Research, Ethics, and Public Discourse:
The Debate on Bilingual Education
Presentation at the National Conference of the
American Association of Higher Education
March 22, 1999
by Jim Cummins
University of Toronto
Few educational issues in North America have become as volatile or as
ideologically loaded as the debate on the merits or otherwise of bilingual
education. Twenty-five years of debate culminated in June 1998 with the
passage of Proposition 227 in California by a margin of 61 to 39 percent.
Proposition 227 was aimed at eliminating the use of bilingual children’s
first language (L1) for instructional purposes except in very exceptional
Research has played a prominent role in this debate. Unfortunately,
the research evidence has been interpreted in very different ways by advocates
and opponents of bilingual education. In this presentation I want to raise
ethical issues about the way in which research evidence has been infused
into the public discourse on bilingual education. Unlike courtroom lawyers
who advocate for their clients regardless of the merits of the case, academics
have an ethical responsibility to analyze the research evidence as objectively
as possible and to recommend policy options that are consistent with the
evidence. There is also a responsibility to address and to reconcile internal
contradictions in their stated positions and interpretation of the research.
The academic debate on bilingual education contrasts markedly with the
treatment of the issue in the media. Media articles on bilingual education
have tended to be overwhelmingly negative in their assessment of the merits
of bilingual programs.  By contrast, the academic debate
lines up virtually all North American applied linguists who have carried
out research on language learning as advocates of bilingual programs against
only a handful of academics who oppose bilingual education. None of those
who oppose bilingual education has a background in the discipline of applied
linguistics. The most prominent of these opposing academics are Rosalie
Pedalino Porter, Keith Baker, Christine Rossell, and Charles Glenn. Others
such as Nathan Glazer and Herbert Walberg have made occasional forays into
the debate to express their scepticism about bilingual education.
In the next section I briefly sketch some interpretations of the research
that I believe would be endorsed by a large majority of applied linguists.
Then I examine contradictions in the claims of some of the opponents of
The Applied Linguistics Perspective
1. Bilingual programs for minority and majority language students
have been successfully implemented in countries around the world. As
documented in a recent volume by Cummins and Corson that documented programs
in more than 30 countries, students educated for part of the day through
a minority language do not suffer adverse consequences in the development
of academic skills in the majority language.  If there
were adverse consequences associated with bilingual instruction, there
would hardly be more than 300,000 English-background students in various
forms of French-English bilingual programs in Canada.
2. Bilingual education, by itself, is not a panacea for students’
underachievement. Underachievement derives from many sources and simply
providing some first language (L1) instruction will not, by itself, transform
students’ educational experience. Bilingual instruction can
make a significant contribution (as Jay Greene’s meta-analysis
outlined below suggests) but the predominant model of bilingual education
(quick-exit transitional programs) is inferior to programs that aim to
develop bilingualism and biliteracy, such as developmental (late-exit)
and two-way bilingual immersion (dual-language). Dual-language programs
also serve English-background students in the same classes as minority
language students with each group acting as linguistic models for the other.
3. The development of literacy in two languages entails linguistic
and perhaps cognitive advantages for bilingual students. There are
close to 150 research studies carried out since the early 1960s that report
significant advantages for bilingual students on a variety of metalinguistic
and cognitive tasks.
4. Significant positive relationships exist between the development
of academic skills in L1 and L2. This is true even for languages that
are dissimilar (e.g. Spanish and Basque; English and Chinese; Dutch and
Turkish). These cross-lingual relationships provide evidence for a common
underlying proficiency that permits transfer of academic and conceptual
knowledge across languages.
5. Conversational and academic aspects of language proficiency are
distinct and follow different developmental patterns. Several large-scale
studies have shown that it usually takes at least five years for second
language learners to catch up academically to their native English-speaking
peers but conversational fluency in English is often attained within two
years of intensive exposure to the language. These data are very much at
variance with the assumptions of Proposition 227 which provides only one
year of intensive English language instruction prior to mainstreaming students
into the regular classroom without specific language support.
