Enrique Soberanes knows one particular spot on the floor of the principal's office at Eastman Avenue Elementary School very well. It's a tile circle covering a metal utility plate held in place by four brass screws. He knows the spot because, as an Eastman student during the early '70s, he used to stand in line there, waiting his turn to get paddled by the assistant principal for fighting on the playground, often with fellow Latino kids who'd taunted him because he spoke little English. In seventh grade, he failed most of his classes, and he didn't do much better in the first semester of eighth grade. In the L.A. of those days, none of this had anything to do with bilingual education, because his schools didn't have it.
“I remember feeling like a second-class citizen,” said Soberanes. “I remember feeling inadequate, that I had to be as white as I could to be accepted. It really does a job on your self-esteem.”
Soberanes, 33, got turned around thanks to his mother, some dedicated teachers and his own hard work. He went on to study biochemistry at Occidental College, although for most of that time he still struggled with writing in English. He would later edit science textbooks, before becoming a third-grade teacher. Despite his eventual success, Soberanes doesn't tout the sink-or-swim approach. “It's a selfish cop-out that whatever worked for you should work for someone else,” he said, noting that you never hear from the people who sink. And many of his Latino classmates, he said, did just that. In the 1960s, the Latino dropout rate was estimated at more than 40 percent.
These days, Soberanes is back at Eastman, as the recently hired director of the East L.A. school's bilingual program. His desk is two feet from the tile circle where he once stood waiting to get paddled. But the school is a very different place. Today, Eastman has a state-of-the-art bilingual-education program, one that Soberanes wishes had been available to him. In it, students with limited English-speaking skills receive most of their instruction, typically for three to four years, in Spanish, the primary language for most of the school's students. “I believe in bilingual education,” said Soberanes. “Bilingual education gives kids the basic building blocks for real achievement – not just enough English for survival, but real academic achievement in English.”
For years, programs like Eastman's were widely accepted as the proper way to educate students lacking sufficient English skills. But now, the approach sits on the verge of extinction in California. On June 2, state voters will almost certainly pass Proposition 227, an initiative that would fundamentally change the course of grade-school education in California by replacing bilingual education with one year of intensive training in English.
This profoundly flawed initiative would install a rigid and largely untested approach to educating many of the state's neediest children. But its appearance on the ballot has forced a major reassessment of bilingual education as practiced in the state, and the resulting picture is deeply disturbing. Bilingual education has certainly been well-intended, and in practice – in certain schools – the results have been entirely respectable. But even Eastman, with its model program, can only claim mixed results. By any objective standard, the state's bilingual-education program has failed.
The disaster's roots go all the way back to the program's California beginnings, when well-meaning bureaucrats imposed a promising but unproven formula for bilingual ed, then sought to make school districts follow it lockstep. Although this prescription did some good by mandating the hiring of much-needed bilingual teachers and insisting on overdue cultural sensitivity, it fell short by never requiring success. The only real demand was orthodoxy to the formula. And in the main, the state's efforts never delivered results. Nowhere is that more clear than in L.A. Unified, which is, by default, the nation's premier working laboratory for bilingual education, with more than 312,000 students – 46 percent of the district's enrollment – who aren't fluent in English.
To explain just how bilingual education has worked in California, it's important first to dispel some common misunderstandings.
As applied here, bilingual education is not at all what the term literally means. That is, it's not at all “bilingual education” – which implies a primary goal of making children literate in two languages. Instead, it is a one-way road from a foreign language (usually Spanish) to the destination of English. Yet, paradoxically, bilingual ed is not about teaching children to speak English as quickly as possible. Rather, the goal has been to promote academic achievement while steadily improving students' language skills, however long that naturally takes.
You can see the philosophy at work in the first-grade classroom of Eastman teacher Phyllis Zepeda, an 18-year veteran who obviously knows what she's doing. On a recent Thursday, her class of 18 was divided into four quiet, happy and efficient groups, all with different tasks, such as writing thank-you letters, editing journals and illustrating poems they wrote for Mother's Day cards. Students who finished early quietly strolled to the well-stocked, in-room library to thumb through books or moved to other stations with neatly organized edu-toys.
Zepeda's class is a virtual essay in good teaching, but it's also almost entirely conducted in Spanish, even though most of these children have been at Eastman since kindergarten, for nearly two full years. The day's grammar lesson, for example, is not about the use of pronouns like him or her, but about differences between masculine and feminine words like ninos and ninas. Although Zepeda sneaks in some instructions in English – because she knows the students understand them – the school's program officially allows for only 50 minutes of formal English-language instruction.
