Skim this before you argue against bilingual education
the myths I'm sure you've heard....
FALLACY 1: ENGLISH IS LOSING GROUND TO OTHER LANGUAGES IN THE UNITED STATES
More world languages are spoken in the United States today than ever before. However, this is a quantitative, not a qualitative change from earlier periods. Concentrations of non-English language speakers were common in the 19th century, as reflected by laws authorizing native language instruction in a dozen states and territories. In big cities as well as rural areas, children attended bilingual and non-English schools, learning in languages as diverse as French, Norwegian, Czech, and Cherokee. In 1900, there were at least 600,000 elementary school children receiving part or all of their instruction in German (Kloss 1998). Yet English survived without any help from government, such as official-language legislation.
FALLACY 2: NEWCOMERS TO THE UNITED STATES ARE LEARNING ENGLISH MORE SLOWLY NOW THAN IN PREVIOUS GENERATIONS
To the contrary, today's immigrants appear to be acquiring English more rapidly than ever before. While the number of minority-language speakers is projected to grow well into the next century, the number of bilinguals fluent in both English and another language is growing even faster. Between 1980 and 1990, the number of immigrants who spoke non-English languages at home increased by 59%, while the portion of this population that spoke English very well rose by 93% (Waggoner, 1995). In 1990, only 3% of U.S. residents reported speaking English less than well or very well. Only eight-tenths of one percent spoke no English at all. About three in four Hispanic immigrants, after 15 years in this country, speak English on a daily basis, while 70% of their children become dominant or monolingual in English (Veltman, 1988).
FALLACY 3: THE BEST WAY TO LEARN A LANGUAGE IS THROUGH "TOTAL IMMERSION"
There is no credible evidence to support the "time on task" theory of language learning--the claim that the more children are exposed to English, the more English they will learn. Research shows that what counts is not just the quantity, but the quality of exposure. Second-language input must be comprehensible to promote second-language acquisition (Krashen, 1996). If students are left to sink or swim in mainstream classrooms, with little or no help in understanding native-their lessons, they won't learn much English. If native-language instruction is used to make lessons meaningful, they will learn more English--and more subject matter, too.
FALLACY 4: CHILDREN LEARNING ENGLISH ARE RETAINED TOO LONG IN BILINGUAL CLASSROOMS, AT THE EXPENSE OF ENGLISH ACQUISITION
Time spent learning in well designed bilingual programs is learning time well spent. Knowledge and skills acquired in the native language--literacy in particular--are "transferable" to the second language. They do not need to be relearned in English (Krashen, 1996; Cummins, 1992). Thus, there is no reason to rush limited-English-proficient (LEP) students into the mainstream before they are ready.
Research over the past two decades has determined that, despite appearances, it takes children a long time to attain full proficiency in a second language. Often, they are quick to learn the conversational English used on the playground, but normally they need several years to acquire the cognitively demanding, decontextualized language used for academic pursuits (Collier & Thomas, 1989).
Bilingual education programs that emphasize a gradual transition to English and offer native-language instruction in declining amounts over time, provide continuity in children's cognitive growth and lay a foundation for academic success in the second language. By contrast, English-only approaches and quick-exit bilingual programs can interrupt that growth at a crucial stage, with negative effects on achievement (Cummins, 1992).
FALLACY 5: SCHOOL DISTRICTS PROVIDE BILINGUAL INSTRUCTION IN SCORES OF NATIVE LANGUAGES
Where children speak a number of different languages, rarely are there sufficient numbers of each language group to make bilingual instruction practical for everyone. In any case, the shortage of qualified teachers usually makes it impossible. For example, in 1994 California enrolled recently arrived immigrants from 136 different countries, but bilingual teachers were certified in only 17 languages, 96% of them in Spanish (CDE, 1995).
FALLACY 6: BILINGUAL EDUCATION MEANS INSTRUCTION MAINLY IN STUDENTS' NATIVE LANGUAGES, WITH LITTLE INSTRUCTION IN ENGLISH
Before 1994, the vast majority of U.S. bilingual education programs were designed to encourage an early exit to mainstream English language classrooms, while only a tiny fraction of programs were designed to maintain the native tongues of students.
