Virtually every American who has studied a foreign language in school, and who has later tried to use it among native speakers, can testify to the difficulty of becoming bilingual.
After years of instruction, reading and writing skills may be adequate in the second language. But attempts at oral communication are likely to be a challenge, demanding the recall of once-learned rules and meanings. And the reply, assuming there is a patient listener, can be perplexing.
Lucky for most Americans, an increasing number of foreigners speak English. By some estimates, English is now spoken by one billion people, nearly one-quarter of the world’s population.
In the United States, fluency in any other language is optional. But language minorities in this country have no choice; the ability to communicate in English is essential to economic and social advancement.
New waves of immigration since the late 1970’s, particularly from Asia and Central America, have put a special burden on the schools to provide quality English-as-a-second-language instruction, especially for children who receive little exposure to English outside the classroom.
But traditional methods of ESL, with their stress on memorization of grammar and vocabulary, have been spectacularly unsuccessful, according to a growing number of educators. Modeled on foreign-language teaching, such approaches train students to compose grammatically correct sentences, but prepare them poorly to communicate in English.
On the other hand, research has shown that children in classrooms that develop native-language skills often score higher in English than their counterparts in ESL-only or English-immersion programs.
This finding is hard to accept at face value. How could native-language instruction promote English acquisition? Why doesn’t intensive study of a language lead to fluency?
To explain these and similar paradoxes, a theory of second-language acquisition has developed over the past 15 years that has revolutionized the practice of both ESL and bilingual education.
Learning vs. Acquiring
Language fluency cannot be learned, says Stephen D. Krashen, one of the theory’s leading exponents. That is, conscious mastery of grammar and vocabulary—while serving an editing function in speech or writing—is an insufficient basis for effective communication.
Instead, such proficiency must be acquired, Mr. Krashen argues, and this happens, he says, “in one fundamental way: We acquire language when we understand it.’'
“What is spectacular about this idea,’' he explains, “is that it happens incidentally, involuntarily, subconsciously, and effortlessly.’'
Language acquisition, in Mr. Krashen’s theory, requires “comprehensible input’'—ideally, second-language messages that make sense, but that are just beyond the listener’s level of proficiency. Not only vocabulary, but grammatical rules are “picked up’’ in this way, he says, through understanding the messages in which they are embedded.
What counts is the quality of second-language exposure, not the quantity. A limited-English-proficient child placed in a “sink or swim’’ classroom primarily hears noise. With contextual clues—such as having a lesson first taught in the native language-English input becomes more comprehensible.
Mr. Krashen adds that “speaking per se does not cause language acquisition,’' but is a result of it, “a result of obtaining comprehensible input.’'
This corollary of the “input hypothesis’’ explains why LEP children typically go through a “silent period,’' and then suddenly begin speaking in English. “When they begin to speak,’' he says, “they are not beginning their acquisition; they are showing off their competence.’'
On the other hand, Mr. Krashen maintains, attempts to teach language through memorization and mimicry, through what he terms mindless exercises that illustrate the “structure of the week,’' are futile for several reasons.
First, there is no complete set of rules for what native speakers have internalized. Linguists, he says, “have described only a fraction’’ of the complex structure of any natural language, and most grammar textbooks are therefore woefully incomplete.
But even if a learner could achieve intellectual mastery of a new syntax—something that few educated adults can do, Mr. Krashen notes—and could successfully “focus on form’’ while using it to communicate, the process would work too slowly to be of use in conversation.
Mr. Krashen’s theory draws on the linguist Noam Chomsky’s hypothesis that a genetically determined “language-acquisition device’’ functions identically in all human beings. When this “mental organ’’ receives intelligible messages in a second language, Mr. Krashen argues, the brain has “no choice but to acquire that language, just as the visual system has no choice but to see, and the pancreas has no choice but to operate as pancreases do.’'
“The obvious implication,’' he says, “is that language teaching should be based on giving people messages they understand. For ESL students, a well-taught geography lesson, if it’s comprehensible, is a language lesson. In fact, it’s better than an ESL class, [where] we’re always wondering what to talk about.’'
