found in a publication of the state’s department of education. Bilingual Program, Policy, and Assessment Issues, issued in 1981, consists of scholarly papers written partly to counteract two highly publicized reports critical of bilingual education — the federally funded American Institutes for Research (AIR) study of Title VII bilingual programs, and Noel Epstein’s monograph on alternatives to bilingual education, which in the late 1970s had shaken public confidence in the bilingual concept. The California report was an attempt to repair the damaged image of the concept and to assert the state’s leadership.
Six nationally distinguished educators were commissioned by the state to lead panels of experts, who then formulated recommendations and prepared written reports for the publication. Chosen to write the most important section, that dealing with non- and limited-English-speaking students, were Heidi Dulay and Marina Burt, researchers with a national reputation for their study of second-language acquisition. n10 Their treatment of the topic in the report explored a number of theoretical and practical facets of bilingual education and concluded with a defense of its effectiveness.
n10 Heidi Dulay and Marina Burt, “Aspects of Bilingual Education for LES/NES Students” in Bilingual Program, Policy, and Assessment Issues, pp. 8-38.
Misses Dulay and Burt roundly criticized the AIR report and the Epstein monograph for methodological shortcomings. They produced a list of nine studies and three demonstration projects that met their supposedly rigid standards and broke these sources down into ten categories. The Cohen study appears in seven of the ten. Further scrutiny shows that although the authors considered nine studies acceptable, only six appear in the categorical listings, and of these, two were conducted in the Philippines and Mexico, where linguistic and sociological situations are so different that valid comparisons with U.S. language programs cannot be made. And so the department of education of California, the state with the nation’s largest population of immigrant children, champions bilingual education on the basis of foreign studies that are not relevant and studies whose results directly contradict the conclusions of the researcher.
If little valid evidence substantiates the bilingual lobby’s claims, then it is logical to ask about the alternatives — programs to help immigrant students learn English rapidly without exhausting the resources of school districts. The most pervasive fiction of the bilingual mythology is that the only alternative to bilingual classes is the “sink or swim” method, in which immigrant students are placed directly in regular classes taught in English and receive little if any special help in learning to speak the new language.
This country has a long history of educational programs for immigrant students. Los Angeles has had special English programs since 1915, and up to the advent of bilingual education in the 1970s, they enabled many tens of thousands of students who spoke little or no English to be integrated into the schools’ regular classes. The programs were organized differently at the primary and secondary levels, and the instructional methodology varied over the years, but the common goal was to provide special English classes away from the regular classroom until the student could function with English-speaking classmates. The best of these programs acknowledged that students have different capacities for learning language and recognized that most of their ability to learn English would come from interaction with English speakers both in and out of school, not from formal language instruction. For this reason most of the old programs were integrative, not segregative, as are bilingual programs.
There is almost no mention in the literature of such programs, probably because before the bilingual revolution the education of immigrant students was not seen as having political potential. Now it appears that bilingual partisans will attempt to suppress any efforts to compare alternatives with the bilingual method. A case in point: The Los Angeles school district, which claims almost 100,000 students in bilingual programs, has not only failed to publish any study that would give objective evidence of those programs’ effectiveness but also appears to have detoured, hidden, and actually aborted studies that would compare other methods with bilingual instruction.
One study that apparently was aborted was an evaluation of the Title VII Bilingual Schools Project for the 1977-78 school year. The experimental group for the study consisted of students in bilingual programs, and the control group was made up of students who received instruction in English as a second language. Both groups were tested in English language skills, and the data were sent to a private research firm, which had a $29,000 contract to evaluate them. After the testing, however, official mention of the study ceases. No formal report of the results, either written or oral, was ever presented to the Los Angeles board of education, which funded the study.
The significance of this study — and of its disappearance — should not be overlooked. Almost all other studies of bilingual education have used control groups of students in regular classes, in a sink-or-swim situation, without special help in learning English. This study would have compared students receiving bilingual instruction with those placed in special classes designed to help them learn to speak, read, and write English.
