Bilingual Education: The Controversy
	By Richard Rothstein Our commonly held notion of how earlier generations of immigrants were educated -- often used as the chief argument in support of English immersion -- is a myth, Mr. Rothstein reveals.  Illustration © 1998 by Jonathan Bouw

BILINGUAL education, a preferred strategy for the last 20 years, aims to teach academic subjects to immigrant children in their native languages (most often Spanish), while slowly and simultaneously adding English instruction.1 In theory, the children don't fall behind in other subjects while they are learning English. When they are fluent in English, they can then "transition" to English instruction in academic subjects at the grade level of their peers. Further, the theory goes, teaching immigrants in their native language values their family and community culture and reinforces their sense of self-worth, thus making their academic success more likely.

In contrast, bilingual education's critics tell the following, quite different, story. In the early 20th century, public schools assimilated immigrants to American culture and imparted workplace skills essential for upward mobility. Children were immersed in English instruction and, when forced to "sink or swim," they swam. Today, however, separatist (usually Hispanic) community leaders and their liberal supporters, opposed to assimilation, want Spanish instruction to preserve native culture and traditions. This is especially dangerous because the proximity of Mexico and the possibility of returning home give today's immigrants the option of "keeping a foot in both camps" -- an option not available to previous immigrants who were forced to assimilate. Today's attempts to preserve immigrants' native languages and cultures will not only balkanize the American melting pot but hurt the children upon whom bilingual education is imposed because their failure to learn English well will leave them unprepared for the workplace. Bilingual education supporters may claim that it aims to teach English, but high dropout rates for immigrant children and low rates of transition to full English instruction prove that, even if educators' intentions are genuine, the program is a failure.

The English First Foundation, a lobbying group bent on abolishing bilingual education, states that most Americans "have ancestors who learned English the same way: in classrooms where English was the only language used for all learning activities."2 According to 1996 Republican Presidential nominee Bob Dole, the teaching of English to immigrants is what "we have done . . . since our founding to speed the melting of our melting pot. . . . We must stop the practice of multilingual education as a means of instilling ethnic pride, or as a therapy for low self-esteem, or out of elitist guilt over a culture built on the traditions of the West."3

Speaker of the House Newt Gingrich chimed in as well:

If people had wanted to remain immersed in their old culture, they could have done so without coming to America. . . . Bilingualism keeps people actively tied to their old language and habits and maximizes the cost of the transition to becoming American. . . . The only viable alternative for the American underclass is American civilization. Without English as a common language, there is no such civilization.4

Bilingual Education: The History

Despite proximity to their homeland, Mexican Americans are no more likely to reverse migrate than were Europeans in the early 20th century. One-third of the immigrants who came here between 1908 and 1924 eventually abandoned America and returned home.5

What's more, the immigrants who remained did not succeed in school by learning English. During the last great wave of immigration, from 1880 to 1915, very few Americans succeeded in school, immigrants least of all. By 1930, it was still the case that half of all American 14- to 17-year-olds either didn't make it to high school or dropped out before graduating. The median number of school years completed was 10.

Far from succeeding by immersing themselves in English, immigrant groups did much worse than the native-born, and some immigrant groups did much worse than others. The poorest performers were Italians. According to a 1911 federal immigration commission report, in Boston, Chicago, and New York 80% of native white children in the seventh grade stayed in school another year, but only 58% of Southern Italian children, 62% of Polish children, and 74% of Russian Jewish children did so. Of those who made it to eighth grade, 58% of the native whites went on to high school, but only 23% of the Southern Italians did so. In New York, 54% of native-born eighth-graders made it to ninth grade, but only 34% of foreign-born eighth-graders did so.6

A later study showed that the lack of success of immigrants relative to the native-born continued into high school. In 1931, only 11% of the Italian students who entered high school graduated (compared to an estimated graduation rate of over 40% for all students). This was a much bigger native/immigrant gap than we have today.

While we have no achievement tests from that earlier period by which to evaluate relative student performance, I.Q. tests were administered frequently. Test after test in the 1920s found that Italian immigrant students had an average I.Q. of about 85, compared to an average for native-born students of about 102. The poor academic achievement of these Italian Americans led to high rates of "retardation" -- that is, being held back and not promoted (this was the origin of the pejorative use of the term "retarded").

