English immersion students in Brockton:
Seyla Nou and Elisia Heak

Eunji Gloria Cho Mantzouranis—Ms. Cho to her students —uses a green marker to put on the whiteboard what should be a simple math problem for fifth- and sixth-graders: Find other ways to express 8 x 7. You could, for example, say (4 x 7) + (4 x 7), which might lead you to 4 x 14 or 2 x 28. The beauty of factoring is that the different combinations all yield the same result.

Numbers may be a universal language, but in Cho’s “sheltered English immersion” classroom, that cliché does not ring true. Some students understand the assignment, but others are lost. Cho crosses her index fingers, forming an “X” to remind students this is multiplication. She tries to clarify factor “pairs” by gesturing to her black platform heels.

“How many pieces of shoes do you need to make a pair?” asks Cho, a native of Korea with long, jet-black hair, whose darting among students to provide help leaves her nearly breathless. She glances at her aide, “Can you translate?”

Yes, the aide can translate—but only into Portuguese. In a classroom in which students speak six different languages, that leaves many stranded, including the new boy from Greece named Jason. He set foot in the Louis Angelo School in Brockton for the first time today. He’s sitting beside the aide, but all he can do is lift his paper and point.

Downstairs from Cho, in second-grade teacher Silvana Resendes’s class of 15, all but three speak Portuguese. Resendes, herself a bilingual success story who speaks Portuguese and English with no accent, is teaching a lesson on counting money. She easily clarifies “one cent” with a quick “un centavo.”

Cho, on the other hand, has 24 students who are at two grade levels (fifth and sixth), at three reading levels (all below the fifth grade), and with backgrounds in six languages (none of which she speaks). Some are progressing quickly. Cho singles out a Portuguese-speaking boy who is “willing to learn English as fast as he can.” But she takes note of another child who “is not opening his mouth after months of being here. He doesn’t understand what is going on.” And there’s only so much she can do to get through, she says.

“I use a lot of gestures and facial expressions,” says Cho, “but there is a limit.”

More than three years after nearly 70 percent of Massachusetts voters approved Question 2, which did away with 31 years of bilingual education in favor of California businessman Ron Unz’s “English for the Children” proposal, confusion reigns.

The confusion is not over the broad mandate, which is refreshingly clear: Teach kids English by teaching them in English. Rather, the challenge is in making it happen, given a growing non-English-speaking population, high-pressure accountability, and the same old six-hour school day.

The simple demand that kids learn English—and quickly—is complicated by real-life factors, including a dearth of qualified teachers, uncertainty over who should be classified as an “English Language Learner,” and tests that, in many cases, cannot differentiate between a kid who doesn’t understand the question and a kid who simply doesn’t know the answer.

English immersion students in Brockton:
Gabriel Teixeira and Aliane Farou,
with teacher Silvana Resendes

It doesn’t help that some non-English-speaking students arrive with no previous schooling of any kind, or with no understanding that one must sit down during class and not run full-out down to the cafeteria when the lunch bell rings. Then there is the uncertainty among some teachers about what they are—and aren’t—allowed to do under the new law. Is pointing at the soles of your shoes a good way to explain “pairs,” or it is an act of desperation?

“People still don’t know what sheltered English immersion is,” says Kathryn Riley, who heads the state Department of Education’s Office of Language Acquisition, referring to the teaching model mandated by the 2002 law. “People can’t run the video in their mind about the type of classroom [they] are trying to create. If you can’t envision the classroom, it is very hard to make.”


Few educational issues attract as much passion as the debate over teaching students classified as limited English proficient (LEP) or English language learners (ELL), terms that are used interchangeably.

That’s partly because the number of LEP students has risen nationwide following a historic burst of immigration in the 1990s. According to a 2005 Urban Institute report, American public school pupils whose parents are immigrants rose from 6 percent in 1970 to nearly 20 percent of students today. A few learn English quickly, but many more —along with some whose parents were born here but don’t speak English—comprise the LEP population.

Most live in a few states (Arizona, California, New Mexico, and Texas), but nearly all states are finding themselves with more LEP students. In Vermont, for example, the LEP school population rose 185 percent from 1990 to 2000. In Massachusetts, 5 percent of public school students, or nearly 50,000, are English language learners, as of March 2005. They speak 112 different languages, but for nearly 55 percent, Spanish is their first language.