A methodology known as “sheltered English’'—subject-matter lessons tailored to LEP students’ level of English proficiency—is increasingly being incorporated into bilingual-education programs, with good results, Mr. Krashen says.
And communication-based ESL techniques are also on the rise. The “natural approach,’' for example, teaches English through extensive use of physical and visual clues, minimal correction of grammatical errors, and an emphasis on communicating messages relevant to students’ needs and interests.
Unfortunately, however, psychological factors can, and usually do, interrupt the smooth functioning of the language-acquisition device, according to experts.
Mr. Krashen’s theory groups these negative influences under the term “affective filter.’' They include anxiety, lack of self-confidence, and inadequate motivation to speak the second language—any or all of which can retard the language-acquisition process by keeping comprehensible input from getting through.
In this respect, young children appear to have an advantage over their elders in learning English. Before puberty, they tend to be less self-conscious about making errors in a new language.
But a lack of self-esteem, characteristic of poor children who come to school speaking a low-status language, can impede their English acquisition by “raising the affective filter,’' Mr. Krashen says.
It is here that native-language instruction can play an important role, according to research. By providing official recognition to a minority language and culture, a school curriculum can enhance a LEP child’s self-esteem and provide a low-anxiety environment for English acquisition—two factors that promote academic progress.
But this is only the beginning of bilingual education’s benefits, Mr. Krashen maintains, benefits he says that he, as a longtime ESL professional, only began to recognize in recent years. In particular, he stresses, bilingual education has the potential “to make English input more comprehensible’’ for LEP children.
While evidence for this conclusion is growing, Mr. Krashen says, it remains controversial, because “this idea is counter-intuitive,’' not unlike other tenets of bilingual education.
The notion of “go East to get West’'—study Spanish to learn English—appears, to many educators as well as laymen, to defy common sense. Mr. Krashen’s conversion, he recalls, came about around 1980, during a discussion of Jim Cummins’s work on native-language literacy.
Genesis of Theory
Mr. Cummins, a researcher at the Ontario Institute for Studies in Education, in the mid-1970’s began formulating a critique of what he regarded as oversimplified notions on both sides of the bilingual-education debate.
First, Mr. Cummins says, there was the “intuitively appealing argument ... that deficiencies in English should be remediated by intensive instruction in English.’' In other words, teaching LEP students in their native language was a costly waste of time.
Second, there was the strongest “common-sense’’ argument for bilingual instruction—in fact, the argument that has shaped federal legislation in the field—that children cannot learn in a language they do not understand.
Neither hypothesis could stand up to theoretical examination, Mr. Cummins concluded, and both were contradicted in practice by successful educational programs.
The “insufficient-exposure hypothesis’'—still a mainstay of opposition to bilingual education—was summed up by the late John Ashbrook, a Republican Congressman from Ohio: “When children come out of the Spanish-language schools or Choctaw-language schools which call themselves bilingual,’' Mr. Ashbrook asked, “how is our educational system going to make them literate in what will still be a completely alien tongue?’'
In a similar vein, Carol Pendas Whitten, director of the U.S. Education Department’s office of bilingual education and minority languages affairs, has warned that prolonging native-language instruction “will hurt the child, because the child is kept out of the mainstream classroom and put in a remedial program.’'
Educational research strongly contradicts these assumptions, Mr. Cummins maintains. “The results of virtually every bilingual program evaluated during the past 50 years,’' he says, “show either no relationship or a negative relationship between amount of school exposure to the majority language and academic achievement in that language.’'
Moreover, he says, “the insufficient-exposure hypothesis’’ implies a bizarre theory of cognitive development: that first and second languages develop separately in the brain. In other words, that knowledge and skills learned in one tongue are not transferable to the other. There is “not one shred of evidence’’ for this model, he says.
Taken to its extreme, Mr. Cummins says, this view would “leave the bilingual in the curious predicament’’ of being unable to “communicate with himself.’' When switching languages, the bilingual would be hard-pressed to translate what he had heard or said.