Today, a majority of bilingual programs continue to deliver a substantial portion of the curriculum in English. According to one study, school districts reported that 28% of LEP elementary school students receive no native-language instruction. Among those who do, about a third receive more than 75% of their instruction in English; a third receive from 40 to 75% in English; and one third of these receive less than 40% in English. Secondary school students are less likely to be instructed in their native language than elementary school students (Hopstock et al. 1993).
FALLACY 7: BILINGUAL EDUCATION IS FAR MORE COSTLY THAN ENGLISH LANGUAGE INSTRUCTION
All programs serving LEP students--regardless of the language of instruction--require additional staff training, instructional materials, and administration. So they all cost a little more than regular programs for native English speakers. But in most cases the differential is modest. A study commissioned by the California legislature examined a variety of well implemented program models and found no budgetary advantage for English-only approaches. The incremental cost was about the same each year ($175-$214) for bilingual and English immersion programs, as compared with $1,198 for English as a second language (ESL) "pullout" programs. The reason was simple: the pullout approach requires supplemental teachers, whereas in-class approaches do not (Chambers & Parrish, 1992). Nevertheless, ESL pullout remains the method of choice for many school districts, especially where LEP students are diverse, bilingual teachers are in short supply, or expertise is lacking in bilingual methodologies.
FALLACY 8: DISPROPORTIONATE DROPOUT RATES FOR HISPANIC STUDENTS DEMONSTRATE THE FAILURE OF BILINGUAL EDUCATION
Hispanic dropout rates remain unacceptably high. Research has identified multiple factors associated with this problem, including recent arrival in the United States, family poverty, limited English proficiency, low academic achievement, and being retained in grade (Lockwood, 1996). No credible studies, however, have identified bilingual education among the risk factors, because bilingual programs touch only a small minority of Hispanic children.
FALLACY 9: RESEARCH IS INCONCLUSIVE ON THE BENEFITS OF BILINGUAL EDUCATION
Some critics argue that the great majority of bilingual program evaluations are so egregiously flawed that their findings are useless. After reviewing 300 such studies, Rossell and Baker (1996) judged only 72 to be methodologically acceptable. Of these, they determined that a mere 22% supported the superiority of transitional programs over English-only instruction in reading, 9% in math, and 7% in language. Moreover, they concluded that "TBE [transitional bilingual education] is never better than structured immersion" in English. In other words, they could find little evidence that bilingual education works.
Close analysis of Rossell and Baker's claims reveals some serious flaws of their own. Krashen (1996) questions the rigor of several studies the reviewers included as methodologically acceptable--all unfavorable to bilingual education and many unpublished in the professional literature. Moreover, Rossell and Baker relied heavily on program evaluations from the 1970s, when bilingual pedagogies were considerably less well developed. Compounding these weaknesses is their narrative review technique, which simply counts the votes for or against a program alternative--a method that leaves considerable room for subjectivity and reviewer bias (Dunkel, 1990).
Meta-analysis, a more objective method that weighs numerous variables in each study under review, has yielded more positive findings about bilingual education (Greene, 1998; Willig, 1985).
Most important, Krashen (1996) shows that Rossell and Baker are content to compare programs by the labels they have been given, with little consideration of the actual pedagogies being used. They treat as equivalent all approaches called TBE, even though few program details are available in many of the studies under review. Researchers who take the time to visit real classrooms understand how dangerous such assumptions can be. According to Hopstock et al. (1993), "When actual practices...are examined, a bilingual education program might provide more instruction in English than...an 'English as a second language' program." Moreover, from a qualitative perspective, programs vary considerably in how (one or both) languages are integrated into the curriculum and into the social context of the school. Finally, simplistic labels are misleading because bilingual and English immersion techniques are not mutually exclusive; several studies have shown that successful programs make extensive use of both (see, e.g., Ramirez et al., 1991).
