English First Foundation Issue Brief

produced by English First Foundation

8001 Forbes Place, Suite 102 Springfield, VA 22151

(permission to reproduce is granted, provided credit is given to English First Foundation)

Evaluations of Current Bilingual Programs

But are the (children) who are enrolled in the (bilingual education) program itself today benefiting from it or not benefiting from it because if they are not benefiting from it, then we are obviously wasting the money? So it is a critical point, it seems to me.<1>

Bilingual education is a term which means different things to different people. Like any other debate among social scientists, it is important to understand the meaning of certain terms. In the interests of absolute fairness, we will consider the terms and definitions of them as they were outlined for the U.S. Commission on Civil Rights by bilingual education advocate Josue M. Gonzalez.<2>

1. Submersion. This approach calls for the placement of LEP (Limited English Proficient) children in classrooms where only English is used. No special attempt is made to help overcome the language problem, and the children's first acquired language (L1), is not used for instruction. For this reason, this approach is often described as the "sink or swim" method.

Many of the Hispanic bilingual education advocates were then, and continue to be, graduates of programs of precisely this nature. Jose A. Cardenas, to take one example of someone we will hear from frequently on the subject, was placed in an all-English first grade in the 1930's.<3> Most readers of this report have ancestors who learned English the same way: in classrooms where English was the only language used for all learning activities.

2. Structured English Immersion. In this approach, instruction is also provided in English, the child's second acquired language (L2). In this respect, this approach is similar to the submersion method, but there are important differences in other respects. In structured English immersion (SEI), it is required that teachers be able to understand the child's first language or L1. The children are permitted to use their language to address the teacher although the latter will always respond in English.

While the details of structured immersion programs have varied over time, the orientation and goals of these programs are the same: to learn another language by speaking, reading and writing that language.

3. English as a Second Language (ESL). ESL students are placed in regular (English-language) classrooms for most of the school day. During part of the day however, they receive extra instruction in English. Generally, this extra help is based on a special curriculum designed to teach English as the second (acquired) language. The child's home language is rarely used in ESL classrooms; when it is used, it is usually for informal, non-instructional uses. Further, an ESL teacher is not required to speak or understand the children's own language.

Since ESL programs do not require teachers who are fluent in foreign languages, they are often much more adaptable to sudden shifts in the linguistic mix of a particular school system's student body. Textbooks are easier to replace than teachers, after all. Because foreign language fluency is not needed to run these programs, they are unpopular with both ethnic culture advocates and foreign language-speaking teachers since neither is assured of employment from programs of this sort.

4. Structured Home Language Immersion. In this approach, children are instructed exclusively in L1 for extended periods of time. The second language (L2) [e.g. English] is not used at all until the children have a mastery of L1 which is commensurate with their age, and extent of formal schooling. It includes skills in reading. . . . Said differently, the child first learns to use one language -- his own -- and having done so, can then learn to use the other.

5. Transitional Bilingual Education. This is the form of bilingual education that developed under the stimulus of federal funds provided mostly by Title VII of ESEA. In it, subject matter is taught in the home language until the student's proficiency in English has been sufficiently developed to allow them to participate successfully in all-English classrooms. In addition to the use of L1 for the teaching of content curriculum, ESL methods and techniques are used to speed the learning of English. Over time, the use of L1 is gradually diminished and English is increased until it becomes the child's only school language.

The reader will note that the two forms of "bilingual" education outlined at first glance differ only in that children are not introduced to a second language (like English) at all in a Structured Home Language Immersion Program, while the second language is only a small part of the school day in a Transitional Bilingual Education Program.

(This almost hair-splitting difference makes it problematic for the outside observer to determine whether language maintenance programs (much beloved by ethnic separatist advocates) are taking place in these programs. During the debate on this question in Congress and in the bureaucracy in the 1970's, there were those who would take advantage of the possibility for confusion to obtain indirectly what Congress might reject explicitly.)