In the same school year a speech therapist in the Los Angeles school district received approval from the board of education to compare primary-school students who spoke mostly Spanish and received bilingual instruction with those receiving special English-only instruction. But the study never got off the ground because, as the speech therapist told me, district officials said it would duplicate a study of wider scope planned at the state level. To date, however, the state department of education has not published any studies even remotely similar to that aborted Los Angeles study.
At least three other studies on immigrant English learning in the Los Angeles schools have been buried in district archives over a number of years. The earliest was a statistical analysis of 150 high school seniors (most of them from Spanish-speaking countries) who had come to the United States in the junior high or senior high school years, studied English in a special program and then graduated. Despite their relatively short residence in the United States (three to six years), these students compared favorably with their English-speaking classmates in class rank, grade-point average, occupational choice, and post-secondary educational plans. The students’ performance in the regular educational program was especially notable, with more than 61 percent achieving a 2.0 (C) average or better in high school.
In 1975 Richard Shea investigated student and parental attitudes toward English and Spanish, including their relative preference for English-only instruction, bilingual instruction, and Spanish-only instruction. n11 The study, undertaken in a San Fernando Valley junior high school with mostly Spanish-surnamed students, found that depending on the subject to be taught, 66.8 to 82.9 percent of the 960 student respondents and 58 to 71.8 percent of the 254 parents preferred English-only instruction.
n11 Richard Shea, “San Fernando Junior High School Needs Assessment Survey,” unpub. report (Los Angeles, 1975).
The third buried study dealt with personality variables in learning English as a second language. n12 Data indicated the importance to the learner of in-school exposure to English in addition to structured English lessons.
n12 Robert Rossier, “Extraversion-Introversion as a Significant Variable in the Learning of English as a Second Language,” unpub. dissertation (University of Southern California, 1975).
Results of those studies have never surfaced in the Los Angeles board of education’s reports. Objective evaluation of bilingual education programs in Los Angeles, then, has been carefully avoided, and there has been a pattern of suppression of district, state, and independent research having potential for showing that alternative programs would be superior to bilingual programs. Is what has happened in Los Angeles typical? Remember that Dr. Troike’s 1978 survey of evaluations nationwide described most as “worthless.” During the early years of bilingual education comparisons of bilingual programs with other methods could easily have been carried out; today it would be difficult, since politicians in many states have given bilingual programs a monopoly, prohibiting other special language programs by law or through federal or state funding regulations.
Why has there been such reluctance to evaluate bilingual programs objectively? Why aren’t bilingual evaluations conducted by individuals or organizations with no vested interest in the perpetuation of these programs?
The most intriguing question will probably not be answered because of the powerful political pressures associated with the bilingual education movement: Does bilingual education provide equal educational opportunity to students who speak languages other than English, as required by Lau v. Nichols? That landmark Supreme Court decision has been interpreted as a mandate for bilingual education. The plaintiffs, certain Chinese students in the San Francisco schools, sued the school district because instead of being given special English instruction, as were some other Chinese students, they were placed in the regular program with English-speaking students. The Court decided that the plaintiffs were being deprived of equal educational opportunity in accordance with the 1964 Civil Rights Act. No particular type of instruction was sought by the plaintiffs, nor was any specified by the Court, although both bilingual classes and special all-English instruction were mentioned as possibilities.
Thus Lau v. Nichols did not mandate bilingual education; it permits any special instruction that would provide equal educational opportunity. The Court found that the instruction must be special in that it would enable the students to learn English outside the regular classroom, where they were at a disadvantage. When they had achieved sufficient proficiency in English, they would enter the regular program to enjoy the same opportunities available to native English-speaking students.
It follows that the test of any special language program should be its efficiency in teaching English. The lack of evidence that bilingual education has any positive effect in promoting English language learning leads to the inescapable conclusion that this method does not meet the Lau test.
