It all started at Ninth Street Elementary. In early 1996, a group of Spanish-speaking parents pulled their children out of the school to protest its failure to teach English. They blamed bilingual education. The boycott lasted nearly two weeks, and it received extensive coverage in the Los Angeles Times — suggesting a new trend, perhaps even a sea-change, in Latino attitudes. These reports caught the eye of a Silicon Valley businessman named Ron Unz. “Parents shouldn’t have to carry picket signs to get English instruction for their kids,” Unz declared. Dipping into his personal fortune, he organized a statewide ballot initiative requiring that “all children … be taught English by being taught in English.” The measure, known as Proposition 227, passed overwhelmingly on June 2, 1998. It virtually outlawed bilingual education in California.

That is the official account, the story as conveyed by the victors and the news media. What actually happened — during the Ninth Street boycott and the Proposition 227 campaign — is more complicated. It is also instructive about the political adversity surrounding bilingual education. California’s experience highlights a second wave of English Only activism, one with a broader appeal than the first. Avoiding the rhetoric of ethnic bigotry, it evokes the principles of equal opportunity, parental choice, and pedagogical effectiveness. Instead of warning about the menace of “bilingualism,” it stresses the importance of “English for the Children.” In effect, it embraces key premises of bilingual programs to undermine their public support. This strategy proved an unqualified success in California, as voters dealt the most serious setback to bilingual schooling since World War I.

Manufacturing a Myth

Located in downtown Los Angeles, on the edge of Skid Row, the Ninth Street School faces more than its share of challenges. It serves about 460 students in pre-kindergarten through the 5th grade. About half are classified as homeless, living in downtown shelters and single-room-occupancy hotels. Others are bused into the area each morning with their parents, mostly recent immigrants from Mexico who work in garment factories nearby. Nine out of ten children are limited-English-proficient (LEP); 99 percent are eligible for free or reduced-price meals. All but a few come from homes where only Spanish is spoken. These factors help to explain why Ninth Street students, like their counterparts in similar schools, score well below national norms when tested in English.

Nevertheless, prospects brightened following the appointment of Eleanor Vargas Page as principal in 1993. The school entered into partnerships with local businesses, civic groups, and social service agencies. Academic expectations increased and so did the amount of English instruction.(1) Ninth Street won a $600,000 Title VII grant to support extended learning opportunities, before and after school, and to expand the school library. Student attendance and parent participation soared. The results were impressive: children’s English scores rose by 35 percent over a four-year period.

So when a reporter called on February 9, 1996, with news that a meeting of 63 Ninth Street parents had just voted to boycott the school, “I was in shock,” Vargas Page recalls. She had always been proud of her close ties with parents and their high attendance at school functions, where complaints about bilingual education had never come up. None of the boycotting parents had asked for their children to be removed from the program; in fact, all had recently signed forms consenting to their children’s enrollment. “The conflict was not here in the school,” the principal believes. “The complaints were initiated by [an outsider], not by the parents.”

The outsider was Alice Callaghan, an Episcopal priest who ran a community center, Las Familias del Pueblo, that provided daycare for about one-quarter of Ninth Street students. Callaghan was also a veteran political organizer who had skirmished with city officials on numerous issues. Now she took up a new cause: abolishing bilingual education. For months she wrote to the Los Angeles Unified School District demanding English-only instruction at Ninth Street. When she received little response, Callaghan urged the parents to take direct action.

Sensational headlines followed: “80 Students Stay Out of School in Latino Boycott … Bilingual Schooling Is Failing, Parents Say.” For the news media, conflict means good copy. This is especially true when the conflict seems unlikely — known in the trade as a “man bites dog” story. That’s what drew journalists to Ninth Street: Downtrodden immigrants, led by a colorful activist, were using civil-rights tactics to protest a “politically correct” program supported by their own ethnic leaders. Amazing! The news accounts neglected, however, to clarify a key point. This was a needless conflict — a drama that was staged precisely to generate sensational headlines.

California, like other states, has long recognized parents’ right to remove their children from bilingual instruction if they so choose. At Ninth Street, all they had to do was come in, meet with their child’s teacher and principal, and hear about the educational options. Then, if they chose to enroll their children in the school’s alternative program taught mostly in English — as a few had done earlier that year — they merely needed to sign a consent form.(2) Vargas Page felt this procedure was essential “for a parent to make an informed decision.” Callaghan called it “harassment” and “intimidation.” She advised parents to refuse to attend any such conference, keep their children out of school, and send them to Las Familias (where “we will speak only English with them”) until the district gave in to their demands. Several days into the boycott, she circulated a form for parents to sign — in English, although few could read the language — authorizing student transfers to the alternative program.(3)

No doubt some of Alice Callaghan’s followers were convinced that bilingual education was to blame for their children’s academic problems. But others told school staff they felt they had no choice but to join the boycott and remove their children from the program, believing that otherwise they would lose the free daycare at Las Familias. In an interview, Callaghan denied making any explicit threats along these lines. Whatever the case, it is clear that she failed to assuage such concerns or to reassure parents that these decisions were entirely their own. Virtually none of those with children at Las Familias resisted her advice to transfer their children out of bilingual education (although a few would later change their minds).