Nationally, such students are concentrated in urban areas, and within those areas, in particular schools. More than half of these students attend schools in which more than 30 percent of their classmates are also LEP, according to the Urban Institute. Half of LEP children have parents with less than a high school education; 51 percent of children whose parents are immigrants live in low-income households.

The case is much the same here. In Lawrence, for example, where Census Bureau figures show a per-capita income of $13,360 (about half the state average), Superintendent Wilfredo Laboy says 90 percent of public school children speak Spanish at home and nearly a quarter are LEP.

Eunji Cho teaches children who speak
six different languages.

While Question 2 was ostensibly about educational policy, it triggered an emotionally charged debate (see “Lost in the Translation,” CW, Education Reform Extra 2002). The English for the Children campaign seemed to tap into a “you’re in America, so speak English” attitude among voters statewide, most of whom had little contact with immigrant communities. On the other side, Question 2 opponents included those who believed transitional bilingual education, even if poorly executed, represented a basic civil right, but they campaigned under the puzzling slogan “Don’t Sue Teachers,” focusing on an obscure enforcement provision in the ballot question.

There were also advocates for language minorities who saw that, politically correct and culturally sensitive as it was, transitional bilingual education wasn’t working. Laboy, for one, called the old bilingual approach “educational apartheid” for its segregation of language-minority children from their English-speaking peers.

Similarly heated debates played out in California, Arizona, and Colorado, with only Colorado rejecting the English-only approach. Jack Jennings, president of the Center on Educational Policy in Washington, DC, says these battles have pushed mainstream practice toward English immersion in recent years. An August 2005 report by the center notes that LEP students receiving all regular-curriculum instruction in English rose from 19 percent in 1993 to nearly 25 percent in 2003.

Jennings says English immersion is also growing for pragmatic reasons. As communities—including small towns —experience influxes of non-English-speaking students from far-flung places, instruction in English is the default approach for teachers who don’t speak 20 or 30 different languages.


In Massachusetts, the text of Question 2 was pointed, asserting that “the public schools of Massachusetts have done an inadequate job of educating many immigrant children.” The solution? “All children in Massachusetts public schools shall be taught English by being taught in English and all children shall be placed in English language classrooms.”

This language—now part of the Massachusetts General Laws, Chapter 71A, Section 4—states that children “who are English learners shall be educated through sheltered English immersion during a temporary transition period not normally intended to exceed one school year.”

In some classrooms, it’s hard to see much difference under the new law: ‘I’ve never had a history book to use in their native language.’

The turnabout in educational approach could not have been more extreme. Under the old law, which dated from 1971, local school districts with 20 or more students who spoke the same non-English language were required to “establish, for each classification, a program in transitional bilingual education for the children therein.” These children were mandated to receive “a full-time program of instruction,” which included “all those courses or subjects which a child is required by law to receive,” taught “in the native language of the children” and “also in English.” They were to be instructed “in reading and writing of the native language” and “in the history and culture of the country, territory or geographic area which is the native land of the parents of the children of limited English-speaking ability.”

While the program was called “transitional” and normally provided up to three years of instruction before students moved into mainstream classrooms, the law also allowed a child to “continue in that program for a period longer than three years,” and many did. Critics echoed a 2001 report by the Lexington Institute, an Arlington, Va.–based conservative think tank, charging that bilingual education in the US was transitional “in name alone.”

On its face, Question 2 brought about a reversal in the way LEP children are taught. Instead of learning subjects like math, science, and social studies in their native tongue while studying English separately, students now study all subjects in English. In theory, at least, children are supposed to spend one year studying English intensively so that they can move into mainstream classrooms.

But “supposed to” is key. “We don’t have enough [English as a Second Language] teachers in the state to do sheltered English immersion,” says Riley, the DOE official in charge of language acquisition. “It is a real puzzle as to how to implement this in a way that follows the law and gives children access to learning English and learning the curriculum.”

Many students ‘were staying six to eight years in separate classes without being integrated and without learning English.’

Critics of the old bilingual education system say that problem is nothing new. Christine Rossell, professor of political science at Boston University and former co-chair of the Massachusetts campaign to pass Question 2, notes that only 23 percent of LEP students were getting true bilingual education. All Question 2 did, she says, was end “all this lying and cheating.” In most districts, she says, it was impossible to provide native-language instruction for all LEP students.