Why, then, has this type of instruction been so vigorously promoted and so little criticized? The answer is that the bilingual education movement has been a political rather than an educational movement.
The leading spokesmen of the movement understood that the establishment of bilingual programs had enormous political and financial potential. They got away with it because language learning offered the perfect shield from criticism: Americans have been spectacularly unsuccessful at learning second languages, and because of their relative lack of linguistic and cultural sophistication, they have been quick to accept the claims of the bilingual activists. Even if they did not entirely accept them, Americans have been reluctant to criticize for fear of displaying ignorance. And so most of the movement’s propaganda has gone unchallenged. Unfortunately, those who have spoken out lack the professional experience or practical background to argue effectively against bilingual education. Furthermore, those affected by bilingual education, the immigrant students and their parents, have not really been given a chance to express their opinion. It is worth noting, however, that in school districts with adult education programs, immigrant parents attend English classes taught in English in great numbers.
Taking advantage of the absence of effective opposition, bilingual activists pushed through Congress the Bilingual Education Act of 1968 and much enabling state legislation that paved the way to funding, jobs, and influence. The predominantly political character of the movement has not gone unnoticed in recent years, even by some liberals earlier taken in by the visionary talk. In a recent article in New Republic, Abigail Thernstrom observed that “from the very start bilingual education was a bad cause for liberals” because it diminished the educational and economic opportunities of disadvantaged, mostly Hispanic children, and it did so “in programs that segregate the children from their black and white peers.” n13 This charge is not new; Noel Epstein criticized the segregative character of bilingual programs some years ago in a monograph that drew outraged howls of protest from bilingual activists. n14
n13 Abigail Thernstrom, “Bilingual Mis-education” in The New Republic, Apr. 18, 1981, p. 16.
n14 Noel Epstein, Language, Ethnicity, and the Schools: Policy Alternatives for Bilingual-Bicultural Education, The George Washington University Institute for Educational Leadership (Washington, D.C., 1977).
If the programs segregate and reduce educational and economic opportunity, why are they still defended? In her article Abigail Thernstrom laid the body on the doorstep: “The programs provide both employment and political opportunities, as schools are forced to hire Hispanics without regular teaching credentials, and as students are molded into an ethnically conscious constituency.”
Her indictment is understated. Many of those involved with bilingual education — teachers, aides, administrators, school board members, legislators, researchers, textbook authors, publishers — have benefited directly from the establishment of bilingual programs in the public schools.
If the reward was not in new teaching jobs, it came in promotions to administrative and supervisory positions and to new slots for consultants and specialists. Many bilingual programs now represent a district within a district and rival the regular program in size.
Bilingual education has been good business for many people outside the schools as well. There has been a veritable revolution in industries that supply educational products. Texts have been translated and rewritten in bilingual form. Signs, posters, films, tapes, records, and other materials have been reworked. Experts have written new books and set up consulting firms that solicit business from every level of government. The universities, too, have profited: Some states require teachers to take cultural and methodological courses and pass proficiency exams in Spanish or other languages.
The ethnic politician in many cases does not care whether bilingual instruction is a valid method of teaching English. In fact, it may be politically advantageous for him if immigrants continue to rely on their native language instead of perfecting their English, since dependence on the home language tends to isolate the group and make it more manipulable. Immigrants who learn English move closer culturally to the general society and even become assimilated, making political control difficult. Thus, the more militant activists in the movement insist that top priority in bilingual education be given to maintenance of the home language, and they therefore relegate the learning of English to a secondary goal.
Many individuals and groups who actively support bilingual education do not pass the test of objectivity. Two questions should be asked of bilingual education advocates. Will they personally benefit from their advocacy? Can they produce substantial objective evidence that bilingual instruction is superior to other special programs for teaching English? There is nothing wrong with advocating an instruction method even if one benefits personally from it provided its superiority can be shown. But bilingual enthusiasts have been unable to demonstrate that, and yet they clamor for the expansion of bilingual education.