Doublethink in Research and Policy
The term doublethink was coined by George Orwell in his
futuristic novel Nineteen Eighty-Four to refer to the simultaneous
belief in two contradictory ideas. The phenomenon is very evident in the
academic arguments that have been made against bilingual education. Rosalie
Pedalino Porter, for example, argues against transitional bilingual education
on the grounds that such programs entail less “time-on-task” than monolingual
English programs.  Yet in the same book she strongly
endorses two-way bilingual immersion programs which have far more L1 instruction
for minority students (usually at least 50% L1 from K through 6). According
to Porter, such programs promise "mutual learning, enrichment, and
respect" (p. 154) and "are also considered to be the best possible
vehicles for integration of language minority students, since these students
are grouped with English-speakers for natural and equal exchange of skills"
Keith Baker has also jumped on the doublethink bandwagon
in providing two totally opposite interpretations of the same program results.
The El Paso program was labeled “bilingual immersion” by the district and
involved a “native language cognitive development” component of 90 minutes
a day at grade 1, gradually reducing to 60 minutes a day by grade 3 and
30 minutes a day by grade 4. In 1992, Baker correctly critiqued Porter’s
mis-interpretation of this program as being an all-English program:
What Porter describes as an all-English immersion program in El Paso
is, in fact, a Spanish-English dual immersion program. The El Paso study
supports the claims of bilingual education advocates that most bilingual
education programs do not use enough of the native language. It does not
support Porter's claims that they should use less. 
Yet, six years later Baker’s interpretation of this program has changed
radically with no explanation of the change for the reader.
El Paso created an SEI [structured English immersion] program in which
Spanish instruction was reduced to 30 minutes a day. The district followed
students from this program and from the state-mandated bilingual education
program for 12 years. The SEI students scored significantly higher on all
tests for 11 straight years. In the 12th year, the SEI students still scored
higher, but their advantage was no longer statistically significant, suggesting
that, after a decade or so, the harm that bilingual education programs
do to learning English is more or less wiped out by continued exposure
to English outside the classroom. 
It is clearly an extreme example of doublethink to be
able to describe in 1992 a program as “a Spanish-English dual immersion
program” whose positive results support the “claims of bilingual education
advocates that most bilingual education programs do not use enough of the
native language” and six years later to describe exactly the same program
as a “structured English immersion” program whose positive results illustrate
“the harm that bilingual education programs do to learning English.”
Charles Glenn’s major concern with bilingual education has been his
perception that such programs segregate bilingual students from the mainstream,
potentially contributing to their long-term marginalization in school and
society. He fails to acknowledge, however, that segregation in schools
is primarily a function of housing and neighborhood concentrations of particular
ethnic groups and will exist regardless of the language of instruction.
Despite his expressed opposition to bilingual education (as illustrated
for example in his support for Proposition 227), Glenn (like Porter) has
been an articulate supporter of two-way bilingual programs for language
minority (and majority) students. He notes, for example:
The best setting for educating linguistic minority pupils – and one
of the best for educating any pupils – is a school in which
two languages are used without apology and where becoming proficient in
both is considered a significant intellectual and cultural achievement.
It should be clear at this stage that considerable common ground is
emerging between “opponents” and “advocates” of bilingual education. So-called
advocates have been highly critical of many quick-exit transitional programs
on the grounds that they do not aspire to develop bilingualism or biliteracy
and also fail to affirm strongly students’ cultural and linguistic identity.
Virtually all applied linguists endorse developmental or two-way bilingual
immersion programs in preference to quick-exit transitional bilingual programs.
Surprisingly, the same appears true of the so-called “opponents” of bilingual
education: they are highly critical of transitional bilingual programs
but have strongly endorsed two-way bilingual immersion programs. They have
not, however, addressed the contradictions that their endorsement of two-way
bilingual programs entails. To argue against bilingual education while
at the same time endorsing the most intensive form of bilingual education
at the very least requires explanation.