“The Chinese kids never were getting bilingual education,” says Rossell. “And now they can call it what it was: sheltered English immersion.”

Principal Suzanne Lee faces tougher goals
and new pupils each year.

Indeed, in some formerly bilingual-education classrooms, it’s hard to see much difference under sheltered immersion. At Dearborn Middle School in Roxbury, for example, 25-year veteran bilingual education teacher Maria Leite says that not all that much has changed in how she instructs her students, who speak Portuguese Creole. After finishing a lesson on forms of government—monarchy, dictatorship, democracy—she notes, “I’ve never had a history book to use in their native language.”

According to Riley, the “sheltered” part of sheltered English immersion refers to teaching strategies that include language goals, time for students to practice speaking, and techniques to help students pick up English from the context of the lesson. For example, Leite discusses the roots of words like “dictatorship” as part of her lesson. One learning objective spelled out on the whiteboard focuses on language: “Students will be able to write sentences with new vocabulary.” She sets aside time in class for students to discuss in groups and write down the different forms of government.

Similarly, Resendes’s second-grade math lesson is about counting money, but it’s just as much about language. She has children copy “Ways to make 15 cents” on their papers, including columns for “dimes,” “nickels,” and “pennies.” She takes note of spelling, telling one child, “Look at your word ‘dimes’ and look at my word ‘dimes.’” He cranes his neck, studies, and then works his eraser.

The lesson is in English, but peppered with Portuguese. Leite, too, uses Creole to clarify and emphasize directions. She says the biggest difference from the past is the amount of English she uses in place of the children’s native language.

In classrooms like Cho’s, in which students speak a variety of languages, sheltering techniques—such as developing word lists for students to draw on, talking aloud with students to help them generate useful words before writing, incorporating visual symbols into instructions (such as a picture of a notebook as a reminder to “get out your notebook”), and repetition of phrases—are more critical.

Still, many teachers are not trained to teach this way. And even when they are, some students find themselves in classes without a clue about what’s being taught or how to begin learning it. In the past, of course, you had a different problem: Students lingered for years in classrooms stewarded by well-meaning but underqualified native language teachers with no incentive to make sure kids learned English.

The new law tries to be—at least on the surface—everything the old law wasn’t. But reality is muddier, beginning with the law itself. “The definition of sheltered English immersion is a very small part of the law,” says Riley. That’s why, she says, “implementation has been so difficult. It’s not specific.” She says some schools “do a terrific job and some do things they would not do with their own children.”

And, much as critics of bilingual ed charged under the old system, poor programs generate little parental outrage. “The parents of these children do not speak English and do not know what they should expect,” says Riley. “Districts do what they wish.” While Riley says the state requires beginning- and intermediate-level LEP students to receive at least 2.5 hours of English as a Second Language instruction each day, she suspects many are not getting even that.

“The abuses of the past were that kids were left in bilingual education too long,” says Jennings. “The abuses of the present are that kids are not being adequately taught English and they are sitting in English classes not knowing what is going on. It would be nice to find some middle ground.”


One true difference between the old era of bilingual education and the new era of English immersion is not related to language learning at all, but to the changed educational environment of MCAS and No Child Left Behind. “Accountability” is the buzzword. Translated, that means data. One pointed criticism of the old bilingual education was the lack of information about how well (or if) kids were learning English or other subjects, and whether they were really “transitioning” into English-language classrooms.

This was a problem not just in Massachusetts. The Urban Institute reports that 18 percent of LEP children in pre-kindergarten through the fifth grade, and 29 percent in the sixth through 12th grades, are American-born children whose parents were also born in this country, suggesting that the parents themselves never learned English adequately. More than half of LEP children in secondary schools are US natives who have not mastered English even after attending American schools for seven or eight years.

Wilfredo Laboy called the old approach
“educational apartheid.”

This situation horrifies at least one early champion of bilingual education. Charles Glenn was director of what was then the Bureau of Equal Educational Opportunity at the state DOE when he pushed for passage of Massachusetts’s bilingual education law, the first in the nation. “I was pressing for bilingual education as part of a general strategy to improve equity,” says Glenn. He saw the measure as a way to ensure the rights of non-English-speaking students to receive the same education as their English-speaking peers. But once the law was enacted, he says, the state created a separate bilingual education office staffed by language educators—not individuals focused on educational equity.