At exactly 1:30 p.m., a singular phenomenon occurs. Zepeda sends three-fourths of her students to other classrooms, while at the same time, students from the rooms of two other teachers arrive at her door. Now, Zepeda will speak in English as much as possible, as she takes these children through an art, music or dance lesson. This part of the day is called “mixing,” and the idea is that you mix English and Spanish speakers together to promote the speaking of English while the teacher focuses on matters less strenuous than math or reading. Playground time and lunch also officially count as mixing, though the academic benefit of munching macaroni and playing dodge ball is hard to document.
But then, even the formal class-time mixing is a compromised process. After all, about 1,050 of the school's 1,450 students are limited-English speakers, and most of the English speakers are concentrated in the upper grades. So there aren't really enough designated English speakers to go around in the primary grades. But the school district says mix, so mix they do.
In this bilingual program, students are tested about once a year for fluency, with the goal of gradually introducing more English in the third, fourth and fifth grades. But actually becoming fluent in English could still theoretically take forever. It just happens when it happens, which, apparently, was perfectly okay with the state and many school districts until the last two or three years. Only recently did L.A. Unified establish loose time limits for how long it ought to take to learn English.
It is clear that most Eastman students are not learning English as quickly as they could. Take the alphabet displayed in another first-grade classroom. To illustrate the letter P, there's a picture of a door (for puerta). The letter Q is demonstrated by a picture of cheese (queso).
Wouldn't it help the cause of learning English if, for example, the illustrated alphabet taught words in English rather than Spanish? Of course it would, but again, teaching English is not the point. Instead, the push is for academic achievement, because students who become literate in their native language will eventually make a smooth transference to English.
The question is, do they?
When bilingual education first became a national education policy, the answer to that question was a resounding maybe.
At its inception, bilingual education was nothing less than a leap of faith, a leap taken without any basis in research when, in 1968, funding for bilingual education became part of President Lyndon Johnson's social-welfare program. But there was, at least, historical precedent. Public-school instruction in languages other than English took place more than 100 years ago in locales such as St. Louis and Milwaukee, Wisconsin – when non-English-speaking immigrants amassed numbers as well as some political pull. In 1917, during World War I, lawmakers forbade German not only in the classroom, but in any public place. And in the years that followed, a instruction in foreign languages continued to bear an unpatriotic taint, making even the few instances of bilingual education virtually disappear.
Then, in 1963, the Dade County, Florida, school system began an experimental bilingual program for children of the first wave of Cuban refugees to arrive there. The program, which aimed for complete bilingualism rather than just English mastery, was widely viewed as successful. But the effort mainly served prosperous, well-educated Castro opponents, so its success said little about how impoverished, sometimes illiterate students would fare. Nonetheless, by 1968, the federal government was handing out bilingual-education grants to school districts across the country. Later revisions of the grant rules would explicitly require participating schools to offer native-language instruction.
Various California districts had run isolated pilot programs, but it was only after a series of court cases that the Legislature, in 1976, required school districts to offer classes in a student's primary language. By this time, positive results from nascent research had persuaded California policymakers that native-language instruction was the way to go.
In the years since, the weight of evidence has solidly favored bilingual ed, especially the basic notion that students benefit from transitional use of their first language while they learn English. Indeed, few in the academic or research community have doubted the notion. The debate was more over specific methods: “maintenance,” which shoots for complete bilingualism after five to seven years, or “transitional,” which lowers its sights to a quick move to English within two or three years. In the political realm, however, the debate has devolved into an all-or-nothing contest between all-English and bilingual approaches. And within school districts, the goal of maintaining the primary language has been largely discarded, with the focus changing to how long students should be taught in their native language.
Any canon of bilingual advocates usually includes the work of researchers Jim Cummins, Virginia Collier and Wayne Thomas. Cummins, and many others after him, have argued that though children can quickly pick up enough English to get by on the playground, the fluency required for “academic proficiency” in English, i.e., for successful use of English in the classroom, takes a minimum of five years to develop. Collier and Thomas upped the ante, saying it takes five to 10 years in even the best situations.