The Lau decision did not specify any particular method as a remedy for students’ language deficiency. It was the Department of Health, Education and Welfare that perverted the sense of the Court’s decision and created a bilingual monopoly with what are known as the Lau remedies, which essentially mandated bilingual education by creating a yardstick for the government to measure compliance. If districts did not comply, they lost federal funds.
The enforcement arm of the Department of Education, the Office of Civil Rights, has helped the activists create a monopoly for bilingual education across the country. A 1980 Los Angeles Times series explained how OCR forced bilingual programs on school districts:
. . . a task force assembled by the Office for Civil Rights and including mainly advocates of bilingual-bicultural education produced a set of guidelines called the “Lau remedies,” which emphasized the bilingual approach.
They required school districts found to be in violation of Lau, and having 20 or more elementary school children from a non-English language group, to provide bilingual-bicultural education for these children in the native language or languages.
The “Lau remedies” . . . became the basis for the Office of Civil Rights’ review of more than 300 school districts, enrolling more than 1 million national-origin minority children. The “remedies” have had a curious history. They were not law, or even official federal regulations, never having been published in the Federal Register. Yet they became the basis for extensive negotiations between the Office for Civil Rights and local school districts. n15
So OCR stacked the deck of the task force that created the Lau remedies, and bilingual activist politicians were able to rig the process of developing remedies that favored their cause.
n15 “Can Bilingual Education Do Its Job?” Los Angeles Times, Sept. 4, 1980, Part I, p. 2.
Furthermore, even though the Lau remedies had no standing in law, they were used to compel school districts to establish bilingual programs where none existed and to expand those already set up. Again, the Los Angeles school district serves as an example. Agitation for bilingual education there dates to the early sixties. In 1969, the first year of federal funding of bilingual programs under the Bilingual Education Act, a controversy arose over the funding of a successful intensive reading program at the Malabar Street School. Critics of the program, including high officials of the federal office of education, cut off funds because it taught English as a second language — not a bilingual approach. Despite intervention by HEW Secretary Robert Finch, who temporarily saved the program by providing funds, it was ultimately eliminated through political action at the district level by bilingual activists who would not tolerate any alternative program.
Meanwhile, bilingual programs were being established in Los Angeles elementary schools. With no opposition, the programs grew rapidly, and by March 1977 the district counted 23,000 students in its programs. Not enough, said Chicano activists mobilized by the intensely political Mexican-American Education Commission. The commission, a quasi-official advisory arm of the Los Angeles board of education, had come into existence after the student walkouts of 1968 and related militant, sometimes violent actions, which had forced the school board to accede to militants’ demands for an ethnic oversight group. One of the prime objectives of the commission, according to a Los Angeles Times article, n16 was the hiring of more Mexican-Americans, who commission spokesmen insisted were needed to staff a vastly expanded bilingual program.
n16 “Intervention by Nixon Aids L.A. School Project,” Los Angeles Times, July 6, 1969, Part B, p. 3.
Leverage for expansion came from OCR, which said the district was in violation of the Civil Rights Act of 1964 because it had “failed to provide what is, in effect, bilingual instruction to limited- and non-English-speaking students” (LES and NES). The 23,000 students already in bilingual programs were not enough; OCR notified the superintendent of schools that the district was not in compliance because fewer than 50 percent of the LES and NES students were receiving “special language assistance as required by law.”