Glenn—who sent his own five children to the two-way bilingual education program at the Rafael Hernandez School in Roxbury, a model exempted from the English immersion requirement by the Legislature soon after Question 2 was approved—insists that he is not a foe of bilingual education, when it’s done right. But in bilingual education as it was actually practiced, Glenn says, he saw good intentions run amok.

“I can’t tell you how many meetings I had with my colleagues in the bilingual office, asking them to do what the law required: To adopt a test to be given to kids in the bilingual program to ensure they learned English,” recalls Glenn, now a professor of educational policy at Boston University. “This was met with, ‘That would be hard on the self-esteem of minority-language students.’”

In the meantime, LEP students were exempted from basic skills tests for years after entering school. As a result, there was little pressure to make sure they learned English—or anything else, for that matter.

In the 1980s, federal District Court Judge W. Arthur Garrity, overseeing Boston school desegregation orders, asked why pupils remained so long in bilingual education. When Glenn analyzed the careers of bilingual-education pupils, he says, he found that “hundreds and hundreds of them were staying six to eight years in separate classes without being integrated and without learning English.”

He remembers a visit he made to a bilingual-ed classroom in the basement of a Lynn school. The teacher was “a very nice Cambodian gentleman whose training was as an engineer,” Glenn says. The teacher’s English was very limited and he had been given “no direction about what to teach or how to teach.” The man seemed to care a lot about the kids. But, recalls Glenn, “He was doing make-work things.”

Over time, bilingual education ossified further, becoming almost a parallel educational establishment, with its own institutional interests. Jennings, of the Center for Educational Policy, blames bilingual educators of the past for not focusing enough on student progress, instead turning “the movement more into a job protection program” than an approach to teaching children English.

By the 1990s, time was running out. When Glenn, who left DOE in 1991, was appointed to Gov. William Weld’s commission on bilingual education, he said he had no choice but to report that “there wasn’t any data to tell whether bilingual education had been a success or a failure.” In 1997, Weld proposed allowing the state to take over any school district that failed to move students out of bilingual education after three years. The idea died in the Legislature’s Education Committee, but it served as a pre-Question 2 shot across the bow.


There is no longer a dearth of data. Credit, in part, Question 2, which required that English language learners in the second grade and above take a “nationally normed written test of academic matter given in English” and a “nationally normed test of English proficiency.” LEP students now take state English speaking-and-listening tests at least once a year and an English reading-and-writing skills test in the fall and spring. They also now take MCAS.

This sea change, however, is not the result of Question 2 alone. The passage of No Child Left Behind in 2002 heightened accountability of schools to federal standards, raising the bar on work begun in Massachusetts under the Education Reform Act of 1993.

Teacher Maria Leite says she just uses
a little less Creole in her classes

When it comes to accountability for educating English language learners, however, two sets of legal roots apply. One is education law, which aims to include LEP students in all aspects of accountability. The second is civil rights law, which mandates access to services. This second strand was asserted in the landmark 1974 Supreme Court decision Lau v. Nichols. The court ruled that the San Francisco school system’s “failure to provide English language instruction to approximately 1,800 students of Chinese ancestry who do not speak English” effectively “denies them a meaningful opportunity to participate in the public educational program” in violation of the Civil Rights Act of 1964, which barred discrimination “on the grounds of race, color, or national origin.” The important part of this law as it stands today is that an LEP child’s services can’t be limited to one year of instruction in sheltered English immersion. So even though Question 2 pushes for the one-year transition, it can’t legally require it to happen.

There is, however, more to scrutinizing education than looking at compliance. Schools may provide services, but that doesn’t mean they are the right services or that children are learning. NCLB aims to address this gap by requiring that teachers be “certified as English language proficient” and that all states “set standards and benchmarks for raising the level of English proficiency.” NCLB demands that states hold schools accountable for making “adequate yearly progress” toward the federal goal of having all students score “proficient” on state standardized tests by 2014.

All these provisions also require the generation of data, where once there was none. The problem is that the numbers aren’t very encouraging.