What was the effect on LEP students? Two years later, only two out of the 74 moved into intensive English instruction had been reclassified as fluent in English. On state-mandated achievement tests, the 5th graders scored at the 11th percentile in reading, the 15th percentile in language, and the 16th percentile in math — well below peers who had remained in bilingual classrooms.(4) Questioned about these disappointing outcomes, Callaghan refused to accept responsibility. “If it’s the older kids, that’s not our fault,” she said. “It’s a result of their terrible bilingual program” provided to these children in the early grades.

Despite its dismal academic results, the boycott was a public relations bonanza for opponents of bilingual education. Soon it mushroomed into a national story. Spanish-speaking parents were quoted — often in translation — citing the importance of English to their children’s future. Los Angeles Mayor Richard Riordan endorsed their protest. Ninth Street was portrayed as indicative of bilingual education’s ineffectiveness and unpopularity among those it was intended to serve. Critics’ views were prominently featured. Misinformation went unchallenged. The Los Angeles Times asserted that bilingual education “trapped” children in all-Spanish classrooms for “six or seven years. … That’s much too long.” Readers never heard an effective response.

This imbalance was not entirely the fault of the news media. Board members and district officials reacted as though the controversy concerned practices at a single school and promised to “look into it.” Few bilingual education advocates came forward to assist their beleaguered colleagues at Ninth Street. Facing a hostile press, Eleanor Vargas Page and her staff were largely on their own in defending the school. Their side of the story usually came out garbled — when it came out at all. By and large, the public never heard a coherent case for bilingual education. L.A. Unified’s failure to provide one conveyed an impression of arrogance and unresponsiveness to parents’ legitimate demands for English.

Thus the Ninth Street myth was born. Over the next two years, it would prove more damaging than anyone had foreseen.

Calls for “Reform”

Critics had long complained that, despite the “sunset” of California’s bilingual education law in 1987, the state kept most of the old requirements in place (see Chapter 10). Citing federal and state civil-rights guarantees for LEP students, the California Department of Education (CDE) continued to mandate — and enforce — the use of native-language instruction “when necessary … [to] provide equal opportunity for academic achievement.” The policy also required districts to staff its bilingual and English-as-a-second-language (ESL) programs with an adequate number of qualified teachers. In the prevailing political climate, however, Sacramento could offer them little practical help in achieving these goals.

To come from behind and win reelection in 1994, Governor Pete Wilson promoted a crack-down on undocumented immigrants known as Proposition 187. One of its provisions would have thrown their children out of public schools.(5) Not surprisingly, the Wilson administration saw little political gain in funding programs to train teachers for language-minority students. By 1997-98, the CDE estimated the shortage of bilingual teachers at 27,000 and rising. This left districts unable to provide bilingual classrooms for more than a minority of their LEP students — about 30 percent, on average, statewide.(6)

Meanwhile, the California Association for Bilingual Education (CABE) resisted proposals in the legislature to allow more “local flexibility” in pedagogical matters. Since the mid-1980s, CABE had built an effective lobbying operation. It remained committed to the prescriptive philosophy, based on experience with districts that had to be prodded constantly to meet their obligations to English learners. The problem remained prevalent, for example, in the Central Valley, where language-minority communities often lacked political influence on school boards. Year after year CABE relied on the clout of Latino legislators in Sacramento to block any attempt to relax state requirements and oversight.

Yet prescriptiveness proved a double-edged sword. On the one hand, districts could be required to respond to the needs of English learners, providing additional resources, trained staff, and better designed programs. On the other hand, mandates inevitably bred resentment. The same had been true in the 1970s, when the U.S. Office for Civil Rights (OCR) imposed Lau plans forcing districts to try native-language approaches (see Chapter 2). Aggressive enforcement created a backlash that forced OCR to retreat — but not before it had established bilingual education as a viable pedagogy. The tradeoff was harder to justify in the 1990s, when that pedagogy was embraced, for the most part, by California’s “education establishment.” Public skepticism, English-only campaigns, and attacks by conservative politicians posed more formidable challenges than resistance at the school district level.

The CDE’s encroachment on local control — and CABE’s vigorous defense of the policy — presented an inviting target for anti-bilingual forces. Their charges of bureaucratic “heavy-handedness” and “one size fits all” schooling were effective, if exaggerated. No one seemed to notice that by 1997 only five school boards — all in conservative Orange County — had petitioned the State Board of Education for “waivers” of the native-language requirement; none was denied. California’s other districts also enjoyed considerable flexibility in designing programs for English learners, as testified by wide variations in program design. Nevertheless, the native-language requirement made bilingual education an easy scapegoat when schools failed these students.

Prescriptiveness also put the burden on advocates of bilingual approaches to prove their superiority in practice over English-only approaches. Given the limited quantity and quality of achievement data for LEP students in California, that was rarely possible. Nationwide, program evaluation research continued to provide relatively weak support — as compared with the findings of basic psycholinguistic research — for the effectiveness of bilingual instruction (see Chapters 5-6). It has become routine to blame bilingual educators for failing to prove the worth of their programs. Yet scientific comparisons of LEP program models are difficult to design and expensive to execute. Federal and state policymakers have seldom been generous in funding such research. Since the Ramírez report, commissioned in 1984 and completed in 1991 (see Chapter 7), the U.S. Department of Education has deemphasized large evaluation studies and concentrated instead on reviews of existing research.