The main difference between "Transitional Bilingual Education" (or TBE) and "Structured Home Language Immersion" is that the word "transitional" is explicitly included in the former term because the native language instruction is intended to be merely temporary. Exactly how temporary this instruction is to be, that is whether Hispanic children, say, should learn their geography, history, and arithmetic in Spanish until third grade, sixth grade, or senior year of high school, has been a matter of much debate. Meanwhile, "Structured Home Language Immersion" programs make no pretense that they are oriented toward integrating non-English speaking children into an English-speaking society.

The reader should also note the frank admission that without the money the taxpayers provide to the federal government, TBE programs would virtually not exist since they were "developed under the stimulus of" federal dollars. It is at this point that we must ask whether the federal government knew what it was buying.

As early as 1970, the then-U.S. Office of Education was preparing draft guidelines on the subject which officially defined bilingual education thusly:

Bilingual education is instruction in two languages and the use of those two languages and mediums of instruction for any part or all of the school curriculum. Study of the history and culture associated with a student's mother tongue is considered an integral part of bilingual education.<4>

By the strict wording of this definition, just about any approach which used the student's home language at some point during the day would qualify as a bilingual program. But a solid ESL or structured immersion program probably would not.

The government's early fixation on bilingual education is particularly startling when we consider that while the federal government was uncertain as to exactly what a program of bilingual education should be exactly (but was already indicating a willingness to require local school districts to provide it), it was equally uncertain whether the principle upon which such programs were based would actually work at all.

Rudolph C. Troike of the Center for Applied Linguistics and National Clearinghouse for Bilingual Education admitted as much in 1978, just four years after the Congress had decided that bilingual education was the only way to help children learn to speak English who did not already do so:

When the Bilingual Education Act (Title VII) was launched in 1968, it was, unlike the moon landing program or federal efforts in public health, undertaken largely as an act of faith, with little research to support it . . . (and) we have very little more of a research base for bilingual programs than we did 10 years ago (emphasis added)<5>

Even today, two full decades after the 1968 beginning of federal funding of bilingual education programs, the search for some proof that these programs actually work to help immigrant children enter the mainstream of English-speaking society continues with little success.

Eight years after Dr. Trioke made his comment, Christine H. Rossell and J. Michael Ross, undertook their own review of the evidence in support of bilingual education. The researchers looked at scientific journals, courtroom testimony and the major studies of the subject to see if transitional bilingual education actually "worked". Their definition of "worked" took advocates of these programs at their word that the goal of such programs was to maximize "the highest English language achievement of which that student is capable."<6>

The first thing Rossell and Ross noticed about the "scientific proof" being waved about in bilingual education's favor was that much of the "proof" was produced by fervent ideologues masquerading as neutral scientists:

(A)s is common with controversial social programs with egalitarian goals,the evaluators and those who review and integrate the research, are also passionate advocates of bilingual education for political or ideological reasons. The disgraceful treatment of linguistic minorities in this country . . . may have influenced many social scientists to believe that any policy that ignores the mother tongue in favor of English is racist, and any policy which maintains the mother tongue, however inadequately, is equitable.<7>

Their opinion of the limitations of much of the research on this subject is confirmed indirectly from a source on the other side of the spectrum. Stanford University law professor Charles R. Lawrence III in his article on "unconscious racism" argues that social science methodology may not be adequate to achieving the results he seeks because of the limitations of its studies:

Social scientists typically argue for or against causal judgments by means of statistical correlations. Physical scientists, in contrast, can usually produce a mechanical model that explains cause and effect. When causal judgments are based upon correlations among observed phenomena, an element of arbitrariness is necessarily introduced by the choice of categories that are correlated and by the selection of samples from which the data is gathered. Moreover, the data or behavior that produce the correlation are subject to rapid change. (emphasis added)<8>

When Rossell and Ross sought more objective data, the first anomaly they were bound to notice was how far the government had gone in the absence of any hard data on American bilingual programs. Even though most schools were required to undertake bilingual education programs by 1975, the first national study of the accomplishments of Title VII bilingual education programs begun in 1968 conducted by the American Institutes for Research (AIR) was not complete until 1977.