Statewide, 90 percent of LEP students were scored as “needs improvement” or “failing” on the 2005 MCAS for 10th-grade English, as opposed to “proficient.” In 10th- grade math, 76 percent were in the bottom two categories. In the fourth grade, 86 percent of LEP students were scored as “needs improvement” or “warning” in English, compared with only 39 percent of other students. In math, 86 percent of LEP students and 53 percent of other students earned scores below “proficient.” Similar results play out across subjects and grades.

Most telling, perhaps, is that LEP students made state goals for improvement in English and math MCAS scores—also known as Annual Yearly Progress—in only 34 percent of schools in English, and only 38 percent of schools in math.

MCAS scores make plain the gap between regular and LEP students. But, asks Rossell of BU, isn’t that the point? In her September 2005 study, Making Uneven Strides, which looks at seven states’ efforts to get LEP students to proficiency under NCLB, Rossell notes that “limited English proficient” means, by definition, student who are not proficient enough in English to score “proficient” or “advanced.” If they were, they wouldn’t be LEP students.

This doesn’t excuse students from learning English or from being tested on their progress toward meeting state subject matter goals. But Rossell argues that the state’s expectations under NCLB are so unrealistic as to be useless. Instead, she advocates exempting LEP students from meeting proficiency standards for five years and, instead, she says, “progress should be assessed for individual students.”

The state is just now beginning to do some of that, for the first time charting the scores of individual students on twice-yearly English proficiency tests. Unfortunately, the results are no advertisement for sheltered English immersion, and scarcely better than the critics’ worst imaginings about the old bilingual education system. Posted without fanfare (no press release or press conference) on the DOE Web site in late December were results from the Massachusetts English Proficiency Assessment in the 53 school districts statewide with large concentrations of English language learners. These results showed that 57 percent of LEP students in third through 12th grades made at least two steps of progress on a seven-step proficiency scale between fall 2004 and spring 2005, exceeding the modest state goal of 50 percent. Among LEP students with three or more years in US schools, 48 percent reached the highest level (“transitioning”), more than meeting the state goal of 40 percent.

Of course, those results mean that 43 percent of LEP students spent last school year in classes in which they made only one step of progress, no progress, or, as Riley observed, “they may have regressed.” And 52 percent of LEP students were not ready to join mainstream classrooms even after three or more years of bilingual and sheltered English education.

The news is even grimmer in many individual districts (see table). In Lawrence, for instance, just 46 percent of students made two steps of progress last year, and only 26 percent had reached “transitioning” after three or more years in US schools (though Lawrence did meet its NCLB Adequate Yearly Progress goal in English for LEP students). In Holyoke, 44 percent made two steps of progress, while 30 percent reached “transitioning” after three or more years. At 48 percent, Boston nearly reached the state benchmark of half of LEP students making two steps of progress, but only 34 percent of students made it to the “transitioning” level. While New Bedford met state standards in progress (53 percent moving up at least two steps), it fell just short of the three-year “transitioning” goal, with 39 percent.

Of the 15 districts with the largest numbers of English language learners, Lowell and Quincy reported the best results, exceeding state goals and statewide averages in both progress and attainment. (Some other districts, including Springfield, Lynn, Brockton, Framingham, Fitchburg, Chelsea, and Somerville, met state targets for progress and attainment after three years, but missed the lower “transitioning” goals of 10 percent for first-year LEP students and 25 percent for second-year students.) But even in these higher-performing districts, at least 30 percent of LEP students failed to make two steps of progress, and a third or more (32 percent in Quincy; 49 percent in Lowell) were still not ready for mainstream classrooms after three years or more.


On a late October morning, Suzanne Lee, principal of the Josiah Quincy School in Boston’s Chinatown, thrusts her hands deep into the pockets of a green parka as a cold wind cuts across the concrete playdecks atop the school. She hardly notices a soccer ball skitter across her path. It is recess for the children, but Lee’s not having fun. In fact, her body language is transparent: She’s ticked off.

The South End News has published the school’s MCAS scores along with the state’s assertion that her students—80 percent of whom are low-income and 67 percent of whom do not speak English as a first language—had barely missed the NCLB-required Adequate Yearly Progress target for improvement in English.

What bothers Lee—besides the fact that parents are upset—is that her school has never before missed Adequate Yearly Progress goals. Despite the palpable economic and language disadvantages her students face, they consistently perform near the state average. On the 2005 English MCAS, for example, 15 percent of fourth-graders earned “advanced” and 34 percent “proficient” scores. The state average was 10 percent “advanced” and 40 percent “proficient.”