Ironically, OCR’s charge of noncompliance was probably caused by the bumbling efforts of Los Angeles school officials to play the numbers game. In their eagerness to demonstrate the need to expand bilingual programs, officials had claimed larger and larger numbers of LES and NES students. Even before the district set up a language survey and testing program in 1978, officials had reported that there were 100,000 affected students in the district. In the fall of 1977 a district publication, the Spotlight, gave a figure of 120,000 (more than one of every five students) and reported that the number was increasing by about 20,000 a year. n17 By the next year another area newspaper stated that district sources put the number at 142,000. n18
n17 “Bilingual-ESL: Program Boosted by More Funding,” Spotlight, Nov. 4, 1977, p. 1.
n18 “Hispanics in L.A. City Schools: Bilingual Proficiency Targeted,” Valley News, June 24, 1978, p. 1.
A challenge to the wildly escalating counts eventually came from a Los Angeles teachers’ organization, the Professional Educators of Los Angeles. PELA Vice President Betty Cordoba in August 1978 asked the board of education to address her skepticism about the numbers.
Skepticism would occur to anyone who looked up the number of language-deficient students a decade earlier. In 1968-69 the number of students in special language classes in all grades was 12,890. Though students were not then identified as LES or NES, the figures indicate how many students needed special instruction to help them learn English. Attendance in such classes had grown steadily but not spectacularly throughout the sixties.
Five years later, in 1973-74, the number of students identified as LES or NES was 56,036, more than four times the 1968-69 figure. The next year there was a modest increase, to 58,041, but in 1975-76 the number jumped almost 50 percent, to 83,822. In 1976-77 the board was told of another huge increase, to 100,000 students. n19 Thus in two years’ time the number of LES and NES students almost doubled. The board members, lacking experience and knowledge in the field, accepted the figures virtually without question and without spotting one obvious problem: How could the counts be accurate if there was no valid method of identifying affected students? It was not until the 1977-78 school year that a testing program was set up to measure students’ English proficiency.
n19 “Comparison of Non and Limited English Speaking Pupils Identified,” from a report prepared by Bilingual-ESL Services Branch of the LAUSD entitled “Services for Non and Limited English Speaking Pupils,” May 26, 1977, Table A.
Perhaps board members attributed the huge increases to the large influx of immigrants and refugees. Then one would expect higher numbers of non-English-speaking youngsters. But the three-year increase in NES students was from 22,000 to only 29,000, about a 30 percent rise. The LES increase, however, was an astounding 108 percent, from 34,000 to 71,000. These LES figures follow a curious path. From 1973-74 to 1974-75 the increase was only 400 students; in the next year the count shot up by almost 25,000 students; then, in the next year, 16,000 more.
None of Mrs. Cordoba’s questions were answered at the time. Instead, an associate superintendent later responded in a letter that he knew would not be placed in the public record.
At best, certain district officials were confused about how many students required help in learning English. At worst, they deceptively claimed enormous growth in the number of these students, thus promoting expansion of the district’s bilingual education program.
Though it is relatively easy to identify NES students — they speak little or no English — the determination of LES students is completely arbitrary, especially at the line between limited and fluent English speakers, between those who still need help and those who will benefit more in a regular classroom. One of the most frequent criticisms of bilingual education is that students are kept in bilingual programs long after they should have been placed in regular classes. The federally funded American Institutes of Research study, conducted in the 1975-76 school year in 117 schools across the United States, confirmed this criticism, finding that “less than one third of the students enrolled in the Title VII classrooms in grades two through six were of limited English-speaking ability.” n20 In interviews 85 percent of the project directors indicated that students remained in bilingual programs after they were able to function in regular classes; only 5 percent said they transferred their students to English-only classrooms once they had learned English well enough.
n20 “Interim Results: Evaluation of the Impact of ESEA Title VII Spanish/English Bilingual Program,” United States Office of Education/Office of Planning, Budgeting and Evaluation, Apr. 1977, pp. 2-5.
That is but one of two practices that make bilingual education a growth industry. The second is to identify new students as LES even though they can function adequately in regular classes. Both practices require only that the LES designation be broad so that more immigrant students are drawn into the bilingual net and then kept there longer. Since empire building requires numbers, the identification and assessment procedures used in bilingual education are critical.