The AIR study found that Title VII, that is, bilingual program students were actually doing worse in English language arts on average than their Hispanic counterparts who were not participating in such programs and had only slightly more interest in staying in school.<9>

Rossell and Ross took the time to review every study they could find on the subject from the 1960's until at least 1984 -- 79 studies in all. Only studies which either followed the scientific requirement of random assignment to either a treatment or a control group or used statistical controls to compensate for the failure to do so were then analyzed.<10> Only 35 studies passed this "methodological soundness" test.

Of the 35 sound studies in the field at the time of their research, just eight found that transitional bilingual education (TBE) helped students learn a second language quicker than those in submersion-type programs ("sink or swim" all-English environments for example) and just one found an improvement in math. Fourteen studies found no difference in English achievement between TBE and doing nothing and six (or 39, if the AIR is counted as 38 separate studies) found TBE had a negative impact.

As Rossell and Ross bluntly put it:

Altogether, seventy-one percent (of the studies) show TBE to be no different or worse than the supposedly discredited submersion technique . . . Altogether, ninety-three percent of the studies show it to be no different or worse than the supposedly discredited submersion technique in developing math proficiency.<11>

In addition:

No study has found transitional bilingual education to be superior to structured immersion and the only one to show no difference between the two in second language learning was conducted seventeen years ago in the Philippines.<12>

Even more surprisingly, in light of all the speeches and litigation on the subject:

Although the plaintiffs have been successful in arguing on a "common sense" basis that bilingual teachers are necessary for teaching limited-English-proficient students, the empirical research does not support this.<13>

Rossell and Ross's findings would seem to indicate that those who have argued that bilingual education works do not mean that it works to increase English-speaking ability among non-English speakers:

How then to conclude, as so many have in and out of court, that transitional bilingual education is superior? One technique, used by Zappert and Cruz (Bilingual Education: An Appraisal of Empirical Research), is to simply redefine the word. As they argue:

"No significant difference should not be interpreted as a negative finding for bilingual education. . . When one adds the fact that students in bilingual education classrooms learn two languages, their native language and a second language, one can conclude that a statistically non-significant finding demonstrates the positive advantages of bilingual education."

The main argument made for transitional bilingual education in the court decisions and the regulations, however, is that it produces greater English language achievement and content mastery than doing nothing, not the same achievement . . .

Another technique used in research reviews to make transitional bilingual education appear to be superior is to include superior performance in Spanish language arts as one of the research findings demonstrating its superiority. Zappert and Cruz also do this. Again, while we agree this is important, it is not the goal of government policy nor the stated object of the court decisions. If we examine the findings of the twelve studies reviewed by Zappert and Cruz for their effect on English language achievement, sixty-three percent of the findings show no difference between transitional bilingual achievement and doing nothing.(emphasis added)<14>

When they turned to the debate on bilingual education as conducted in the nation's courtrooms, Rossell and Ross discovered that federal judges in case after case were equally at a loss to find proof of the benefits of bilingual education, even as they were, in some cases, making it compulsory. In Serna v. Portales Municipal Schools (351 F. Supp at 1283) for example, the plaintiffs demanded a bilingual program to reduce the difference between the IQ test scores of Hispanic and Anglo children in the school district. Yet the lowest scores were found at the only school with a bilingual education program.<15> Despite this almost unavoidable evidence, the court ordered that the program be expanded to the entire school system.<16>

(The Serna decision is also noteworthy because native English speakers in these classrooms were required to enjoy the "benefits" of a bilingual education. As the Center for Applied Linguistics put it: "English-dominant Chicanos and Anglo students were required to receive some bilingual (i.e. Spanish) instruction" for a minimum of 30 minutes per day beginning in grade 1.)<17>

While the details of the various key bilingual education cases are not relevant to this study, it is useful to look at some of these cases, as Rossell and Ross did, in the context of the evidence for bilingual education which was presented in court when the judges were making their rulings.

Serna came before the Tenth Circuit Court of Appeals while the Supreme Court was considering Lau v. Nichols in its 1973 term. The Supreme Court chose not to follow the Serna court in attempting to ordain a bilingual society by making two-language instruction mandatory for all American students.

In fact, the Supreme Court chose not to make any kind of bilingual education mandatory for non-English speaking students, as long as the school did something.