National Research Council Report

The most significant of these reviews appeared in 1997. Improving Schooling for Language-Minority Children: A Research Agenda summarized the findings of an expert panel of the National Research Council. In addition to its chair, Kenji Hakuta of Stanford, several other participants were prominent supporters of bilingual education. On the effectiveness question, however, their conclusions were equivocal:


It is difficult to synthesize the program evaluations of bilingual education because of the extreme politicization of the process. 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. The beneficial effects of native-language instruction are clearly evident in programs that are labeled “bilingual education,” but they [??] also appear in some programs that are labeled “immersion.” There appear to be benefits of programs that are labeled “structured immersion,” although a quantitative analysis of such programs is not yet available. There is little value in conducting evaluations to determine which type of program is best. The key issue is not finding a program that works for all children and all localities, but rather finding a set of program components that works for the children in the community of interest, given the community’s goals, demographics, and resources.


Complaints about politicization had been heard before. When the discussion of LEP education policy became polarized in the 1980s, focusing almost exclusively on the language of instruction, it tended to discourage open-minded efforts to analyze research evidence and improve programs. The restructuring of Title VII grant categories in 1994 was one attempt to mitigate this problem (see Chapter 12).

What was new in the NRC report was its suggestion that the debate is no longer relevant. That both bilingual and English-only program models can be beneficial. That the federal government should stop funding expensive yet futile attempts to determine which is superior. That researchers on both “sides” needed to behave more like scientists and less like advocates.(7) “We need to think in terms of program components,” the panel recommended, “not politically motivated labels.” It called for “theory-based interventions” that could be evaluated more scientifically and for “a developmental model … for use in predicting the effects of program components on children in different environments.”

The report also raised eyebrows with its generous, albeit ambiguous, words about “immersion.” Its favorable assessment relied heavily on studies of an El Paso program whose English-only character had long been disputed; this so-called “bilingual immersion” model featured substantial amounts of Spanish instruction (see Chapter 7). Some suspected that, with its even-handed findings, the NRC panel was stretching the evidence to fit a preconceived agenda of its own.

Depoliticizing the research debate, considering the diversity of LEP students and their needs, restoring a measure of scientific detachment … these goals sounded worthy on paper. In reality they proved problematic. Opponents of bilingual education were quick to seize on the NRC’s findings as vindication of their views. Rosalie Porter’s READ Institute, a project funded in part by U.S. English, published a lengthy analysis of the report by Charles Glenn of Boston University. Its main theme was that, after a generation of experience with bilingual education, we still know virtually nothing about whether it works. Hakuta disputed this interpretation, condemning what he called “the far from impartial attempt by READ to place its own political spin on this matter.” His statements, however, did little to settle what the NRC panel had meant to say. Like the research literature it criticized, the report had something for everyone, enabling partisans to pick and chose findings that served their purposes.

Meanwhile, the political divide showed no sign of narrowing. Language of instruction was becoming, if anything, more contentious. Journalists and commentators highlighted the panel’s complaints about the quality of program evaluation studies — “design limitations … poorly articulated goals … extreme politicization” — and pronounced all research in the field useless to policymakers. If the science was inconclusive about what works, why not encourage experimentation and flexibility? they argued, reviving a familiar theme of the 1980s (see Chapter 4).

In California, State Senator Deirdre Alpert and Assemblyman Brooks Firestone sponsored a bipartisan bill along these lines. It proposed to relax the mandate for native-language instruction and allow local districts to choose their own approach, while requiring them to assess LEP students annually and show progress over time. Programs would have to be restructured if children failed to meet goals within three years. Some bilingual education advocates saw the measure as a compromise worth exploring, believing that few districts were likely to dismantle existing programs. Others objected that its accountability provisions were too weak and its sanctions too vague, thus encouraging schools to scale back their efforts for LEP children. Although the Alpert-Firestone bill easily passed the California Senate in the summer of 1997, for the third year in a row CABE managed to kill it in the Assembly. Self-described “moderates” expressed frustration and blamed bilingual educators for obstructing change. It was an opening tailor-made for Ron Unz.

English Only, Phase II

A software millionaire, aged 36 and single, with no children in school and no background in education, Unz seemed an unlikely antagonist. His first foray in politics had come three years earlier, when he challenged Pete Wilson for the Republican gubernatorial nomination. He won about a third of the vote, mainly from Far Right critics of the incumbent. Yet he also emerged as a staunch opponent of Proposition 187, earning “pro-immigrant” credentials that distinguished him from other English Only advocates. Unz believed that Republicans needed to face demographic realities and reach out to fast-growing minorities in states like California. Rather than side with nativists, he argued, the party should cater to the “natural conservatism” of Hispanic and Asian Americans on issues like welfare, crime, and abortion. Campaigning in 1994, the challenger preached the gospel of upward mobility through assimilation, while denouncing “the poisonous brew of bilingual education, multiculturalism and other ethnic-separatism policies.”