The proficiency test, or assessment instrument, as it is called in education jargon, must not be too narrow in its determination lest some fish slip through the net. In the 1977-78 school year, the first year of state-mandated assessment of English proficiency, the Los Angeles school district apparently chose an assessment instrument that did not produce the desired number of LES students. Those figures, which came from a proficiency testing program, were junked for figures of the preceding year, when the determination of LES students was left to local school personnel, experienced and inexperienced, with whatever method was at hand and without any districtwide or objective definition of the term. That 1976-77 count was used not because it was more valid but because it was some 12,000 students greater and thus closer to the district’s earlier claims of 100,000 to 142,000 LES and NES students.
Invalid and Unreliable
The next year officials decided not to use the disappointing proficiency test, the San Diego Observation Assessment Instrument, and chose another that they said had higher validity and reliability ratings — the Bilingual Inventory of Natural Language (BINL). The differences between them are not great. Both are pictorial elicitation tests, that is, tests in which pictures are shown and students are asked to describe what they see. Both have low validity ratings, despite the district’s claims. Indeed, schools experienced difficulties with BINL from the start. Stories circulated that some native U.S. students were identified as LES and even NES. In one junior high school, for example, a native-born student who had attended regular classes in the Los Angeles schools since kindergarten and had a B average in regular classes was labeled NES; another NES pupil was foreign-born but had attended school here since the first grade and had a C average in regular classes; two others had lived in the United States only three and four years each but were straight-A students in regular classes. n21
n21 Information from the author, who as Lau coordinator in the school at the time had access to the data.
There were other indications that the validity and reliability of BINL were suspect. In February 1979 a committee of secondary school administrators and bilingual specialists who had met to discuss progress in implementing the district’s Lau remedies wrote in their meeting record: “The question remains of validity and reliability of the BINL.” n22 But the committee members did not feel any urgency in making a change. BINL is still used in Los Angeles as the principal instrument for determining English proficiency, and there have been no appreciable changes in its procedures.
n22 “Summary of Meeting Regarding Implementation of Lau Plan in Secondary Schools,” Junior High Principals Organization-Educational Development Committee, LAUSD, Feb. 23, 1979.
Even more damaging to the validity of BINL was a report published by the Northwest Regional Education Laboratory, which rated BINL near the bottom in a ranking of fourteen oral tests. n23 The evaluation used a three-point scale (good, fair, poor) to rate four criteria: measurement validity, examinee appropriateness, technical excellence, and administrative usability. On only one measure, examinee appropriateness, did BINL receive even a fair rating. On the others, including validity, it received poor ratings.
n23 Oral Language Tests for Bilingual Students: An Evaluation of Language Dominance and Proficiency Instruments (Portland: Northwest Regional Education Lab, 1976), pp. 124-125.
If BINL is such a poor instrument, why was it selected? Why is it still being used in Los Angeles? Perhaps the test was chosen because it produced a large number of LES students to support demands for expanded bilingual education programs.
Fat Cats and Lucre
The motivating force for bilingual education, then, has been to provide opportunity not for students to learn but for bilingual educators and others to build empires that offer lucrative and satisfying power bases. In Los Angeles the transition from special English programs to bilingual education programs produced explosive growth in jobs and budgets at the supervisory and administrative levels.
During the middle sixties the district had two language consultants, paid on the teacher salary schedule, one at the elementary and one at the secondary level. By the 1976-77 school year Los Angeles was paying a director of bilingual services $37,739, two assistant directors $32,899 each, and eleven advisers $20,899 each — a salary slightly higher than that of the district’s highest-paid teacher. Another $288,362 was budgeted for bilingual support, and $204,968 met costs to administer a bilingual program funded by the state. In all, more than $1,200,000 of the district’s total bilingual funds of $15,000,000 was budgeted for administrative costs.