Even when program descriptions are available, Rossell and Baker sometimes ignore them. For example, they cite a bilingual immersion program in El Paso as a superior English-only (submersion) approach, although it includes 90 minutes of Spanish instruction each day in addition to sheltered English. The researchers also include in their review several studies of French immersion in Canada, which they equate with all-English, structured immersion programs in the United States. As the Canadian program designers have repeatedly stressed, these models are bilingual in both methods and goals, and they serve students with needs that are quite distinct from those of English learners in this country.
FALLACY 10: LANGUAGE-MINORITY PARENTS DO NOT SUPPORT BILINGUAL EDUCATION BECAUSE THEY FEEL IT IS MORE IMPORTANT FOR THEIR CHILDREN TO LEARN ENGLISH THAN TO MAINTAIN THE NATIVE LANGUAGE
Naturally, when pollsters place these goals in opposition, immigrant parents will opt for English by wide margins. Who knows better the need to learn English than those who struggle with language barriers on a daily basis? But the premise of such surveys is false. Truly bilingual programs seek to cultivate proficiency in both tongues, and research has shown that students' native language can be maintained and developed at no cost to English. When polled on the principles underlying bilingual education for example, that developing literacy in the first language facilitates literacy development in English or that bilingualism offers cognitive and career-related advantages--a majority of parents are strongly in favor of such approaches (Krashen, 1996).
REFERENCES
California Department of Education (CDE). (1995). "Educational demographics unit. Language census report for California public schools." Sacramento: Author.
Chambers, J., & Parrish, T. (1992). "Meeting the challenge of diversity: An evaluation of programs for pupils with limited proficiency in English. Vol. IV, cost of programs and services for LEP students." Berkeley, CA: BW Associates.
Collier, V. P., & Thomas, W. P. (1989). How quickly can immigrants become proficient in school English? "Journal of Educational Issues of Language Minority Students," 5, p26-39.
Cummins, J. (1992). Bilingual Education and English Immersion: The Ramirez Report in Theoretical Perspective. "Bilingual Research Journal," 16, p91-104.
Dunkel, P. (1990). Implications of the CAI effectiveness research for limited-English-proficient learners. "Computers in the Schools," 7, p31-52.
Greene, J. P. (1998). A meta-analysis of the effectiveness of bilingual education. Claremont, CA: Tomas Rivera Policy Institute.
Hopstock, P., Bucaro, B., Fleischman, H. L., Zehler, A. M., & Eu, H. (1993). "Descriptive Study of Services to Limited English Proficient Students." Arlington, VA: Development Associates.
Kloss, H. (1998). "The American Bilingual Tradition." Washington, DC and McHenry, IL.: ERIC Clearinghouse on Languages and Linguistics and Delta Systems Inc.
Krashen, S. D. (1996). "Under Attack: The Case Against Bilingual Education." Culver City, CA: Language Education Associates.
Lockwood, A. T. (1996). Caring, Community, and Personalization: Strategies to Combat the Hispanic Dropout Problem. "Advances in Hispanic Education, 1." Washington, DC: U.S. Department of Education.
Ramirez, J. D., Yuen, S. D., & Ramey, D. R. (1991). "Final report: Longitudinal study of structured immersion strategy, early-exit, and late-exit transitional bilingual education programs for language-minority children. Executive summary." San Mateo, CA: Aguirre International.
Rossell, C., & Baker, K. (1996). The Educational Effectiveness of Bilingual Education. "Research in the Teaching of English," 30, p7-74.
Veltman, C. (1988). "The Future of the Spanish Language in the United States." Washington, DC: Hispanic Policy Development Project.
Waggoner, D. (1995, November). Are Current Home Speakers of Non-English Languages Learning English? "Numbers and Needs, 5."
Willig, Ann C. 1985. A Meta-Analysis of Selected Studies on the Effectiveness of Bilingual Education. "Review of Educational Research," 55, p269-317.
This Digest is drawn from the National Clearinghouse for Bilingual Education (NCBE) report Best Evidence: Research Foundations of the Bilingual Education Act (1997), by James Crawford. For the complete report, see the NCBE home page at http://www.ncbe.gwu.edu.
James Crawford is author of "Bilingual Education: History, Politics, Theory, and Practice," 4th ed. (Los Angeles: Bilingual Educational Services, 1999) 800-448-6032.