The Court's lack of a specific remedy for a school's failure encouraged lower court judges to read their own ideas into the law. If the Court had no definitive ideas, then whatever a judge could justify would probably pass muster. It was for that reason that the Tenth Circuit upheld the lower court ruling in Serna shortly after the Lau decision was handed down:

[T]he Portales school curriculum, which has the effect of discriminating . . . violates the requisites of Title VI . . . We do not believe that under the unique circumstances of this case the trial court's plan is unwarranted.<18>

The judge who ordered bilingual education in the Serna case for every student despite evidence that it wasn't working was, regrettably, not unique. During the 1970's, a number of jurists ordered such programs despite the lack of agreement among the "experts" summoned to hold forth on the subject in their courtrooms.

The plaintiffs who wanted bilingual education in Otero v. Mesa County School District (408 F. Supp. 162 (1975)) lost only after a defense expert pointed out that less than three percent of the Mexican-American students affected had any knowledge of Spanish, even if a large number of them came from homes in which Spanish was spoken.<19> Spanish language instruction for these children would seemingly do more to help them communicate at family reunions than it would for their ability to do their schoolwork.

The Otero court specifically noted:

Listening to these experts causes one to conclude that if psychiatrists' disagreements are to be compared to differences between educators, psychiatrists are almost of a single mind.<20>

Meanwhile, the plaintiffs in Aspira of New York v. Board of Education of the City of New York (394 F. Supp. 1161 (1975)) managed to win a bilingual program for students who scored higher in Spanish than in English on the Language Assessment Battery Test despite the court's finding that "the most vivid point to emerge from all the argumentation was the enormous amount of speculation and uncertainty."<21>

The pro-bilingual "experts" even conceded at times that their testimony has little basis in fact. Courtney Cazden of the Harvard Graduate School of Education, an expert witness in both Rios v. Reed (73 F.R.D. 589 (1977)) and U.S. v. Texas Education Agency (506 F. Supp. 405 (1981)) cases stated that her testimony was based simply on what she termed "common sense" and "court precedent."<22>

One judge broke from this mold when he applied a bit of "common sense" to the bilingual education question in Castaneda v. Pickard (648 F.2d 989 (1981)). In that case, the plaintiffs had argued that the lower achievements of Hispanic students were the result of an overemphasis of English reading and writing in the school district's transitional bilingual program for students in grades K-3. In simple terms, the suit accused the school district of letting non-English speaking children spend too much time learning English.

The court found that no such overemphasis existed as a matter of school policy, but did suggest another reason why Hispanic children in a bilingual program might lag behind their classmates: such students were bound to spend less time learning other subjects if they attended school for the same number of hours as students not in the bilingual program. (The court did not add, though it might have, that "the time children in bilingual education programs spend perfecting their native tongue will also take time away from their other subjects, including English.")<23>

The "common sense" belief of Courtney Cazden and her allies that bilingual education will somehow magically help the non-English child do better in school simply finds little support among those studies which attempt to look at the matter objectively. The Rossell and Ross study was neither the first nor would it likely be the last to find that the bilingual education emperor has no clothes.

Researchers like Rossell and Ross are seldom professional political advocates. It is perhaps for that reason that they seemed so determined to find an innocent explanation for the persistence of bilingual education advocates in the face of this mountain of evidence to the contrary. They seem to assume that pro-bilingual education scholars are merely managing to make methodological errors, and will happily change their views when properly instructed in statistical methods. There is however, a less sanguine explanation: for many advocates of bilingual-bicultural education, it is precisely the gain in Spanish or other non-English language arts that billingual education does achieve that is their goal, explicitly or implicitly.