A survey of New York City's retarded students (liberally defined so that a child had to be 9 years old to be considered retarded in the first grade, 10 years old in the second grade, and so on), found that 19% of native-born students were retarded in 1908, compared to 36% of Italian students. The federal immigration commission found that the retardation rate of children of non-English-speaking immigrants was about 60% higher than that of children of immigrants from English-speaking countries.7 The challenge of educating Italian immigrant children was so severe that New York established its first special education classes to confront it. A 1921 survey disclosed that half of all (what we now call) "learning disabled" special education children in New York schools had Italian-born fathers.8

As these data show -- and as is the case today -- some groups did better than others, both for cultural reasons and because of the influence of other socioeconomic factors on student achievement. If Italian children did worse, Eastern European Jewish children did better. This is not surprising in light of what we now know about the powerful influence of background characteristics on academic success. In 1910, 32% of Southern Italian adult males in American cities were unskilled manual laborers, but only one-half of 1% of Russian Jewish males were unskilled. Thirty-four percent of the Jews were merchants, while only 13% of the Italians were. In New York City, the average annual income of a Russian Jewish head-of-household in 1910 was $813; a Southern Italian head-of-household averaged $688.   9

But even with these relative economic advantages, the notion that Jewish immigrant children assimilated through sink-or-swim English-only education is a nostalgic and dangerous myth. In 1910, there were 191,000 Jewish children in the New York City schools; only 6,000 were in high school, and the overwhelming majority of these students dropped out before graduating. As the Jewish writer Irving Howe put it, after reviewing New York school documents describing the difficulties of "Americanizing" immigrant children from 1910 to 1914, "To read the reports of the school superintendents is to grow impatient with later sentimentalists who would have us suppose that all or most Jewish children burned with zeal for the life of the mind."10 There may have been relatively more such students among the Jewish immigrants than in other immigrant communities, Howe noted, but they were still a minority.

Immersing immigrants in an English-language school program has been effective -- usually by the third generation. On the whole, immigrant children spoke their native language; members of the second generation (immigrants' native-born children) were bilingual, but not sufficiently fluent in English to excel in school; members of the third generation were fluent in English and began to acquire college educations. For some groups (e.g., Greek Americans), the pattern more often took four generations; for others (e.g., Eastern European Jews), many in the second generation may have entered college.

This history is not a mere curiosity, because those who advocate against bilingual education today often claim that we know how to educate immigrant children because we've done it before. However, if we've never successfully educated the first or even second generation of children from peasant or unskilled immigrant families, we are dealing with an unprecedented task, and history can't guide us.

To understand the uniqueness of our current challenge, compare the enormous -- by contemporary standards -- dropout rate of New York City Jewish students in 1910 with that of Mexican students in the Los Angeles school district today. Like New York in 1910, Los Angeles now is burdened with a rising tide of immigrants. In 1996, there were 103,000 Hispanic students in grades 9-12 in Los Angeles (out of the city's total K-12 Hispanic population of 390,000). Hispanic high school students were about 26% of the total Hispanic student population in Los Angeles in 1996,11 compared to 3% for Jews in New York in 1910 (only 6,000 high school students out of 191,000 total Jewish enrollment). In Los Angeles today, 74% of Mexican-born youths between the ages of 15 and 17 are still in high school; 88% of Hispanic youths from other countries are still in attendance.12 More than 70% of Hispanic immigrants who came to the United States prior to their sophomore year actually complete high school (compared to a 94% high school completion rate for whites and a 92% rate for blacks).13 English immersion programs for Jews early in this century (and certainly similar programs for Italians) cannot teach us anything that would help improve on today's immigrant achievement or school completion, much of which may be attributable to bilingual education programs, even if imperfectly administered.

If the notion is misleading that English immersion led previous generations of immigrants to academic success, so too is the claim that bilingual education repudiates the assimilationist approach of previous immigrants. In reality, today's Hispanics are not the first to seek bicultural assimilation. Some 19th- and early 20th-century European immigrants also fought for and won the right to bilingual education in the public schools.14 Native-language instruction was absent from 1920 until the mid-1960s only because a fierce anti-German (and then anti-immigrant) reaction after World War I succeeded in banishing it from American classrooms. Even foreign-language instruction for native-born students was banned in most places. If Chicago's Bismarck Hotel found it necessary to rename itself the "Mark Twain," it should not be surprising that bilingual education programs were also abolished.

Before World War I, immigrant groups often pressed public schools to teach children in their native language. The success of these groups depended more on whether adult immigrant activists had political power than on a pedagogical consensus. The immigrants' objective, as it is today, was to preserve a fragment of ethnic identity in children for whom the pull of American culture seemed dangerously irresistible. In this, they were supported by many influential educators. William Harris, the school superintendent in St. Louis and later U.S. commissioner of education, argued for bilingual education in the 1870s, stating that "national memories and aspirations, family traditions, customs and habits, moral and religious observances cannot be suddenly removed or changed without disastrously weakening the personality." Harris established the first "kindergarten" in America, taught solely in German, to give immigrant students a head start in the St. Louis schools.15

Nineteenth-century immigrant parents were often split over the desirability of bilingual education, as immigrant parents are split today. Many recognized that children were more likely to succeed if schools' use of the native language validated the culture of the home. But others felt that their children's education would be furthered if they learned in English only.