FALLACY 1: ENGLISH IS LOSING GROUND TO OTHER LANGUAGES IN THE UNITED STATES
More world languages are spoken in the United States today than ever before. However, this is a quantitative, not a qualitative change from earlier periods. Concentrations of non-English language speakers were common in the 19th century, as reflected by laws authorizing native language instruction in a dozen states and territories. In big cities as well as rural areas, children attended bilingual and non-English schools, learning in languages as diverse as French, Norwegian, Czech, and Cherokee. In 1900, there were at least 600,000 elementary school children receiving part or all of their instruction in German (Kloss 1998). Yet English survived without any help from government, such as official-language legislation.
FALLACY 2: NEWCOMERS TO THE UNITED STATES ARE LEARNING ENGLISH MORE SLOWLY NOW THAN IN PREVIOUS GENERATIONS
To the contrary, today's immigrants appear to be acquiring English more rapidly than ever before. While the number of minority-language speakers is projected to grow well into the next century, the number of bilinguals fluent in both English and another language is growing even faster. Between 1980 and 1990, the number of immigrants who spoke non-English languages at home increased by 59%, while the portion of this population that spoke English very well rose by 93% (Waggoner, 1995). In 1990, only 3% of U.S. residents reported speaking English less than well or very well. Only eight-tenths of one percent spoke no English at all. About three in four Hispanic immigrants, after 15 years in this country, speak English on a daily basis, while 70% of their children become dominant or monolingual in English (Veltman, 1988).
FALLACY 3: THE BEST WAY TO LEARN A LANGUAGE IS THROUGH "TOTAL IMMERSION"
There is no credible evidence to support the "time on task" theory of language learning--the claim that the more children are exposed to English, the more English they will learn. Research shows that what counts is not just the quantity, but the quality of exposure. Second-language input must be comprehensible to promote second-language acquisition (Krashen, 1996). If students are left to sink or swim in mainstream classrooms, with little or no help in understanding native-their lessons, they won't learn much English. If native-language instruction is used to make lessons meaningful, they will learn more English--and more subject matter, too.
FALLACY 4: CHILDREN LEARNING ENGLISH ARE RETAINED TOO LONG IN BILINGUAL CLASSROOMS, AT THE EXPENSE OF ENGLISH ACQUISITION
Time spent learning in well designed bilingual programs is learning time well spent. Knowledge and skills acquired in the native language--literacy in particular--are "transferable" to the second language. They do not need to be relearned in English (Krashen, 1996; Cummins, 1992). Thus, there is no reason to rush limited-English-proficient (LEP) students into the mainstream before they are ready.
Research over the past two decades has determined that, despite appearances, it takes children a long time to attain full proficiency in a second language. Often, they are quick to learn the conversational English used on the playground, but normally they need several years to acquire the cognitively demanding, decontextualized language used for academic pursuits (Collier & Thomas, 1989).
Bilingual education programs that emphasize a gradual transition to English and offer native-language instruction in declining amounts over time, provide continuity in children's cognitive growth and lay a foundation for academic success in the second language. By contrast, English-only approaches and quick-exit bilingual programs can interrupt that growth at a crucial stage, with negative effects on achievement (Cummins, 1992).
FALLACY 5: SCHOOL DISTRICTS PROVIDE BILINGUAL INSTRUCTION IN SCORES OF NATIVE LANGUAGES
Where children speak a number of different languages, rarely are there sufficient numbers of each language group to make bilingual instruction practical for everyone. In any case, the shortage of qualified teachers usually makes it impossible. For example, in 1994 California enrolled recently arrived immigrants from 136 different countries, but bilingual teachers were certified in only 17 languages, 96% of them in Spanish (CDE, 1995).
FALLACY 6: BILINGUAL EDUCATION MEANS INSTRUCTION MAINLY IN STUDENTS' NATIVE LANGUAGES, WITH LITTLE INSTRUCTION IN ENGLISH
Before 1994, the vast majority of U.S. bilingual education programs were designed to encourage an early exit to mainstream English language classrooms, while only a tiny fraction of programs were designed to maintain the native tongues of students.