These views and the wherewithal to promote them gave Unz access to prominent conservatives like Linda Chavez. He helped to endow her Center for Equal Opportunity, a Washington advocacy group that opposes affirmative action and bilingual education on “civil rights” grounds. Impressed with the ideas of Peter Salins, he commissioned the sociologist to write Assimilation, American Style, a tendentious history that portrays earlier immigrants as eager to abandon their languages in favor of English. Meanwhile, Unz kept his distance from traditional English Only lobbies.

The influence of groups like U.S. English appeared to reach a high-water mark on August 1, 1996. By a vote of 259-167, the House of Representatives passed the so-called “English Language Empowerment Act,” the first official-English bill at the federal level (see Appendix D). Mindful of a veto threat by President Clinton, the Senate declined to consider the measure. If enacted, it would have banned most federal publications in languages other than English, repealed bilingual voting rights, mandated English-only naturalization ceremonies, and shielded English speakers from “discrimination.” Republicans claimed the legislation was essential to preserve the nation’s “common bond” and “empower” immigrants by motivating them to learn the language. Democrats condemned it as divisive, mean-spirited, and potentially unconstitutional in its restrictions on minority access to government. In practice, the bill would have affected relatively few people. It was the precedent that stimulated interest on both sides.(8)

Likewise, in criticizing bilingual education, English Only advocates had always stressed symbolism over substance. Rhetorical attacks aside, they never dared to mount a legislative campaign to destroy the program. Fearing they would be seen as callous toward children, even the most draconian official-English proposals featured an exemption for bilingual education. Evidently they failed to grasp its vulnerability.

Ron Unz brought a different approach to English Only politics. Rather than rely on official language declarations to oppose “bilingualism” in government, he launched a frontal assault on the most important bilingual program. Instead of gauzy rhetoric about English and “American” identity, he used specific arguments about educational effectiveness. Rather than blame immigrants for failing to learn English, he posed as their advocate against unresponsive schools.

It was a more sophisticated strategy than the visceral politics of resentment that guided Phase I of the English Only movement. Unz’s arguments sounded rational and public-spirited by contrast. They also exploited ignorance about language acquisition, which extended into liberal and progressive sectors of the electorate. As a result, he effectively changed the terms of the debate. Instead of “Should bilingual education be ‘reformed’ and made more ‘flexible’?” the question became: “Should bilingual education be eliminated?”

Proposition 227

Unz’s first step was to disassociate himself from California’s anti-immigrant fringe. In the spring of 1997, he recruited prominent Latinos to help spearhead his ballot initiative, which he dubbed “English for the Children.” These included his cosponsor, Gloria Matta Tuchman, a 1st grade teacher from Santa Ana and a perennial candidate for state superintendent of public instruction.(9) Jaime Escalante, the legendary calculus teacher of Stand and Deliver fame, agreed to serve as “honorary chairman.” Several lesser luminaries also lent their support. Alice Callaghan helped to kick off the campaign with a media event at Las Familias, where scenes of brown-skinned children provided a backdrop for speeches attacking bilingual education. Unz established his headquarters nearby, just a few blocks from the Ninth Street School.

Petitions for Proposition 227 began circulating in July. Qualifying an initiative statute for the ballot required sponsors to gather 433,000 signatures from registered voters — an enormous hurdle for grassroots volunteers, but not for those able to pay. Unz simply opened his checkbook and spent $500,000 to hire canvassers.

He made a point of circulating petitions in East Los Angeles, home to a large Mexican American community. Claims of minority support became a key selling point for the initiative after a Los Angeles Times Poll, conducted in October, reported it was favored by 84 percent of Latinos. Small wonder, since the survey portrayed the measure as primarily an effort to improve English instruction for children who needed to learn the language.(10) Few Californians had heard about its extreme provisions, much less read the fine print (see Appendix E).

Proposition 227 was a complex, often confusing proposal crafted to meet conflicting political and policy goals. Unz sought to outlaw a program that many parents wanted without appearing to restrict parental choice; to tie the hands of school boards that favored bilingual education without seeming to usurp their authority; and to eliminate protections for LEP students while shielding the law from civil-rights litigation. Its highlights are as follows:


  • LEP students must be taught in “sheltered English immersion” classrooms “during a temporary transition period not normally intended to exceed one year.” Their instruction must be delivered “overwhelmingly [in] English” by teachers who “possess a good knowledge of the English language.” Students may be mixed by age and grade. They must be transferred to mainstream classrooms after they have attained “a good working knowledge of English.”

  • Parents may request “waivers” of the English-only rule under limited circumstances: (a) if their children already score at or above grade level in English; (b) if they are at least 10 years old and educators believe a bilingual program would foster “rapid English acquisition”; or (c) if they are under 10 and have “special physical, emotional, psychological, or educational needs” that would better be served through native-language instruction.

  • Teachers, administrators, and school board members who “willfully and repeatedly” violate the law’s provisions may be sued by parents and held personally liable for financial damages and plaintiffs’ legal fees.

  • The California legislature must appropriate $50 million each year to provide ESL instruction for adults who agree to tutor children in English.

  • Proposition 227 may be repealed or amended only by a two-thirds vote of the legislature and approval of the governor, or by another ballot initiative.