The first bilingual public school in New York City was established in 1837 to prepare German-speaking children for eventual participation in regular English schools. The initial rule was that children could remain in German-language instruction only for 12 months, after which they would transfer to a regular school. But the German teacher resisted this rule, believing that, before transferring, the children needed more than the limited English fluency they had acquired after a year of German instruction. The record is unclear about how often the rule was stretched.

Many immigrant children, not just Germans, did not attend school at all if they could not have classes in their native language. In his 1840 address to the New York legislature, Gov. William Seward (later Lincoln's secretary of state) explained that the importance of attracting immigrants to school -- and of keeping them there -- motivated his advocacy of expanded native-language instruction: "I do not hesitate to recommend the establishment of schools in which [immigrant children] may be instructed by teachers speaking the same language with themselves." Only by so doing, Gov. Seward insisted, could we "qualify . . . [them] for the high responsibilities of citizenship."

Buoyed by Seward's endorsement, Italian parents in New York City demanded a native-language school as well, and in 1843 the Public School Society established a committee to determine whether one should be established. The committee recommended against an Italian-language school, claiming the Italian community was itself divided. "Information has been obtained," the committee stated, "that the more intelligent class of Italians do not desire such a school, and that, like most [but not, apparently, all] of the better class of Germans, they would prefer that those of their countrymen who come here with good intentions should be Americanized as speedily as possible."16

Bilingual education, though sometimes controversial, was found nationwide. In Pennsylvania, German Lutheran churches established parochial schools when public schools would not teach in German; in 1838, Pennsylvania law converted these German schools to public schools. Then, in 1852, a state public school regulation specified that "if any considerable number of Germans desire to have their children instructed in their own language, their wishes should be gratified."17

In 1866, succumbing to pressure from politically powerful German immigrants, the Chicago Board of Education decided to establish a German-language school in each area of the city where 150 parents asked for it. By 1892 the board had hired 242 German-language teachers to teach 35,000 German-speaking children, one-fourth of Chicago's total public school enrollment. In 1870, a public school established in Denver, Colorado, was taught entirely in German. An 1872 Oregon law permitted German-language public schools to be established in Portland whenever 100 voters petitioned for such a school. Maryland, Iowa, Indiana, Kentucky, Ohio, and Minnesota also had bilingual education laws, either statewide or applying only to cities with large immigrant populations. In Nebraska, enabling legislation for bilingual education was enacted for the benefit of German immigrant children as late as 1913.18

There was considerable variation in how these programs arranged what we now call the "transition" to English. In St. Louis, Harris' system introduced English gradually, beginning in the first grade. The 1888 report of the Missouri supervisor of public instruction stated that "in some districts the schools are taught in German for a certain number of months and then in English, while in others German is used part of the day and English the rest. Some of the teachers are barely able to speak the English language." Ohio's 1870 rules provided that the lower grades in German-language public schools should be bilingual (half the instructional time in grades 1 through 4 could be in German), but in grades 5 through 8 native-language instruction had to be reduced to one hour a day. Baltimore permitted public schools in the upper grades to teach art and music in German only, but geography, history, and science had to be taught in both English and German. In some midwestern communities, there was resistance to any English instruction: an 1846 Wisconsin law insisted that public schools in Milwaukee must at least teach English (as a foreign language) as one academic subject.19

While Germans were most effective in demanding public support for native-language instruction, others were also successful. In Texas in the late 19th century, there were seven Czech-language schools supported by the state school fund. In California, a desire by the majority to segregate Chinese children seemed to play more of a role than demands by the Chinese community for separate education. San Francisco established a Chinese-language school in 1885; the city later established segregated Indian, Mongolian, and Japanese schools.20

San Francisco's German, Italian, and French immigrants, on the other hand, were taught in their native languages in regular public schools. Here, bilingual education was a strategy designed to lure immigrant children into public schools from parochial schools where they learned no English at all. According to San Francisco's school superintendent in 1871, only if offered native-language instruction could immigrant children be brought into public schools, where, "under the care of American teachers," they could be "molded in the true form of American citizenship."21