‘Common Underlying Proficiency’
In opposing the notion of separate language development, Mr. Cummins advances a theory of “common underlying proficiency.’' That is, that languages develop in the same part of the brain, reinforcing each other at the base, while differing at the surface.
In this “iceberg model,’' literacy-related aspects of language ability—features that are most cognitively demanding and detached from contextual clues—share much in common. They “are interdependent across languages,’' he says, while “surface aspects’’ may differ substantially between, say, English and Swahili.
Thus, according to the theory, once a child has learned to read and think in one language, the transition to a second language will be more efficient.
Successful bilingual-education programs provide considerable empirical support, Mr. Cummins says, for what he has termed the “interdependence hypothesis.’' Studies in Sweden and the United States, for example, have shown that children who immigrate at the age of 10 or 11, after learning to read in their native tongue, tend to learn a second language quickly, and in other subjects soon outpace language-minority children who received little or no native-language instruction.
Canadian studies have indicated that academic skills learned in one language need not be relearned when instruction shifts to a second language. Such is the case for mathematics and social studies, as well as literacy skills. The full range of proficiencies involved in the ability to read appears to transfer readily, according to researchers, even when there are radical differences between alphabets.
The implications of such findings for Mr. Krashen’s “input hypothesis’’ are significant. A LEP child who has kept up in mathematics through Spanish instruction, for example, will benefit doubly when studying the subject in a “sheltered English’’ classroom, Mr. Krashen says. “He will not only get more math; he’ll get more English’’ than a counterpart who is behind in mathematics.
“The goal of bilingual programs is English literacy,’' he continues. “The route is through the first language. ... You learn to read by reading, by making sense out of print. Vocabulary grows, grammar grows, spelling ability grows, good writing style grows. It’s easier to make sense out of print in a language you understand.’'
And once acquired, those skills provide more context for comprehensible input in the second language.
Two Types of Language Proficiency
Mr. Cummins also takes issue with a prevailing rationale for bilingual education, the conventional wisdom that children cannot learn when there is a mismatch between the languages of home and school.
The U.S. Commission on Civil Rights summed up this position in 1975: “Lack of English proficiency is the major reason for language-minority students’ academic failure. Bilingual education is intended to ensure that students do not fall behind in subject-matter content while they are learning English.’'
On its face, Mr. Cummins says, this “language-mismatch hypothesis’’ is refuted by the overwhelming success of French immersion programs among English-speaking Canadian children. At the outset of these programs, children learn exclusively through the second language—geared to their level of proficiency—and English instruction is phased in later.
And theoretically, he says, this view reflects “an inadequate understanding of what is meant by ‘English proficiency,’'' a shortcoming manifested, he contends, in the practice of “transitional bilingual education’’ in the United States.
In transitional programs, the emphasis is on hurrying children into mainstream classes and replacing their native language with English—an objective written into the Bilingual Education Act. But Mr. Cummins warns that this approach “is likely to result in the creation of academic deficits in language-minority students.’'
While LEP children, usually within two years of entering school, develop what Mr. Cummins has called “basic interpersonal communications skills,’' or BICS, in English, they are not yet equipped for mainstream classrooms, he says. Sometimes dubbed “playground English,’' BICS is heavily dependent on context—conversational responses, gestures, visual cues, physical interactions.
Children may appear fluent in English, but this type of proficiency is insufficient for the cognitively demanding activities encountered in school, Mr. Cummins insists, citing the high incidence of failure among children who pass through “quick exit’’ bilingual programs.
The long-term outcome for such students is often “limited bilingualism,’' he says. While students have developed conversational skills in two languages, they lack the full range of linguistic tools they need for thinking.
Attainment of “cognitive-academic language proficiency,’' or CALP, Mr. Cummins argues, is necessary for children to succeed in the “context-reduced, cognitively demanding’’ activities of reading, writing, mathematics, science, and other school subjects. This typically takes five to seven years, he says, and is best achieved by building on the linguistic base language-minority children bring to school, rather than trying to replace it.
Native-language instruction, in his view, is “much more than an interim carrier of subject-matter content.’'