When the initiative was unveiled, California’s education and civil rights communities were naturally alarmed. A long list of advocates came together to plot strategy, including CABE, the California Teachers Association, the Association of California School Administrators, the Mexican American Legal Defense and Educational Fund, California Tomorrow, and the Northern California Coalition for Immigrant Rights. But an opposition campaign was slow to get under way. Finally, in mid-November, a coalition calling itself Citizens for an Educated America held its first press conference — four months after Ron Unz had launched his campaign. Meanwhile, his charges against bilingual education went largely unchallenged. Though not unreported.

News media were drawn to the high-stakes controversy and covered it in lavish detail. Unz took to the campaign trail full-time. He carefully stayed “on message,” repeating the same arguments at every stop. His pitch was straightforward: “Begun with the best of theoretical intentions some twenty or thirty years ago, bilingual education has proven itself a dismal practical failure. … Enormous numbers of California children today leave years of schooling with limited spoken English and almost no ability to read or write English.” He blamed “government efforts to prevent young immigrant children from learning English,” despite the wishes of parents at schools like Ninth Street. “During the past decade, the number of these non-English-speaking immigrant children has more than doubled. Yet under the current system, centered on bilingual education, only about 5 percent of these children each year are found to have gained proficiency in English. Thus our state’s current system of language instruction has an annual failure rate of 95 percent” (his emphasis).

Unz’s mantra was simplistic but powerful. It was widely disseminated by journalists, who seldom challenged his evidence or assumptions. Few prospective voters understood that, because of teacher shortages, fully 70 percent of California’s LEP students were not enrolled in bilingual classrooms. If the “current system” was indeed failing, it was more logical to blame the scarcity, not the excess, of native-language instruction. Moreover, Unz’s arbitrary standard of success — one year to learn English — bore no relation to the realities of second language acquisition. Educators understood that, while children quickly acquire “playground English,” they typically need four to seven years to acquire the decontextualized, cognitively demanding English required to excel in school. The annual “redesignation rate” — which averaged 7 percent statewide in 1997 — fluctuated significantly from district to district. It was affected by demographic patterns, variations in assessment procedures, and other factors unrelated to achievement. Using this dubious statistic to “hold schools accountable” implied that rapid English acquisition was the best measure of student progress. Yet research showed the opposite: programs that stressed a gradual transition to English were correlated with long-term academic success.(11)

On the other hand, there was limited research supporting the “sheltered English immersion” approach that Unz sought to impose by law. There was none whatsoever showing it could successfully mainstream children within 180 school days. One of the few rigorous studies in this area, the 1991 Ramírez report, found that after one year in English immersion programs only 4 percent of students had become fluent in English. After four years 67 percent had been redesignated, as compared with 72 percent of those in transitional bilingual education.

While such rebuttals occasionally appeared in the press, they did not lend themselves to snappy sound-bites. Voters tended to remember the “95 percent failure rate.” In candid moments Unz conceded the figure was misleading, but he found it too useful to abandon. Journalists never exposed the fraud. Appealing to folk wisdom about how languages are acquired — the younger the better … through “total immersion” … without the “crutch” of bilingual support — he dismissed research in the field as worthless, motivated by ethnic politics or ivory tower “looniness.” Many reporters, after a cursory glance at the NRC report, were inclined to agree.

Unz also played to their habitual cynicism, painting bilingual educators as a vested interest — resistant to legislative reform and more interested in taxpayer subsidies than student achievement. The news media recycled these charges, while rarely questioning Unz’s own motives once it was established that he was not a nativist zealot.

Meanwhile, it seemed that every story trumpeted his lead in the polls, implying that Proposition 227 was unbeatable. This seemed to intimidate the traditional allies of bilingual education, including Latinos and liberal Democrats. Few spoke out aggressively against the measure. Those who finally did, in the late stages, conceded the program needed a complete overhaul.

How to Respond?

Bilingual education advocates were in a quandary. Anti-immigrant bias was an obvious factor in the initiative’s popularity; yet Unz had immunized his campaign from the charge of racism. Voters who truly cared about LEP children were badly misinformed; yet there was little time to give them a crash course in the theory and practice of bilingual education. Media bias had become a major obstacle; yet the No on 227 campaign needed the help of journalists to get its message out. The California Teachers Association was a potential source of funds to buy TV advertising; yet the giant union was ambivalent in its support for bilingual programs. A deal remained within reach on “reform” legislation that might head off the initiative; yet the coalition was divided over whether to compromise.

Citizens for an Educated America recognized its need for professional help. It hired campaign and media consultants, along with pollsters to conduct surveys and focus groups. The professionals recommended a counter-attack against Proposition 227. This election should not be a referendum on bilingual education, but a referendum on Unz’s proposal, they argued. Highlighting its extreme provisions would help to win over “swing voters” who had yet to form an opinion. Such voters were not to be found among immigrants, many of whom were noncitizens, or among Latinos and other language minorities, who usually turned out in small numbers. Given California’s electoral realities, the swing voters would have to be found among white, affluent, older, English-speaking moderates. In particular, the consultants argued, the No campaign should target “Republican women over 50.” A winning message — one that would appeal to the undecided — could not be built on challenging the conventional wisdom. It could not claim success for programs that were widely perceived to be failing. In short, they recommended: “Don’t defend bilingual education.”