Today, a majority of bilingual programs continue to deliver a substantial portion of the curriculum in English. According to one study, school districts reported that 28% of LEP elementary school students receive no native-language instruction. Among those who do, about a third receive more than 75% of their instruction in English; a third receive from 40 to 75% in English; and one third of these receive less than 40% in English. Secondary school students are less likely to be instructed in their native language than elementary school students (Hopstock et al. 1993).
FALLACY 7: BILINGUAL EDUCATION IS FAR MORE COSTLY THAN ENGLISH LANGUAGE INSTRUCTION
All programs serving LEP students--regardless of the language of instruction--require additional staff training, instructional materials, and administration. So they all cost a little more than regular programs for native English speakers. But in most cases the differential is modest. A study commissioned by the California legislature examined a variety of well implemented program models and found no budgetary advantage for English-only approaches. The incremental cost was about the same each year ($175-$214) for bilingual and English immersion programs, as compared with $1,198 for English as a second language (ESL) "pullout" programs. The reason was simple: the pullout approach requires supplemental teachers, whereas in-class approaches do not (Chambers & Parrish, 1992). Nevertheless, ESL pullout remains the method of choice for many school districts, especially where LEP students are diverse, bilingual teachers are in short supply, or expertise is lacking in bilingual methodologies.
FALLACY 8: DISPROPORTIONATE DROPOUT RATES FOR HISPANIC STUDENTS DEMONSTRATE THE FAILURE OF BILINGUAL EDUCATION
Hispanic dropout rates remain unacceptably high. Research has identified multiple factors associated with this problem, including recent arrival in the United States, family poverty, limited English proficiency, low academic achievement, and being retained in grade (Lockwood, 1996). No credible studies, however, have identified bilingual education among the risk factors, because bilingual programs touch only a small minority of Hispanic children.
FALLACY 9: RESEARCH IS INCONCLUSIVE ON THE BENEFITS OF BILINGUAL EDUCATION
Some critics argue that the great majority of bilingual program evaluations are so egregiously flawed that their findings are useless. After reviewing 300 such studies, Rossell and Baker (1996) judged only 72 to be methodologically acceptable. Of these, they determined that a mere 22% supported the superiority of transitional programs over English-only instruction in reading, 9% in math, and 7% in language. Moreover, they concluded that "TBE [transitional bilingual education] is never better than structured immersion" in English. In other words, they could find little evidence that bilingual education works.
Close analysis of Rossell and Baker's claims reveals some serious flaws of their own. Krashen (1996) questions the rigor of several studies the reviewers included as methodologically acceptable--all unfavorable to bilingual education and many unpublished in the professional literature. Moreover, Rossell and Baker relied heavily on program evaluations from the 1970s, when bilingual pedagogies were considerably less well developed. Compounding these weaknesses is their narrative review technique, which simply counts the votes for or against a program alternative--a method that leaves considerable room for subjectivity and reviewer bias (Dunkel, 1990).
Meta-analysis, a more objective method that weighs numerous variables in each study under review, has yielded more positive findings about bilingual education (Greene, 1998; Willig, 1985).
Most important, Krashen (1996) shows that Rossell and Baker are content to compare programs by the labels they have been given, with little consideration of the actual pedagogies being used. They treat as equivalent all approaches called TBE, even though few program details are available in many of the studies under review. Researchers who take the time to visit real classrooms understand how dangerous such assumptions can be. According to Hopstock et al. (1993), "When actual practices...are examined, a bilingual education program might provide more instruction in English than...an 'English as a second language' program." Moreover, from a qualitative perspective, programs vary considerably in how (one or both) languages are integrated into the curriculum and into the social context of the school. Finally, simplistic labels are misleading because bilingual and English immersion techniques are not mutually exclusive; several studies have shown that successful programs make extensive use of both (see, e.g., Ramirez et al., 1991).