The pressure on Lee is clear. And yet, with new non-English speaking students arriving in the school each year, how will she ever get all students to “proficient?”

Question 2 may have put a halt to kids being warehoused in bilingual education programs that failed to teach them English. But it left behind a new question: How fast can a kid who doesn’t speak English get in synch with English-speaking peers?

In overcoming the language barrier, Quincy School students suffer from the same deficit as children from other close-knit immigrant communities, and one that schools alone have a tough time compensating for. Third-grade sheltered English immersion teacher Lai Lai Sheung puts it simply: “Exposure to the language.”

“If you live in Chinatown, you don’t need to speak English to survive,” says Sheung. “You wake up, you hear the news in Chinese, you go to school for six hours, you go home and speak Chinese.” In that respect, she says, things are getting harder, not easier. “Before, children would go home and watch Sesame Street in English. Now, with cable and VCR, it’s all in Chinese.”

There might be no better argument for English immersion than this phenomenon. And Sheung, an immigrant from Hong Kong who spends summers doing volunteer lab work for Earthwatch Institute and sends students home with English-language story tapes, likes the new approach. She knows her kids must learn English.

But she also knows that there is not enough time in the school day to get the results she is expected to get. When third-graders take the MCAS reading test in the spring, Sheung estimates that only half of her 24 pupils will score “proficient.”

Several miles away, but still in Boston, is the Paul A. Dever School. There, the 2005 MCAS shows 75 percent of fourth-grade students scoring “Needs Improvement” or “Warning” in math while 72 percent earned similar scores in English. In the third grade, 80 percent failed to earn “proficient.” The school did make Annual Yearly Progress goals in math in 2004 and 2005—the only times since 1999. It has never met these goals in English.

Like the Josiah Quincy School, the Dever has a student body that is poor—nearly 90 percent come from low- income households. Some 36 percent of the students are LEP, and nearly half speak English as a second language. Most are Hispanic.

In a second-grade classroom, the benefits, and challenges, of the new law are apparent as teacher Christine Cronin calls a reading group to the back table and passes out a slim book about things people like to do “alone” and “together.” After children offer examples from their own lives, she asks them to “read in your brains”—that is, silently.

A boy named Luis, who is wearing a Hawaiian shirt, inadvertently demonstrates why it is more difficult for a child from a different language background to learn English than for those born into English-speaking households. As he moves his fingers along the text, some words make no sense to him; he can’t even guess at them. One that stumps him is “enjoy.”

Luis looks at the “j” over and over, each time making an “h” sound—the way “j” is pronounced in Spanish. An English native learning to read would make a hard “j” and might easily figure out that “en-j” might be “enjoy,” in part from the context. Luis doesn’t make the same connection and, frustrated, just turns the page.

“Things don’t sound wrong to these kids,” observes Cronin, who has taught bilingual education for nine years and has degrees from University of Vermont and Harvard Graduate School of Education. Cronin was disappointed when Question 2 passed, but concedes that “there were a number of classrooms where too much of the native language was being spoken.”

The new law, says Cronin, has brought more English exposure. She says that helps about two-thirds of her 18 students. The other six, she says, are new to the country and, some of them, new to schooling. These students, she says, need more than “clarification” in Spanish, their native language.

“They have no English vocabulary skills, and of those six, I have three who have no literacy skills. They don’t know letters, don’t know sounds. They don’t know that letters make sounds,” she says. Cronin tries to teach them letters in English, but they look at an apple for “A” and say “manzana.” She has a set of flash cards with English and Spanish words corresponding to the same letters and sounds—like “leon” and “lion” for “L” and “tortuga” and “turtle” for “T”—but the supply of such examples barely covers the alphabet.

“They are not learning-disabled,” says Cronin of these students. But they might as well be. “These are six kids who don’t know what they are doing.” How much Spanish can she use in class, under the law? Cronin doesn’t know.


Cronin is not the only teacher who is confused. “There is a general lack of understanding of what the law actually did,” says Kathy Kelley, president of the Massachusetts Federation of Teachers, which represents teachers in mostly urban districts, including Boston, Lawrence, Lowell, and Lynn. “After it passed, there was very little done in terms of preparing teachers. Now they have students mainstreamed into their classrooms.”