Consider, for example, a 1971 federal government manual for project applicants and grantees for funding under Title VII of the Elementary and Secondary Schools Act. Now Congress had surely not stated that it wished to see the American education system move toward a bilingual society on the Canadian model. Yet federal funds were to be spent to make a beginning at achieving this precise result:

It must be remembered that the ultimate goal of bilingual education is a student who functions well in two languages on any one occasion.<24>

Canadian-style official bilingualism is one of those ideas that turns up in Congressional testimony before bilingual education advocates realized how counterproductive to their goals such honesty was. Dr. Josue M. Gonzales of Southern Methodist University testified before House General Subcommittee on Education, Committee on Education and Labor in March of 1974 and had this to say:

In Canada in 1969 the legislature of that country passed the Official Languages Act. That statute requires that the government of that country make available all of its services in either English or French, the two official languages of that country, as may be requested by an individual citizen. Perhaps it is a bit too early to think seriously about a comparable bill in this country. But if the present growth of Spanish-speakers continues, it is not unlikely that we will soon have to consider that as a very definite possibility. . . What we are proposing here by way of bilingual/bicultural education will make it possible for us to be better prepared for the complexity of such demands when they do come and to demonstrate that this government is truly a representative democracy which is sensitive to the needs of its variegated populations (emphasis added).<25>

Dr. Rudolph Troike of the Center for Applied Linguistics admitted on at least one occasion:

Others (bilingual/bicultural supporters) see the potential for building greater linguistic resources in this country, or recognizing the advantages of bilingualism and rewarding it by strengthening rather than diminishing it, and of contributing to cross-cultural respect and understanding and the growth of a pluralistic society (emphasis added).<26>

Dr. Jose A. Cardenas, Executive Director of the Intercultural Development Research Association (San Antonio, TX) once outlined his conception of an ideal bilingual program:

Though there is a scattering of bilingual materials in almost every subject at almost every grade level, there is no complete set which includes every subject matter area and extends over several years. . . America's minorities request their ethnicity be represented in art, music, sociology, anthropology, science, reading, arithmetic and all other aspects of the instructional program.<27>

(Dr. Cardenas' notion of an "ethnic" science or a "minority" arithmetic is not only absurd on its face. It is also hardly the road to greater toleration in this country. Adolf Hitler, after all, used to believe there was an "Aryan" physics and a "Jewish" physics, a belief which resulted in America's benefiting from the brains of the greatest German-Jewish physicists.)

Frederico M. Carrillo issued a monograph of his own on this question. Like Dr. Cardenas, his views can be characterized as "bilingual-education-is-a-mere-step-toward-a-bilingual-nation":

Bilingual-bicultural education should not be looked upon as a tool for assimilation, as another form of compensatory education, or as merely a bridge to learning the national language and culture. Likewise, the development of bilingual-bicultural skills should not be considered a process that must be terminated at a specific grade level. Instead, both should be viewed as a continuing social force that must be nourished to sophisticated level (sic) if we are to expect our students to become truly bilingual-bicultural individuals--functional in and appreciative of two cultures and languages . . . Minority groups are demonstrating a desire to have their languages and cultures become integral components of the total school curriculum, not only in grades one to twelve, but also on the college level.<28>

Some Anglo politicians have all but joined the "Make Spanish Equal to English in America" movement. Senator Edward M. Kennedy had this to say on the Senate floor on May 20, 1974 during the debate on what would become the 1974 Bilingual Education Act:

When the United States is the fifth largest Spanish-speaking country in the world and when a near majority of people in this hemisphere speak Spanish, surely our educational system should not be designed so that it destroys the language and culture of children from Spanish-speaking backgrounds.<29>

Some of the folks who were at least implicitly endorsing a bilingual country are not militantly anti-American. Some, like Frank Carrillo, seem merely pro-Hispanic, eager to explain that those who declared his people to be failures in American society were putting the blame on the wrong side:

Part of the present day turmoil we have witnessed on college and high school campuses is directly attributable to the ethnocentric philosophy and curricula of schools that offer only the "assimilate or perish" alternative for minorities and then employ scapegoating rationalizations for the subsequent minority failure and dropouts . . . (E)ducational programs have been committed to making Mexican-Americans into the image of the Anglo-American middle class.<30>

Others have not the excuse of ethnic pride or mistaken science. For these bilingual education advocates, non-English speaking children or children from poorer families who do not speak English well are mere cannon fodder for a greater ideological crusade. It matters not to them whether the program works, so long as traditional American society is weakened:

American schooling may be structured in a way that undercuts the most basic freedoms of democracy. For at the heart of American school ideology is the belief that schooling decisions, like most government decisions, are the proper province of the political majority. . . The Supreme Court has eliminated religious indoctrination in public schools, but . . . the imposition of secular values may constitute as significant an interference with first amendment values as the imposition of religious beliefs. . . Alienation from the dominant value structure may generate disciplinary problems or cause poor performance on standardized tests . . . In a case several years ago . . . the school demanded that they walk and talk and wear their clothes in a way that made their white teachers comfortable . . . In class, they were ignored or ridiculed because, unlike their upper-middle-class white classmates, they did not sound like their teachers when they spoke. . . . While the most obvious form of bias in the standardized tests that permeate our education system is the bias of language, there is often a cultural and ideological bias as well.<31>

Much as honest researchers like Rossell and Ross might wish otherwise, it seems clear that bilingual education advocates are concerned about much more than whether or not their programs help children speak English better than they might otherwise. A program that has been a demonstrable failure at its stated objective must be achieving a result someone wishes it to.

Notes

<1> Inquiry of Rep. Augustus F. Hawkins, Chairman, (House) Subcommittee on Elementary, Secondary, and Vocational Education, Reauthorization of Expiring Federal Elementary and Secondary Education Programs: Bilingual Education, March 24, 1987, vol. 4, p. 27.

<2> Josue M. Gonzalez, Hispanics, Bilingual Education and Desegregation: A Review of Major Issues and Policy Directions, U.S. Commission on Civil Rights, January, 1982, IV-15 to IV-17.

<3> Jose A. Cardenas, "The Role of Native Language Instruction in Bilingual Education", Intercultural Development Research Association Newsletter, January, 1984, page 1.

<4> Bernard J. McFadden, "Bilingual Education and the Law", Journal of Law and Education, vol. 12, no. 1, January, 1983, p. 2.

<5>Rudolph C. Troike, Research Evidence for the Effectiveness of Bilingual Education, (National Clearinghouse for Bilingual Education, 1978), pp. 1-2.

<6> Christine H. Rossell and J. Michael Ross, "The Social Science Evidence on Bilingual Education," Journal of Law and Education, vol. 15, no. 4, (Fall, 1986) [Hereafter cited as Rossell], p. 386.

<7> Ibid.

<8> Charles R. Lawrence III, "The Id, the Ego, and Equal Protection: Reckoning with Unconscious Racism," Stanford Law Review, vol. 39, January, 1987, p. 361, note 204.

<9> Noel Epstein, Language, Ethnicity, and the Schools: Policy Alternatives for Bilingual-Bicultural Education, (Washington, D.C., Institute for Educational Leadership, 1977), p. 10.

<10> Rossell, p. 396.

<11> Ibid. pp. 399-401.

<12> Ibid. p. 405.

<13> Ibid. p. 409.

<14> Ibid. pp. 401-402.

<15> Ibid. p. 389.

<16> Ibid.

<17> Center for Applied Linguistics, Bilingual Education: Current Perspectives-Law, (Arlington, VA, 1977), p. 11.

<18> McFadden, op. cit., p. 10.

<19> Rossell, p. 389.

<20> Cited in Epstein, op. cit., p. 29.

<21> Rossell, p. 390.

<22> Ibid. p. 391.

<23> Ibid. p. 393.

<24> Epstein, p. 20.

<25> Ibid., pp. 35-6.

<26> Ibid., p. 41.

<27> Ibid. pp. 73, 77.

<28> Frederico M. Carrillo, The Development of a Rationals and Model Program to Prepare Teachers for Bilingual-Bicultural Secondary School Programs (San Francisco, R & E Research Associates, 1977), p. 45.

<29> Dr. Susan Gilbert Schneider, Revolution, Reaction, or Reform: The 1974 Bilingual Education Act, (New York, Las Americas Publishing Company, 1976), p. 61.

<30> Carrillo, p. 31.

<31> Stephen Arons and Charles Lawrence III, "The Manipulation of Consciousness: A First Amendment Critique of Schooling", Harvard Civil Rights-Civil Liberties Law Review, vol. 15, no. 2, (Fall, 1980), pp. 324-333.