The other side, the anti-bilingual-ed forces, have few friends inside academia. But one of the most vocal is Rosalie Porter, who is not a researcher, but a former teacher and school administrator whose book, Forked Tongue: The Politics of Bilingual Education, is based on her own experience in Massachusetts. She relies largely on the common-sense notion that the best way to teach English to children is by teaching them in English. It's known in the jargon as the “time-on-task” approach, based on the idea that the more time spent on the task of teaching English, the better the results. Porter cites the studies of researchers Christine Rossell and Keith Baker for support in arguing that English immersion is more effective than teaching in the native language.
Rossell and Baker had indeed decided that immersion is more effective at promoting English reading skills. They based this conclusion not on their own data, but on a 1996 review of 300 previously published studies. They eliminated all but 72 of these as methodologically flawed. Other researchers, using the same data, have reached a conclusion opposite to that of Rossell and Baker.
But even the work of Rossell and Baker does not point toward a Proposition 227-like solution, because they also find fault with the time-on-task theory, calling it unsupported by research. Moreover, they accept that “there is some initial period of time when native-tongue instruction actually benefits students,” opting for a brief period of first-language instruction, followed by a transition to all-English classes. The approach in Proposition 227, by contrast, is both more extreme (using little or no native language) and more curtailed (limiting the immersion program to one year). Opponents of Proposition 227 make the case that such an intensive one-year program to teach English may have the practical effect of costing students a year of academic instruction. The current debate also has inspired a middle ground of pro-bilingual-ed pragmatists who've broken away from the hardcore academics. These political leaders and educators, some of them Latino, have begun to flinch at the notion of fifth-graders still taking classes in Spanish after six years of “transition.” The pragmatists, who include outgoing state Assembly Speaker Cruz Bustamante, some school-board members and L.A. schools chief Ruben Zacarias, have begun to coalesce politically around a two- to three-year maximum for native-language instruction, regardless of what the gurus claim.
And there's some reason to question the gurus, according to a review last year by the National Research Council. “What has happened in this area of research is that most consumers of the research are not researchers who want to know the truth, but advocates who are convinced of the absolute correctness of their positions,” wrote the study's authors. “Because advocacy is the goal, very poor studies that support an advocated position are touted as definitive.” Moreover, “because there are studies that support a wide range of positions, advocates on both sides end up with plenty of 'evidence.'”
Despite this caveat, the research council concluded that, while “structured immersion” does offer benefits, providing some native-language instruction produces more clearly beneficial results. But even those positive benefits ascribed to structured immersion by the council don't apply to the program mandated by Proposition 227, which allows only one year for the process.
Some of the most compelling data on bilingual ed were collected in the 1980s at Eastman Elementary and four other California schools. The Eastman students showed small but statistically significant gains in English reading skills after the school's bilingual-ed program was revamped under the leadership of state specialists. The key features of the Eastman model include hiring a bilingual staff (and offering them regular training sessions), installing a complex and carefully structured bilingual program, and teaching academic subjects almost entirely in Spanish until the fourth grade, when the conversion to English is gradually made. Over time, this blueprint has begun an incremental march of conquest across L.A. Unified, existing in some form now at 22 schools under the name Project MORE. But the approach used at Eastman has not always proved successful. In one school that was part of the original study, test scores went down after its installation.
Ultimately, the National Research Council experts a reached a point of enlightened futility in their review of existing research when it came to determining what works best. They concluded that there was, in fact, “little value in conducting evaluations to determine which type of program is best.” There's not likely to be a one-size-fits-all approach, they said. In fact, “successful bilingual and immersion programs may contain many common elements.” The focus should be on what is most likely to work in a particular local setting, and it's okay, they said, to piece together elements from different approaches. Perhaps most notable of all the council's conclusions was that past studies fell short of the mark in answering vital questions about bilingual ed, leaving much of the basic research yet to be done.
In the late '70s, when California was coming onboard in the thrust to help non-native speakers, bilingual ed looked like the sure ticket, according to research then available. But more than that, the leadership at the state education department was of a philosophical bent to embrace new thinking. The state's Office of Bilingual Education was led by educators and bureaucrats, most in their early 30s and fresh out of grad school, who were neither traditionalists nor gradualists, but conscious trendsetters who saw, in bilingual education, a vehicle to orchestrate the academic and social rescue of an entire generation of impoverished immigrant children who were sweeping into the state's public schools.