Even when program descriptions are available, Rossell and Baker sometimes ignore them. For example, they cite a bilingual immersion program in El Paso as a superior English-only (submersion) approach, although it includes 90 minutes of Spanish instruction each day in addition to sheltered English. The researchers also include in their review several studies of French immersion in Canada, which they equate with all-English, structured immersion programs in the United States. As the Canadian program designers have repeatedly stressed, these models are bilingual in both methods and goals, and they serve students with needs that are quite distinct from those of English learners in this country.
FALLACY 10: LANGUAGE-MINORITY PARENTS DO NOT SUPPORT BILINGUAL EDUCATION BECAUSE THEY FEEL IT IS MORE IMPORTANT FOR THEIR CHILDREN TO LEARN ENGLISH THAN TO MAINTAIN THE NATIVE LANGUAGE
Naturally, when pollsters place these goals in opposition, immigrant parents will opt for English by wide margins. Who knows better the need to learn English than those who struggle with language barriers on a daily basis? But the premise of such surveys is false. Truly bilingual programs seek to cultivate proficiency in both tongues, and research has shown that students' native language can be maintained and developed at no cost to English. When polled on the principles underlying bilingual education for example, that developing literacy in the first language facilitates literacy development in English or that bilingualism offers cognitive and career-related advantages--a majority of parents are strongly in favor of such approaches (Krashen, 1996).
REFERENCES
California Department of Education (CDE). (1995). "Educational demographics unit. Language census report for California public schools." Sacramento: Author.
Chambers, J., & Parrish, T. (1992). "Meeting the challenge of diversity: An evaluation of programs for pupils with limited proficiency in English. Vol. IV, cost of programs and services for LEP students." Berkeley, CA: BW Associates.
Collier, V. P., & Thomas, W. P. (1989). How quickly can immigrants become proficient in school English? "Journal of Educational Issues of Language Minority Students," 5, p26-39.
Cummins, J. (1992). Bilingual Education and English Immersion: The Ramirez Report in Theoretical Perspective. "Bilingual Research Journal," 16, p91-104.
Dunkel, P. (1990). Implications of the CAI effectiveness research for limited-English-proficient learners. "Computers in the Schools," 7, p31-52.
Greene, J. P. (1998). A meta-analysis of the effectiveness of bilingual education. Claremont, CA: Tomas Rivera Policy Institute.
Hopstock, P., Bucaro, B., Fleischman, H. L., Zehler, A. M., & Eu, H. (1993). "Descriptive Study of Services to Limited English Proficient Students." Arlington, VA: Development Associates.
Kloss, H. (1998). "The American Bilingual Tradition." Washington, DC and McHenry, IL.: ERIC Clearinghouse on Languages and Linguistics and Delta Systems Inc.
Krashen, S. D. (1996). "Under Attack: The Case Against Bilingual Education." Culver City, CA: Language Education Associates.
Lockwood, A. T. (1996). Caring, Community, and Personalization: Strategies to Combat the Hispanic Dropout Problem. "Advances in Hispanic Education, 1." Washington, DC: U.S. Department of Education.
Ramirez, J. D., Yuen, S. D., & Ramey, D. R. (1991). "Final report: Longitudinal study of structured immersion strategy, early-exit, and late-exit transitional bilingual education programs for language-minority children. Executive summary." San Mateo, CA: Aguirre International.
Rossell, C., & Baker, K. (1996). The Educational Effectiveness of Bilingual Education. "Research in the Teaching of English," 30, p7-74.
Veltman, C. (1988). "The Future of the Spanish Language in the United States." Washington, DC: Hispanic Policy Development Project.
Waggoner, D. (1995, November). Are Current Home Speakers of Non-English Languages Learning English? "Numbers and Needs, 5."
Willig, Ann C. 1985. A Meta-Analysis of Selected Studies on the Effectiveness of Bilingual Education. "Review of Educational Research," 55, p269-317.
This Digest is drawn from the National Clearinghouse for Bilingual Education (NCBE) report Best Evidence: Research Foundations of the Bilingual Education Act (1997), by James Crawford. For the complete report, see the NCBE home page at http://www.ncbe.gwu.edu.
James Crawford is author of "Bilingual Education: History, Politics, Theory, and Practice," 4th ed. (Los Angeles: Bilingual Educational Services, 1999) 800-448-6032.