Unlike other academic “opponents” of bilingual education, Christine
Rossell has not explicitly endorsed two-way bilingual immersion programs.
Yet, her arguments for structured English immersion programs are based
overwhelmingly on the documented success of bilingual and trilingual programs.
Rossell and Baker reviewed a large number of program evaluations and cite
ten research studies which they claim show structured immersion to be superior
to transitional bilingual education (TBE).  Specifically,
they claim that in comparisons of reading performance in TBE versus Structured
Immersion, no difference was found in 17 percent and structured immersion
was significantly superior to TBE in 83 percent of studies. These statistics
sound impressive but they obscure the fact that nine out of ten of the
so-called “structured immersion” programs were actually bilingual or trilingual
programs. Thus, even though Rossell does not publicly endorse bilingual
education, the fact that she relies on the success of bilingual and trilingual
programs to make her point constitutes an implicit endorsement of bilingual
It is worth noting that the Tomas Rivera Center published a review of
essentially the same evaluation database by Jay Greene which showed that
participation in a bilingual program (defined as one that had instruction
through two languages) in the United States contributed a significant increment
to academic achievement in comparison to participation in a monolingual
English program. 
The Ethics of Policy-Related Research
There will always be legitimate differences of opinion in the interpretation
of academic research. Scientific progress is made possible by means of
dialogue, discussion, and further research designed to resolve the differences.
This process of dialogue has not happened in the area of bilingual education.
A negative spin on the research to the tune of “bilingual education doesn’t
work” has been fed directly to the media and has polluted public discourse
on this topic. I use the strong label pollution to convey
the fact that the message broadcast by the media ignores the consensus
among virtually all North American researchers that (a) countless successful
bilingual programs have been implemented in countries throughout the world,
and (b) two-way bilingual immersion programs have produced consistently
positive outcomes for both language minority and majority students and
constitute a viable policy-option for helping to reverse bilingual students’
academic underachievement. I believe that academics, in contrast to courtroom
lawyers, have an ethical responsibility to clean up the info-pollution
or to publicly admit that they have abandoned academic standards in favor
of the standards of adversarial discourse where the goal is to win rather
than to contribute to effective policy grounded in solid research.
1. McQuillan, J., & Tse, L. (1996). Does research
matter? An analysis of media opinion on bilingual education 1984-1994.
Bilingual Research Journal, 20 (1), 1-27.
2. Cummins, J. & Corson, D. (1997). Bilingual
education. Vol. 5, International Encyclopedia of Language and Education.
Dordrecht, The Netherlands: Kluwer Academic Publishers.
3. Porter, R. P. (1990). Forked tongue: The politics
of bilingual education. New York: Basic Books.
4. Baker, K. (1992, Winter/Spring). Review of Forked
Tongue. Bilingual Basics, pp. 6-7.
5. Baker, K. (1998). Structured English immersion: Breakthrough
in teaching limited-English-proficient students. Phi Delta Kappan
(November), 199-204. It is worth noting that Baker’s second account of
the El Paso findings are highly inaccurate. For example, differences between
the programs had disappeared by grade 7, not grade 12.
6. Glenn, C. L. & LaLyre, I. (1991). Integrated
bilingual education in the USA. In K. Jaspaert & S. Kroon (Eds.), Ethnic
minority languages and education (pp. 37-55). Amsterdam: Swets &
7. Rossell, C. H., & Baker, K. (1996). The effectiveness
of bilingual education. Research in the Teaching of English, 30,
8. Greene, J. P. (1998). A meta-analysis
of the effectiveness of bilingual education. Claremont, CA: Tomas
Rivera Policy Institute. Greene reports that participation in a bilingual
program over a period of two years made a difference of about 1/5 of a
standard deviation in achievement. Thus, if the English-only student performed
at the 26th percentile at the end of those two years, the bilingual student
would be at the 34th percentile.
COPYRIGHT NOTICE: Copyright © 1999 by
Jim Cummins. All rights reserved. Reprinted by permission.