The office was headed by Guillermo Lopez, a former boxer and onetime high school dropout who worked his way up to a doctorate in education before becoming a school-district administrator in Sacramento and moving on to the state bureaucracy. His key subordinates included three administrators who remain top state ed officials to this day: Norm Gold, fresh from a year in Costa Rica with the Peace Corps and a doctoral program at the University of Massachusetts, Amherst; Fred Tempes, another Peace Corps grad, who went on to study bilingual education with a teacher-corps project in Ventura; and Dennis Parker, a school administrator from Riverside County. Some outside academics also played influential roles, including Stephen Krashen of USC and Jim Cummins, who was flown in from the University of Toronto. Both continue to be among the leading academic advocates for bilingual ed.
To capture the latest thinking, the state ed department sponsored three seminar-style think tanks between 1979 and 1986 on bilingual education, English immersion and the effect of cultural factors on language learning, respectively – each time flying in noted experts from around the world. Books grew out of all three gatherings. “We thought it was a golden age,” recalled Krashen in an interview. The atmosphere “was like a high-level university seminar. There was no discussion of politics, of getting ahead.”
It was, indeed, a heady era, marked by the free exchange of ideas, but it also was more about concepts and good will than hard science. Many of these same academics also were sweeping out traditional notions about reading and math instruction, making policy decisions that became highly controversial. Since bilingual-ed curriculum was itself coming into being, this area became a natural home for these new teaching methods and theories. But the biggest decision was to push for a program rooted in primary-language instruction.
The revolutionary nature of the approach encompassed more than just trendy pedagogy. Prior to the push for bilingual education, school districts across the state practiced what can only be characterized as institutional racism. Latino students were frequently segregated from Anglo students, provided unequal resources and tracked unquestioningly into manual labor and factory jobs. Then, suddenly, in nothing less than a second-wave civil rights movement, school districts were told that they not only had to start educating Latinos, but they had to do it in their own language.
“Many school districts resisted providing any special help to students who spoke limited English,” said Dennis Parker, a program coordinator for the state. Or they would fail to identify all the students who needed help or refuse to make a serious effort to hire or train bilingual teachers. For the most part, however, “by the mid-'80s school districts stopped asking why and started asking how they could best serve these students.”
There was still a potent political backlash, namely, a latent English-only movement that spurred then-Governor George Deukmejian to veto an extension of the state's bilingual-education legislation, which officially expired in 1987. English-only advocates thought they had killed off bilingual education, but not so. The state education department cited both case law and the fine print of the expired laws to justify continued enforcement of the rules.
And enforcing the rules is what the state bilingual-ed division does best. The state's mandate has been a difficult one, requiring school systems to identify every student who's not fluent in English and provide a primary-language teacher for that student for five years or more. Even cooperative school districts have had to play perpetual catch-up to hire the needed teachers. And although 99 percent of limited-English students receive some special services, only 30 percent of the state's 1.4 million limited-English students are in classrooms that meet state specifications, that is, classes taught by a fully certified teacher fluent in the students' primary language. This teacher shortage remains a key point in the arsenal of bilingual-ed defenders, who've singled out the lack of resources and bilingual teachers as a principal reason for low test scores and the sluggish learning of English.
And they have a point. In L.A. Unified, 42 percent of the bilingual teachers have not completed their teacher training and are working with emergency credentials, compared to about 11 percent of the rest of district teachers.
School districts that couldn't achieve the ideal (which was all of them, really) had to show the state they were headed in the mandated direction. There was little room for variation and little thought about bilingual options that might be less ambitious, but perhaps more successful in the bargain.
“That department is a true-believer department,” said former state schools Superintendent Bill Honig in an interview. “The bilingual community has not been open enough to quality issues; it's been ideological issues instead.” Honig, who ran state ed from 1982 to 1993, said it took an interdepartment war to win for school districts the legal option of trying modified bilingual approaches through a “waiver,” and even then, state bilingual specialists would erect roadblocks and nitpick proposals.
In all, just 23 of the state's 1,000-plus school districts have ever received temporary waivers for portions of their bilingual programs. Four school districts have received general waivers since the rules were recently relaxed.
Of course, state officials have long operated on the premise that they know best for the children of California. Why shouldn't they make it hard for school districts trying to get over with a “lesser” plan? The state, however, in monitoring adherence to its bilingual recipe never gave sufficient thought to the resulting stew. Other than the Eastman study, which is more than 10 years old, state officials never challenged their own assumptions or followed up to see how their doctrine was faring.