Support for bilingual education was rarely unanimous or consistent. In San Francisco, the election of an "anti-immigrant" Republican school board majority in 1873 led to the abolition of schools in which French and German had been the primary languages of instruction and to the firing of all French- and German-speaking teachers. After protests by the immigrant community, bilingual schools were reestablished in 1874. In 1877, the California legislature enacted a prohibition of bilingual education, but the governor declined to sign it. William Harris' bilingual system in St. Louis was dismantled in 1888, after redistricting split the German vote and the Irish won a school board majority.22

In 1889, Republican Gov. William Hoard of Wisconsin sponsored legislation to ban primary-language instruction in public and private schools, claiming the support of German immigrant parents. The Milwaukee Sentinel published a front-page story about "a German in Sheboygan County . . . who sent his children away to school in order that they might learn English." The father, reported the Sentinel, complained that "in the public schools of the town, German teachers, who . . . did not know English . . . had been employed . . . , [and] he felt it essential to the welfare of his children, who expected to remain citizens of this country, to know English."23

But both the newspaper and Wisconsin's Republican politicians had misjudged the immigrants' sentiments. In response to the anti-bilingual law, enraged German Americans (who had previously supported Republican candidates) mobilized to turn the statehouse over to Democrats and to convert the state's 7-to-2 Republican majority in Congress to a Democratic majority of 8-to-1. The Democrats promptly repealed the anti-bilingual education law.

An almost identical series of events took place in Illinois, where formerly Republican German American voters mobilized in both East St. Louis and Chicago to elect a liberal Democrat, Peter Altgeld, governor in 1890, largely because of his bilingual school language policy. These upheavals in two previously safe Republican states played an important role in the election of Democrat Grover Cleveland as President in 1892. Nonetheless, the controversy continued, and in 1893 the Chicago Tribune began a new campaign against German-language instruction. In a compromise later that year, German instruction was abolished in the primary grades but retained in the upper grades, while Chicago's mayor promised German Americans a veto over future school board appointments to ensure that erosion of primary-language instruction would not continue.24

But these controversies ended with World War I. Six months after the armistice, the Ohio legislature, spurred by Gov. James Cox, who was to be the Democratic Presidential candidate in 1920, banned all German from the state's elementary schools. The language posed "a distinct menace to Americanism," Cox insisted. The New York Times editorialized in 1919 that, although some parents "want German to be taught [because it] pleases their pride . . . , it does not do their children any good." Within the following year, 15 states in which native-language instruction had flourished adopted laws requiring that all teaching be in English. By 1923, 35 states had done so.25 Only when Nebraska went so far as to ban native-language instruction in parochial as well as public schools did the Supreme Court, in 1923, strike down an English-only law.26

During the next 30 years, bilingual instruction had its ups and downs, even where English was not the native language. In 1950, Louisiana first required English, not French, to be the language of public school instruction. In the Southwest, where teaching in Spanish had long been common, the practice continued in some places and was abolished in others. Tucson established a bilingual teaching program in 1923, and Burbank established one in 1931. New Mexico operated bilingual schools throughout most of the 20th century, up until the 1950s. The state even required the teaching of Spanish to English-speaking children in elementary school. But in 1918, Texas made teaching in Spanish a crime, and, while the law was not consistently enforced (especially along the Mexican border), as recently as 1973 a Texas teacher was indicted for not teaching history in English.27 In the same year, Texas reversed itself and adopted bilingual education as its strategy.

When bilingual education began to reemerge in the 1970s -- spurred by a Supreme Court finding that schools without special provisions for educating language-minority children were not providing equal education -- the nation's memory of these precedents had been erased. Today many Americans blithely repeat the myth that, until the recent emergence of separatist minority activists and their liberal supporters, the nation had always immersed its immigrant children in nothing but English and this method had proved its effectiveness.
 

Bilingual Education: Mixed Evidence

This mixed history, however, does not prove that bilingual education is effective, any more so than English immersion or intense English-language instruction. To an unbiased layperson, the arguments of both advocates and opponents of bilingual education seem to make sense. On the one hand, it's reasonable to insist that children who don't speak English continue their education in a language they understand in history, literature, math, and science, while they learn English. It's also reasonable to expect, however, that this might make it too tempting to defer English-language instruction. Moreover, the best way to do something difficult -- e.g., making the transition to English -- is simply to do it without delay. It makes sense to acknowledge that children may adapt better to school if the school's culture is not in conflict with that of the home. But some immigrant parents may be more intent on preserving native culture for their children than are the children themselves.