It is rather, he says, “the means through which the conceptual and communicative proficiency that underlies both [native-language] and English literacy is developed.’'
Late-Exit Bilingual Model
Mr. Cummins’s theory of language proficiency implies an instructional model that breaks with quick-exit, transitional bilingual education. It entails several years of native-language instruction—not only literacy, but concept development in all subjects. At the same time, children should be receiving communication-based ESL and sheltered-English classes. The payoff takes time—until the 5th or 6th grade—but Mr. Cummins argues that, because of the transferability of cognitive-academic learning proficiency, it comes “at no cost to English.’'
At this point, however, the theory raises a logical series of questions: If successful immersion programs have shown that children can learn through a second language, why can’t CALP be developed through English instruction? Why is bilingual education necessary for language-minority children in the United States? And why not pour all available resources into developing quality ESL and English-immersion programs?
Mr. Cummins answers that “sociocultural determinants of minority students’ school failure, [which] are more fundamental than linguistic factors,’' dictate special treatments.
He points out that there are academic benefits in encouraging language-minority children “to take pride in their cultural background,’' and in developing proficient bilingualism. In an English-dominant society, he says, both goals are endangered by “English-only’’ instructional approaches.
In addition, he says, a larger solution lies in “empowering minority students.’' Bilingual education, he maintains, is merely part of the educational intervention that is necessary to “counteract the power relations that exist within the broader society.’'
The Canadian experiment with immersion provides a concrete illustration of these arguments. One of the most successful language programs ever studied, French immersion has played a major role in the development of second-language-acquisition theory.
It began in the mid-1960’s, when a group of English-speaking parents in St. Lambert, a suburb of Montreal, began meeting to discuss their frustrations over the linguistic and cultural segregation of their community. In particular, they were disturbed about their children’s inability to gain fluency in French, despite a 12-year school curriculum in that language.
The prevalent “audio-lingual method’’ of grammar and vocabulary drills was clearly not producing functioning bilinguals, the parents believed. Working with Wallace E. Lambert and Wilder Penfield, two researchers on bilingualism at McGill University, they pressured the St. Lambert school district to try a new approach.
In the fall of 1965, administrators reluctantly agreed to an experimental French immersion program. Its goals were “to provide ... functional competence in both written and spoken aspects of French; to promote and maintain normal levels of English development; [and] to ensure achievement in academic subjects commensurate with students’ academic ability and grade level.’'
The program also sought “to instill in the students an understanding and appreciation of French Canadians, their language, and culture without detracting in any way from the students’ identity with and appreciation for English-Canadian culture.’'
The initial class of 26 English-speaking kindergarteners entered a school program conducted entirely in French. Not until the 2nd grade, after children had learned to read in the second language, was English language arts introduced for one period a day. Gradually, the proportion of English was increased in other subjects until it reached about 60 percent by the 6th grade.
To minimize the students’ anxiety, teaching methods avoided over-correction of errors in French. Children were motivated to acquire the language “to engage in meaningful and interesting communication,’' according to Fred Genesee, a McGill University psychologist.
Recognizing that “comprehension of language usually comes before their production skills,’' he says, the program did not require children to use French among themselves or with the teacher until the end of the 1st grade.
The outcomes of the experiment came as a pleasant surprise to everyone, and especially to skeptical district officials. According to Jim Cummins, children achieved “native-like levels in the receptive aspects of the language, in their listening skills and their reading skills.’'
While their speaking and writing skills were distinguishable from French natives, he reports—probably because of insufficient social interactions with French-speakers, he says—the immersion students became “quite fluent and quite comfortable in speaking French for the most part.’'
And while the children picked up the second language, Mr. Cummins adds, they made excellent progress in English and other academic subjects. The social goal of cross-cultural understanding appears to have been met as well, based on assessments of students’ attitudes toward their French neighbors.
As English-speaking parents learned about the success of immersion, demand for the program spread throughout Canada, and enrollment has reached about 200,000, according to Mr. Cummins. Variants now include delayed immersion, beginning in the middle elementary grades, and late immersion, beginning early in secondary school.