This advice came as a bitter pill for many practitioners and advocates of the program. It seemed like a betrayal of everything they had worked for, a capitulation to demagogues who would risk children’s futures for political gain. Lies about bilingual education were everywhere. How could Citizens fail to refute them? Still, it was hard to deny the situation was desperate. With their polling results and political savvy, the professionals were convincing. “Put aside your personal feelings,” they said in effect. “This strategy is the only hope of saving bilingual education. Given a free rein and sufficient resources, we can beat Proposition 227.” After much internal debate, the coalition agreed.

Others viewed the decision as not merely misguided but suicidal. In private, some said, the strategy could have hardly served Ron Unz’s purposes better if he had designed it himself. Indeed, when school officials defended bilingual education, Unz warned they could be prosecuted under a California law that prohibited public employees from taking sides on pending legislation. Now bilingual educators and even immigrant advocates working in Latino communities were feeling the same pressure from the Citizens consultants. Whenever the program came up in debates or press interviews, they were advised to change the subject, saying: “I’ll be happy to discuss the merits of different bilingual education programs on June 3 [the day after the election] — assuming the Ron Unz Initiative fails and we can still have a meaningful conversation.” Unz cited the “Don’t Defend” strategy as evidence that the program was indefensible.

Journalists were incredulous, not to mention frustrated, when Citizens refused to respond to his attacks. But they did not stop writing about issues of educational effectiveness. For the press and the public, Proposition 227 remained very much a referendum on bilingual education.

Some advocates chose to go their own way. Notwithstanding his long alliance with CABE, Stephen Krashen was among those who broke with its leadership. He saw the initiative as a rare opportunity to educate Californians about second-language acquisition. The more they learned about the rationale for bilingual education, he reasoned, the more likely they were to support it. Krashen joined with like-minded colleagues in a project to influence media coverage by debunking Unz’s claims.(12) Numerous other researchers, teacher trainers, and school administrators came forward — as individuals — to defend bilingual education in public forums. Yet without central coordination by Citizens, these efforts were sporadic and unfocused; their impact on the press was minimal.(13) Bilingual teachers and immigrant advocates formed local committees, especially in Latino communities, to activate parents, staff phone banks, and get out the vote. They enjoyed some success. Apart from fundraising drives, however, grassroots organizing received limited support from the No on 227 campaign.

Citizens needed large sums because its strategy relied heavily on TV advertising. Based on private opinion surveys, the campaign singled out one feature of Proposition 227 for special attack: the $50 million annual appropriation for adult English instruction. Even though that was barely 1/6 of one percent of the state’s K-12 education budget, Californians resented such spending to benefit immigrants, according to the Citizens polls. Ironically, Unz had inserted this provision to guard against attacks from the Left. It was intended to fortify his “pro-immigrant” guise and draw criticism from anti-immigrant zealots, making him look moderate by comparison. Now Citizens took the bait. Advocates who had lobbied over the past decade for more immigrant ESL classes suddenly sounded like fiscal conservatives, denouncing “a new spending program — not in our schools — but to teach adults English.” The hypocrisy was hard to conceal. Unz charged that his opponents were so desperate that they were willing to abandon cherished principles. For once he was right.

A lion’s share of No on 227 spending — which exceeded $4.7 million for the campaign — went for media buys in the final two weeks. Unz ran little advertising; there was no need, considering the lopsided polls. Contrary to expectations, Citizens outspent him nearly five to one overall.(14) Proposition 227 still passed easily, by a margin of 61 to 39 percent.

The only surprise was the Latino vote — two to one against the initiative — exactly the opposite of pollsters’ predictions.(15) In at least this one respect, bilingual educators were vindicated. The program’s strongest constituency had not forsaken it, as opponents claimed. Indeed, a survey by Spanish-language media found no decline in approval rates: 68 percent of Latinos in Los Angeles favored bilingual education, including 88 percent of those with children in such classrooms.(16) Spanish-speaking parents campaigned actively to defeat Proposition 227 in some areas. In Santa Barbara and Orange County, respectively, hundreds of them staged boycotts and filed litigation against local decisions to dismantle bilingual programs. (Journalists paid little attention, however, even as they continued to publicize the two-year-old protest at Ninth Street.)

Nevertheless, parent activism against the initiative remained more the exception than the rule — an unhealthy sign. When beneficiaries of a controversial program offer mainly passive support, political trouble cannot be far away. Thus the Proposition 227 story is a cautionary one for bilingual educators across the country. Ron Unz has announced plans to finance an anti-bilingual initiative in Arizona; similar efforts are expected in other states.


