Home     resources

Summary of paper presented at the International Conference on Bilingualism, Bristol, April 20, 2001

  The Academic and Political Discourse of Minority Language Education: Claims and Counter-Claims About Reading, Academic Language, Pedagogy, and Assessment as They Relate to Bilingual Children’s Educational Development

 Jim Cummins 

University of Toronto

Jump to: Language Development, Reading, Bilingual Instruction and Assessment.  

Academic Language Development

Naïve Reactionary View:

1.          English can be “picked up” rapidly by young children;

2.          1 year of intensive immersion is sufficient to learn English; 

Naïve Progressive View:

1.          All bilingual children require 5+ years of ESL or bilingual support to achieve academically; 

Research Findings:

1.          Progress to grade-appropriate academic language performance usually requires 5+ years of L2 academic learning;

2.          Acquisition of peer-appropriate conversational performance usually requires 1-2 years of exposure;

3.          Conversational abilities (Gee’s primary discourse) are conceptually distinct from academic language abilities (one form of secondary discourse); differences are apparent in developmental patterns (compare 6- and 12-year olds) and in linguistic forms (e.g. high-frequency Anglo-Saxon lexicon v. low-frequency Graeco-Latin lexicon [Corson, Nation]; syntax [Biber]).

Reading

Naïve Reactionary View:

1.          Intensive systematic teaching of phonemic awareness and phonics in English is the key to reversing reading difficulties and academic underachievement among low-income and bilingual children; 

Naïve Progressive View:

1.          Children learn to read by reading; explicit, systematic phonics instruction is unnecessary and potentially harmful (this view is often identified by opponents as the whole-language position);

2.          Initial reading instruction through L2 will result in academic failure; therefore bilingual children should be taught to read in L1;

3.          Reading instruction in L2 should be delayed until L1 reading is well-established. 

Research Findings:

1.          Decoding skills are a necessary but not sufficient condition for reading comprehension development; many low-income students who appear to perform well on standardized tests in the early grades experience a “grade 4 slump” when reading comprehension rather than decoding becomes the primary focus of standardized tests of reading;

2.          The most effective approaches to developing initial reading skills (decoding) are those that combine extensive and varied exposure to meaningful print with explicit and systematic instruction in phonemic awareness and letter-sound correspondences;

3.          Systematic phonics instruction can enable second language learners to acquire word recognition and decoding skills in their second language to a relatively high level, despite the fact that their knowledge of the second language is still limited. These decoding skills, however, do not automatically generalize to reading comprehension or other aspects of second language proficiency;

4.          The order in which reading is introduced in a bilingual program (L1 first, L2 first, L1/L2 together) is not, in itself, a significant predictor of reading development or academic progress;

5.          After the initial grades, reading comprehension is predicted primarily by the amount that students actually read; extensive reading provides access to a wide range of vocabulary which has consistently been shown to be the strongest predictor of readability; psychometrically, vocabulary knowledge is virtually indistinguishable from reading comprehension.

Bilingual Instruction

Naïve Reactionary View:

1.          If children are deficient in English, they need maximum exposure to English in school; bilingual instruction dilutes exposure to English and consequently will result in lower English achievement (maximum exposure/time-on-task hypothesis);

2.          Children in transitional bilingual programs should be mainstreamed into English-only instruction as rapidly as possible to enable them to learn English. 

Naïve Progressive View:

1.          If children are exposed to a home-school language switch, academic retardation is inevitable because children cannot learn through a language they do not understand (linguistic mismatch hypothesis);

2.          Initial reading instruction should be in the language children know best;

3.          Reading and writing instruction in L2 should be delayed until L1 literacy is well-established;

4.          L1 and L2 should be kept rigidly separate in bilingual programs.  

Research Findings:

1.          Instruction through a minority language a well-implemented bilingual program entails no adverse effects on academic development in the majority language (for either minority-language or majority-language L1 students);

2.          Moderately strong relationships have consistently been reported between L1 and L2 academic development in bilingual programs both for cognate and non-cognate languages (e.g. Basque-Spanish) (the interdependence hypothesis);

3.          Transfer of academic language skills can be two-way (from L1-L2 and L2-L1 – Verhoeven’s research);

4.          Continued development of L1 and L2 through elementary school is associated with enhanced awareness of language and, to a lesser extent, increased cognitive flexibility;

5.          For minority-language students, the most positive outcomes for L2 and L1 academic development have been reported in dual-language and developmental programs that promote L1 literacy throughout elementary school (e.g. Thomas/Collier research);

6.          The primary causes of underachievement among marginalized students are rooted in coercive power structures in the wider society; reversal of underachievement requires that schools actively challenge the operation of this power structure; some “bilingual” programs may reinforce coercive power structures while some English-medium programs may challenge coercive power structures.

Assessment

Naïve Reactionary View:

1.          Regular assessment of students’ progress by means of standardized tests will increase schools’ accountability and improve student achievement;

2.          Virtually all students should be included in standardized testing programs regardless of English language proficiency or length of residence;

3.          The outcomes of standardized assessment for particular schools reflect the quality of instruction in those schools;

4.          English-only assessment in the early grades can validly reflect the quality of instruction in bilingual programs.

Naïve Progressive View:

1.          Standardized tests measure only “nonsense” and are inappropriate for all children, under any circumstances and for any purpose;

Research Findings:

1.          Standardized reading tests reflect decoding skills in the early grades and increasingly comprehension skills in later grades (3+); thus, early grade results are not necessarily a good indicator of later progress (the grade 4 slump);

2.          High stakes testing can dramatically constrict instruction and reinforce “banking education” (Gándara, McNeil research)

3.          Standardized (or non-standardized) English-only academic language measures will underestimate bilingual students’ academic progress and potential for at least 5 years after they start learning English;

4.          Use of standardized verbal ability tests for special education diagnosis and placement has resulted in significant overrepresentation of bilingual and culturally diverse students in classes for handicapped students (e.g. “learning disability”) and under-representation of such students in classes for gifted and talented students;

5.          When administered appropriately, standardized reading tests assess the same underlying academic language construct as more authentic measures such as cloze tests, miscue analysis, etc.

6.          Teaching-to-the-test is likely to be much less effective in the long run than promoting extensive reading and writing, and language awareness activities (Focus on Meaning, Focus on Language, Focus on Use).

Sources:

Cummins, J. (in press). Negotiating Identities: Education for Empowerment in a Diverse Society. 2^nd edition.

Los Angeles: California Association for Bilingual Education. (www.bilingualeducation.org; 213/532-3850 – available June 2001)

The Case for Bilingual Education Why Bilingual Education? by Stephen Krashen ERIC® Clearinghouse on Assessment and Evaluation ... read more. How effective is bilingual education?  Elizabeth Howard, Center for Applied Linguistics... read more. Visit James Crawford's Language Policy Web  Possibly the most in-depth bilingual education site on the web. Highlighted ESL Sites and Tools Instant Multi-Language Translator Great tool for the ESL writing workshop classroom.  SETTING EXPECTED GAINS for Non and Limited English Proficient Students Edward De Avila, Ph.D. Mathematics For Students with Learning Disabilities from Language-Minority Backgrounds: Recommendations for Teaching Diane Torres Raborn Rethinking Schools:  Online Urban Educational Journal. Dr. James Cummins is a leader in second language learning and literacy development research. This website is an expanding resource for educators the world over. Top of Page i teach i learn.com © 1999-2003 Educators. Technology. Connected.