Modern research findings on bilingual education are mixed. As with all educational research, it is so difficult to control for complex background factors that affect academic outcomes that no single study is ultimately satisfying. Bilingual education advocates point to case studies of primary-language programs in Calexico, California; Rock Point, Arizona; Santa Fe, New Mexico; New Haven, Connecticut; and elsewhere that show that children advance further in both English and other academic subjects when native-language instruction is used and the transition to English is very gradual. Opponents point to case studies in Redwood City and Berkeley, California; in Fairfax, Virginia; and elsewhere that prove that immersion in English or rapid and intensive English instruction is most effective.28 Overall, the conflicting evidence from these case studies does not suggest that abolition of bilingual education or even the substitution of parental choice for pedagogical expertise in determining whether bilingual approaches should be used would improve things much.

The problem is especially complex because not only economic factors but also generational variation apparently affects the achievement of immigrant youths. In 1936, the principal of a high school in New York City that enrolled large numbers of Italian immigrants wrote:

The problem of juvenile delinquency . . . baffles all the forces of organized society. . . . The highest rate of delinquency is characteristic of immigrant communities. . . . The delinquent is usually the American-born child of foreign-born parents, not the immigrant himself. Delinquency, then, is fundamentally a second-generation problem. This intensifies the responsibility of the school.29

In the absence of persuasive evidence, the issue has become politicized. In a country as large as ours, with as varied experience, there is virtually no limit to the anecdotes and symbols that can be invoked as substitutes for evidence.

Opponents of bilingual education promote Hispanic parents to the media when they claim they want their children to learn English without bilingual support; the clear implication is that only liberal ideologues and separatists support native-language instruction. These claims, like those circulated by the Milwaukee Sentinel a century ago, may not reflect the feelings of most parents. And the technology of teaching a new language to immigrant children is complex; both bilingual education advocates and opponents claim their goal is full English literacy as rapidly as possible. But there's no reason to expect that politicized parent groups are the best judges of language acquisition research.

There are also successful adult immigrants who brag of their English fluency, acquired either with or without bilingual education. As always, such anecdotal evidence should be treated with caution. Richard Rodriguez' autobiography, Hunger of Memory, describes his successful education in an English-only environment. But Rodriguez, unlike most immigrants, was raised in a predominantly English-speaking neighborhood and was the only Spanish speaker in his class.31 His experience may be relevant for some immigrants, but not relevant for many others.

Whichever method is, in fact, more effective for most immigrant children, there will be many for whom the other method worked well. It may be the case that immigrant children's social and economic background characteristics should affect the pedagogy chosen. Even if some Russian Jewish immigrants did not require bilingual education to graduate from high school, perhaps Italians would have progressed more rapidly if they'd had access to bilingual instruction. Today, the fact that some (though not all) Asian immigrants seem to progress rapidly in school without native-language support provides no relevant evidence about whether this model can work well for Mexican or Caribbean children, especially those low on the ladder of socioeconomic status and those whose parents have little education. Nor does it tell us much about what the best pedagogy would be for Asians who generally do less well in school, such as Hmong, Laotian, and Cambodian children.32

It is certain, however, that the American "melting pot" has never been endangered by pluralist efforts to preserve native languages and cultures. Bilingual instruction has never interfered with the powerful assimilationist influences that overwhelm all children whose parents migrate here. And this is equally true of Spanish-speaking children today.

After the last 20 years of bilingual education throughout America, Spanish-speaking children continue to assimilate. From 1972 to 1995, despite rapidly accelerating immigration (more Hispanic youths are first-generation immigrants today than 20 years ago), the Hispanic high school completion rate has crept upward (from 66% to 70%). Hispanic high school graduates who enroll in college jumped from 45% to 54% (for non-Hispanic whites, it's now 64%). And the number of Hispanic high school graduates who subsequently complete four years of college jumped from 11% to 16% (for non-Hispanic whites, it's now 34%).33 A study of the five-county area surrounding Los Angeles, the most immigrant-affected community in the nation, found that from 1980 to 1990, the share of U.S.-born Hispanics in professional occupations grew from 7% to 9%, the share in executive positions grew from 7% to 10%, and the share in other administrative and technical jobs grew from 24% to 26%.   34 Overall, 55% of U.S.-born Hispanics are in occupations for which a good education is a necessity, in an area where bilingual education has been practiced for the last generation.

Perhaps we can do better. Perhaps we would do better with less bilingual education. But perhaps not. All we can say for sure is that the data reveal no apparent crisis, and the system for immigrant education with which we've been muddling through, with all its problems, does not seem to be in a state of collapse.

The best thing that could happen to the bilingual education debate would be to remove it from the political realm. Sound-bite pedagogy is no cure for the complex interaction of social, economic, and instructional factors that determine the outcomes of contemporary American schools.