Meeting at Lake Tahoe in August 1998, veterans of No on 227 shared their views about what had gone wrong. Some defended the campaign strategy and argued that defeat was inevitable, given the demographics of California voters and their appetite for racist initiatives. Others insisted the outcome might have been different if Citizens had offered a straightforward rationale for bilingual education — something most voters had never heard — instead of insulting their intelligence with diversionary gimmicks. No consensus was achieved on these matters. Still, there was no shortage of mea culpas in explaining the victory of Proposition 227:


  • Inattention to the public image of bilingual education. Facing increasing opposition since the mid-1980s, the field has failed to respond proactively. Professional organizations have rarely used the news media to showcase success stories or to elaborate the mission of bilingual education. School districts have often neglected to communicate with parents, for example, to assuage worries about programs that stress a gradual transition to English. Researchers have seldom strived to make their findings accessible to a broad audience. Bilingual educators have tended to circle the wagons and complain to each other about unsympathetic colleagues rather than challenge their misconceptions. No wonder the field has become politically isolated.

  • Limited efforts to marshal data on outcomes. The field has tended to ignore the importance of test scores to combat skepticism about bilingual education. Notwithstanding the difficulties in assessing LEP students, the public is not unreasonable to demand accountability for programs that have been in place for years. To their credit, a few California districts scrambled during the campaign to release data comparing the outcomes of children in bilingual and English-only classrooms. Unfortunately, the data sometimes had flaws that tended to discredit legitimate claims of success. Trained researchers are needed to assist in analyzing these results. Virginia Collier and Wayne Thomas have developed a promising model for such studies and, in the process, have greatly expanded the database on program effectiveness. Since 1994, they have reported overall patterns that confirm theories underlying bilingual education, such as a correlation between native-language development and long-term academic achievement. Their work seemed to offer powerful evidence against the claims of Ron Unz. Yet Collier and Thomas declined to release sufficient data — in the view of many colleagues — for others to assess their findings. As a result, the researchers missed an opportunity to influence the debate over Proposition 227.

  • Resistance to legislative “reform.” Over the past decade — an era of conservatism, fiscal and otherwise — bilingual educators have adopted a defensive posture, relying on backroom deal-making to block change. They have stressed mandates to maintain funding levels and force schools to meet their obligations to LEP students. In short, they have become defenders of the status quo. Refusal to compromise may be “principled,” but it can also sacrifice chances to win needed improvements — for example, to remedy the chronic shortage of qualified teachers. In addition, it strengthens the stereotype of bilingual educators as a vested interest rather than a group of dedicated professionals, sometimes provoking more extreme attacks like Proposition 227.

  • Loss of ties to the grassroots. Bilingual education was not a gift from above, but a victory of mass struggle below. Without the efforts of determined parents and community leaders, the Bilingual Education Act and the Lau v. Nichols decision would never have materialized. Gradually, the program was supported by government, accepted by school boards, studied by researchers, and sustained by a corps of experts, lawyers, and bureaucrats. In short, it became institutionalized. To the extent that it became a domain of professionals, it became less of an activist cause, less of a social movement. Parents, once its strongest political base, were reduced to a passive role — as consumers of bilingual education rather than participants.


Whether any decision by No on 227 could have altered the outcome is impossible to say. There is little doubt, however, that greater attention to these problems during the campaign would have left advocates in a better position to cope with the initiative’s aftermath. Because of legal challenges(17) and ambiguous language, the final interpretation of Proposition 227 could take years to sort out. Ultimately, its impact on children will depend not only on litigation but on political pressures that each side can bring to bear to influence school boards, district administrators, and state officials.

Meanwhile, the fallout from Proposition 227 is spreading beyond California’s borders. In several states with the ballot initiative process, bilingual education is vulnerable to Unz-like assaults. Elsewhere, there are proposals to impose arbitrary time limits — typically three years — on children’s enrollment in language assistance programs.

Rep. Frank Riggs (R-Calif.) sponsored the latter proposal in Congress during the summer of 1998, borrowing Unz’s argument that students are languishing “too long” in bilingual programs. His bill, approved by the House on a party-line vote, would have also turned Title VII into a “block grant” program administered by the states, while voiding numerous Lau plans, federal court decisions, and consent decrees involving the civil rights of LEP students. With an eye on the mid-term elections and the voters’ impatience to mainstream English learners, the Clinton White House was tempted to endorse a version of the three-year limit. But following an angry response from the Congressional Hispanic Caucus, it reconsidered and blocked the Riggs bill.

Nevertheless, the issue was hardly settled. It is expected to resurface during 1999 deliberations to reauthorize the Improving America’s Schools Act. Similar battles loom in several states. Bilingual education will likely face jeopardy as never before. How advocates respond — whether they can regroup and formulate effective strategies to defend the program — will determine the future of language-minority schooling for years to come.


1. According to Eleanor Vargas Page, students receive a minimum of two hours of English daily in the early years, with the amount of Spanish instruction decreasing until children are reclassified as fluent English proficient (FEP), usually by the 4th or 5th grade.

2. Critics have frequently charged that children are misassigned to bilingual programs, typically on the basis of Hispanic surnames, and that parents’ requests for English-only instruction are denied. Such mistakes may sometimes occur. There is no evidence, however, that such problems are pervasive. Norman Gold, the official in charge of enforcing civil rights guarantees for English learners in California, says his department receives “scores of written complaints” each year from parents unable to get bilingual instruction for their children. “But records going back over more than a decade show that there have been no complaints alleging that parents have been unable to remove their children from bilingual instruction.”