That was five years ago. The position of director has since been elevated to that of assistant superintendent; the salary has been elevated, too, to $50,000. Salaries for administrative and supervisory posts below this level have also increased substantially.
Although the bilingual movement has consistently ridiculed and even suppressed studies of other methods of teaching English to immigrant children, there now exists a large-scale, thorough review of the literature on the effectiveness of bilingual education. n24 Written by two staff members in the U.S. Department of Education, Keith Baker and Adriana de Kanter, it supports key ideas expressed in this paper, especially the belief that school districts should be free to establish alternatives to bilingual education.
n24 Keith A. Baker and Adriana A. de Kanter, Effectiveness of Bilingual Education: A Review of the Literature, report of the Office of Planning, Budget, and Evaluation, Department of Education (Washington, D.C., 1981).
The review is limited to two central questions, whether transitional bilingual education leads to better performance in English, and whether it leads to better performance in nonlanguage subjects. It examines more than 300 documents relating to bilingual education and rejects the majority of the studies and evaluations as not methodologically applicable to their two questions. The conclusion, based on twenty-eight studies that met the authors’ requirements, is that bilingual education has not proven its case: “These findings do not add up to a very impressive case for the effectiveness of transitional bilingual education.” Further, Mr. Baker and Ms. de Kanter found “. . . no firm empirical evidence that [transitional bilingual education] is uniquely effective in raising language-minority students’ performance in English or in non-language subject areas.”
The most significant conclusion of the review follows logically: There is a need to try other methods. The authors recommend that each school district be given the freedom to decide what kind of special program would be “most appropriate for its own unique setting” because the prescription of a single, federal remedy is not realistic when so little is known about teaching English to children. In March 1982 the Department of Education withdrew the guidelines that had imposed bilingual education on more than 500 school districts in the nation. Schools are now permitted to use “any effective approach including total immersion in English,” another recommendation made in the Baker-de Kanter review. This policy change applies only to the federal level, however, and does not affect California and other states that have more stringent bilingual education laws.
Open the Door
Future immigrant education programs should recognize the unique linguistic, social, and educational situation in each school district but should also be built solidly on three principles of language learning.
* English can be mastered only through linguistic interaction in English, not through translation. Although all proficient bilinguals find themselves occasionally translating unfamiliar words or phrases, oral comprehension and speech are seriously impaired if the learner must translate. Accordingly, effective teachers of immigrant students do not have to be bilinguals. What is important is that they understand how language is learned and are good models of English. In fact, bilingual teachers must be disciplined in their use of translation so that their students do not become habituated to translating from one language to another.
* Though almost all immigrant students can profit from initial language instruction, such help provides only a small fraction of what they need to reach proficiency. For the most part, language teaching provides the learner with information, some of it invaluable, about the language. That is not the same as speaking and understanding the language. It is imperative that from the beginning immigrant students not be segregated from their English-speaking classmates for the total school day. The total staff — administrators as well as teachers — must use every opportunity to integrate these English language learners into activities with English-speaking students. Even beginners can join certain classes in which English does not play a large part in the instruction: physical education, art, music, elementary industrial arts, home economics.
* Some students learn more rapidly and efficiently than others. Language learning is an individual matter, not a group effort. Students in special language programs should therefore be integrated into regular classes on an individual rather than a group basis.
The United States has received great benefit over the years from its immigrants. We should not now close the door to opportunity on young people from other countries who enter our schools seeking to learn the language that binds us all in our common pursuit of social and economic well-being. Bilingual education has been given more than an adequate chance to prove itself, and it has failed miserably. It is now time that other methods be tried — methods that put the welfare of the student ahead of that of the adults who run the programs.
Nothing written here is to be construed as necessarily reflecting the views of The Heritage Foundation or as an attempt to aid or hinder the passage of any bill before Congress.
Robert E. Rossier, head of an English-Spanish bilingual family, is a former teacher with twenty-five years’ experience in working with immigrant students.