Where school administrators have remained resistant, Mr. Genesee says, efforts to persuade them have usually succeeded with the help of an advocacy group known as Canadian Parents for French.
Additive vs. Subtractive Bilingualism
There are virtually no English immersion programs for French-speaking children in Canada. Nor should there be, say the immersion researchers.
The approach, they insist, is appropriate only for language-majority children, whose native language is not jeopardized by low social status. Even in Quebec, where 80 percent of the population speaks French, Anglophone children’s first-language skills develop regardless of the quantity of English instruction they receive in school, according to immersion research.
For them, French immersion is a process of “additive bilingualism,’' an enrichment program in which they acquire a socially and economically valuable skill, without undermining their English competence or identification with the majority culture.
For French-speaking children, by contrast, the social and cultural dominance of English makes the native language “vulnerable to neglect and replacement,’' according to Mr. Lambert, one of the designers of French immersion. For language-minority children, he says, an English-immersion program “would result in a slow subtraction of the students’ French and its replacement by English.’'
Far from a desirable goal, this phenomenon of “subtractive bilingualism’’ entails adverse “cognitive and educational consequences,’' Mr. Lambert says. “The trouble is that, for most language-minority children, the home language has been the critical linguistic system associated with the development of basic concepts from infancy on.’'
“It would be an enormous mental-gymnastic feat for these children to replace and reprogram these concepts in English ... and at the same time, try to keep up with English-speaking peers in subject matters that introduce new ideas which build on basic concepts,’' he says.
It should come as no surprise that, when placed in this situation, a disproportionate number of language-minority children in the United States fall behind in their subjects, drop out of school, and question their cultural identity, he adds. They have been “placed in a psycholinguistic limbo where neither the home language nor English is useful as a tool of thought and expression, a type of semilingualism.’'
Bilingual education is the answer for such students, according to Mr. Lambert. The development of native-language proficiency can protect against the loss of that language—and maintain cultural pride—while they develop the English skills needed to succeed academically. Quality bilingual programs, he says, can “turn subtractive bilingualism into additive bilingualism.’'
In deciding what instructional treatment can best promote proficient bilingualism, Mr. Lambert and Richard Tucker of the Center for Applied Linguistics have formulated the following guideline: “Priority in early years of schooling should be given to the language ... least likely to be developed otherwise—in other words, the language most likely to be neglected.’'
This principle, in the view of Canadian immersion researchers, argues against English immersion as an alternative to bilingual education for language-minority children in the United States. Early results from a federally financed study of English immersion appear to confirm this hypothesis.
Mr. Cummins of the Ontario Institute concludes that “power and status relations between minority and majority groups exert a major influence on school performance.’' The lower the status of a “dominated group,’' he says, the lower the academic achievement.
He cites research on the school failure of Finnish immigrants in Sweden, where they are considered a low-status group, as compared with their success in Australia, where being a Finn carries no social stigma. In addition, he says, children of the Burakumin minority “perform poorly in Japan, but as well as other Japanese students in the United States.’'
Reacting to messages about their inferior status—which tend, Mr. Cummins says, to be perpetuated by the schools—minority children frequently exhibit what he calls “bicultural ambivalence’': shame of the first culture and hostility toward the second.
Bilingual educators report, for example, that some Mexican-American children refuse to speak their native language, even though their English skills are poorly developed.
Interactions with educators, Mr. Cummins says, are important factors in “empowering’’ or “disabling’’ language-minority students; whether schools revere or reject a minority language is near the top of the list. The benefits of successful bilingual programs, he says, may be more a result of their affirmation of the LEP child’s cultural identity than to their fostering of native-language cognitive skills.
Conversely, he asks, “Is the failure of many minority students in English-only immersion programs a function of cognitive/academic difficulties or of students’ ambivalence about the value of their cultural identity?’'
Either way, he says, bilingual education is the more appropriate treatment.
A version of this article appeared in the April 01, 1987 edition of Education Week as Language-Acquisition Theory Revolutionizing Instruction