3. One parent who checked off the English-only box wrote below: “Quiero que mi hijo siga en la clase bilingüe porque quiero que es mejor para su futuro” (I want my son to continue in the bilingual classroom because I believe it is better for his future).

4. By contrast, 5th graders in bilingual programs at Ninth Street scored at the 27th, 31st, and 38th percentiles respectively. Scores are from the Stanford 9 test administered in the spring of 1997. Direct comparisons are difficult because some students were excused from taking the test because of their limited English proficiency.

5. This and most other features of Proposition 187 were ruled unconstitutional in 1998.

6. Even in these classrooms, about one-third of teachers were still working toward their bilingual certification. According to the CDE’s annual language census, another 21 percent of LEP students received lessons taught in English with “native-language support” from paraprofessionals. Various forms of English-only instruction were provided to 32 percent, and 16 percent received no special help whatsoever.

7. The plague-on-both-houses theme came through even stronger in an NRC press release announcing the study on January 14, 1997: “Political debates over how children with limited English skills should be taught are hampering research and evaluation of educational programs established to meet the needs of these children. … Much research has been used in trying to determine which type of instruction is better — English-only or bilingual. However, there is little value in using research for this purpose. … Instead of attempting to single out one method for all students, research should focus on identifying a variety of educational approaches that work for children in their communities, based on specific local needs and available resources. … Evaluations have proved inconclusive about which teaching approaches work best. … Because many current studies are attempting to compare different types of programs that vary widely in such areas as funding, classroom setting, student background, and subject matter, the studies are unlikely to settle the debate over which type of instruction is best. … Advocates on many sides of the issue have been able to use research to uphold their arguments because there are study results that support a wide range of positions. These debates confuse policy-makers and muddle research agendas. …The committee called for a model for research and development that would be grounded in knowledge about the linguistic, social, and cognitive development of children. …”

8. Speaker Newt Gingrich championed the English Only cause during House floor debate, hoping to give Republicans a boost in the fall campaign. But it seemed to have little impact at the polls except to drive Latinos further from the party. Gingrich would not make the same mistake again. In the 105th Congress, he refused to allow a vote on the legislation.

9. Cosponsoring Proposition 227 gave Tuchman the name recognition to wage a serious challenge to the incumbent, Delaine Eastin, in 1998. But she lost in the November election.

10. Here’s how the question was posed: “There is a new initiative trying to qualify for the June primary ballot that would require all public school instruction to be conducted in English and for students not fluent in English to be placed in a short-term English immersion program. If the June 1998 primary election were being held today, would you vote for or against this measure?” Overall, 80 percent of likely voters said yes and 18 percent said no. Later surveys by the Times and the Field Poll were nearly as misleading. So was California’s official ballot summary, which failed to mention the virtual ban on bilingual programs. It seems likely that many voters never got that message.

11. E.g., in a 1998 study of San Francisco students who had been redesignated as fluent in English, David Ramírez found that children had spent an average of 4.8 years in language assistance programs (bilingual and otherwise). After being mainstreamed, they usually outscored all other groups at the secondary level, including native English speakers.

12. The author was also an organizer of this effort, known as UnzWatch.

13. Spokespersons hired by Citizens, who had little background in education or knowledge of research, devoted most of their energies to debating Ron Unz and, later, to influencing editorial boards to oppose Proposition 227. But they held few press conferences and generated few news stories of any kind. Meanwhile, Unz faxed daily press releases to journalists — setting the agenda and defining terms of the debate.

14. English for the Children raised $1,289,815 but spent only $976,632, according to its reports to the California Secretary of State. Of this amount, $752,738 came from Ron Unz. Citizens for an Educated America raised and spent $4,754,157, including $2.1 million from the California Teachers Association and $1.5 million from Jerrold Perenchio, owner of the Spanish-language network Univision.

15. A week before the election, the Los Angeles Times Poll reported 62 percent Latino support for Proposition 227. By contrast, its exit interviews with voters on June 2 found that Latinos had opposed the initiative by 63 to 37 percent. African Americans also voted No, 52-48, while Asian Americans voted Yes, 57-43, and whites voted Yes, 67-33, according to the exit poll.

16. Still, because of confusion about Proposition 227, 43 percent were inclined to support it, too. The poll was conducted in February 1998 by the newspaper La Opinión and KVEA-TV, the Los Angeles affiliate of Telemundo.

17. The day after the vote, bilingual education advocates sued to overturn Proposition 227 on civil rights and constitutional grounds. A federal district judge denied their petition for a preliminary injunction in the case, Valeria G. v. Wilson; so the law took effect as scheduled on August 2, 1998. Litigation on the merits continues. Meanwhile, Ron Unz and Alice Callaghan have threatened to bring their own lawsuits against districts whose creative interpretations displease them. For example, some educators have cited Unz’s claim of an “overwhelming” victory with 61 percent of the vote to argue that a curriculum taught 61 percent in English should satisfy the law’s requirement.

“Copyright ? 1999 by James Crawford. All rights reserved. Reposted by permission of James Crawford ([email protected]).

From Bilingual Education: History, Politics, Theory, and Practice, 4th ed. (Los Angeles: Bilingual Educational Services, 1999); 800-448-6032.

Comments are closed.