DOE has issued guidelines and provides on its Web site lists of resources and how-to documents like the one titled “Identifying Limited English Proficient (LEP) Students,” issued in October 2004. But other things have not been as clear.

Eighth-grader Nuria Teixeira is a model student
but still struggles with English.

Some school districts, for example, were initially confused about whether or not they had to provide services to LEP students who didn’t choose to attend a school that already had an English immersion class. In a few districts, large numbers of LEP students were effectively categorized as having “opted out” of any language support whatsoever. The Department of Education has since explained that LEP students have a right to English immersion, regardless of which school they choose to attend. Confusion has also arisen, for example, over teacher qualifications and how to decide when a student is no longer LEP.

It is also a challenge to dovetail English immersion rules with common sense, notes James Crawford, executive director of the National Association of Bilingual Education in Washington, DC. Crawford says research shows it takes years—not months—to learn English, especially for academic purposes. Immersion, he says, does nothing to change that fact.

“This approach has not had the effect of speeding up the acquisition of English,” says Crawford. A 2005 University of Arizona study examining that state’s stringent English-only law showed that more than 70 percent of LEP students tracked during the 2003-04 school year made no gain in English acquisition; some even lost ground.

Students may speak English in the hallways, but ‘they need more time’ to get it in class.

What happens to kids who have had a year of English immersion and are now expected to keep up, academically, in the language of their new country? Students in the US for six months may speak English in the hallways, and be able “to navigate our city,” says Guadalupe Guerrero, principal of the Dever School. “But when it comes to academic language, they need more time.”

At Dearborn Middle School in Roxbury, Leite says most of her eighth-grade sheltered English immersion pupils aren’t ready to move on after one year.

“Even at this grade, we have students who come in with a second-grade education, third-grade reading,” she says. The task is not just to teach them English, but also to catch them up.

One of her top students, Nuria Teixeira, arrived from Cape Verde in September 2004. A 14-year-old with a wide smile and wavy black hair pulled back into a ponytail, Nuria decided to be a doctor when she was six. She practices English with her brother and uncle and displays all the marks of an academically driven student: Her hand is always raised and she cares a lot about her grades. But even as she works hard at school, Nuria concedes, “Sometimes I don’t understand.”

That’s a worry shared by Chris Coxon, Boston Public Schools deputy superintendent for teaching and learning, who oversees bilingual education and professional development. Coxon sees LEP students struggling with test scores “much worse than other years,” leading him to pinpoint a distressing new problem in the post-bilingual education era: the large number of students who are English language learners but are not in sheltered English immersion classrooms.

Since Boston places students in school by choice, not mandatory assignment, Coxon believes nearly all the district’s 4,000 classrooms must become, in effect, sheltered English immersion, or SEI, classes.

“Every classroom that has at least one ELL child is entitled to SEI instruction,” he says. Coxon says LEP students who have completed a year of English immersion, or even more, still may not be capable of functioning in a mainstream English-speaking classroom. All Boston classrooms must be ready to serve LEP kids, says Coxon, but that’s easier said than done.

“If you walk around our schools, there are still [teachers] who are not sure about how to develop good content objectives” for their regular lessons, he says. “Then, if you ask them to plan a good language objective [required for SEI classes], that is a challenge.”

Coxon says the confusion has been at many levels. He says the state “did not define a program” for sheltered English immersion, leaving districts to search for one. And until last summer, he says, districts didn’t know what the state meant by the requirement that they have a “qualified teacher” in every sheltered English immersion classroom. Only later did Coxon learn that a “qualified” teacher is one who uses SEI pedagogical approaches and has 75 hours of specific training. But that did not solve his problem.

“I control 18 hours a year of my teachers’ professional development,” says Coxon. “You can imagine how long it will take me to get everyone trained.”

Meanwhile, Coxon is also looking for better ways to track LEP students. How much and what kind of support do they need as they enter mainstream classes? Then there is the matter of definitions. “We are in the process now of trying to clarify, ‘How do we figure out when a child is no longer an English language learner?’” observes Coxon.

But more to the point: Are kids learning English any better now than before? Coxon doesn’t know. “The fact that I can’t answer that,” he says, “is part of the challenge.”