DISABLING LANGUAGE: THE OVERREPRESENTATION OF EMERGENT BILINGUAL STUDENTS IN SPECIAL EDUCATION IN NEW YORK AND ARIZONA.

• Author: Hulse, Emma Curran Donnelly

Introduction 382 I. Disproportionality and Its Consequences 387 II. Restrictive Language Policies and Educator Discretion Drive Disproportkmality 393 A. Contextualizing the Classroom: The Rise and Fall of Federal Support for Bilingual Education 394 B. Arizona: The Nativist Campaign for English-Only Education 397 C. New York: The Collateral Consequences of Accountability 399 D. Educator Discretion Under Restrictive Language Policies 403 III. Patterns of Overrepresentation in New York and Arizona 407 A. Measuring Disproportionality: The Data Sets and Their Limitations 408 B. Methodology 412 C. Findings 414 i. Emergent Bilingual Students 414 ii. Students of Color 417 iii. Implications for Enforcement 421 IV. Towards Race-Conscious Remedies 422 A. Strengthening Transparency and Accountability 423 B. From Procedural Compliance to School Transformation 428 Conclusion 437 Appendix 433 INTRODUCTION

At the age of four, Luz seemed poised for academic success. The firstborn daughter of two Mexican immigrants, Luz was a spunky, assertive child who spoke in Spanish without hesitation. Both of her parents were deeply invested in her education. Her mother, Alicia, attended weekly classes on how to design at-home learning experiences for her children and received support from a trained facilitator through home visits. Luz regularly read with her mother and her father when he was home from his restaurant job. The spring before she entered kindergarten, she won a coveted spot at a high-achieving neighborhood school that had just opened in a new state-of-the-art building. The facilitator assured Alicia that Luz was well prepared for kindergarten and would thrive at her new school.

But by the middle of first grade, Luz's progress had ground to a halt. She was not advancing in reading or writing, and teachers reported that she was not engaged in classroom activities or discussions. At home, she refused to do homework or read in Spanish with her mother. Alicia reluctantly agreed to a special education evaluation to determine if Luz had a speech or language impairment (SLI). At the Individualized Education Plan (IEP) (1) meeting, both Luz's teacher and the school principal insisted that she required special education services and pressured Alicia to accept a placement in a self-contained classroom. (2) However, the psychologist who had conducted the evaluation had come to a very different conclusion: Luz was simply navigating the difficult process of acquiring academic English.

Parents and students like Alicia and Luz are not unique in the New York City school system. Mothers, fathers, grandparents, and guardians share a fierce commitment to their children and a willingness to fight to ensure they receive the education they deserve. But many of them face a system that lacks the institutional knowledge and resources to support their children's emergent bilingualism. (3)

The overrepresentation of emergent bilingual and Latinx students in special education is a civil rights issue, implicating two intersecting but distinct legal regimes: the federal Individuals with Disabilities Education Act (IDEA) and state laws regarding English Language Learners (ELLs). First passed in 1975, the IDEA guarantees children with disabilities a right to a free and appropriate public education in the least restrictive environment. (4) To qualify for services, an evaluator must find that a student has a disability encompassed in one of 13 categories and that this disability adversely affects their educational progress. (5)

While Supreme Court precedent (6) and the federal Equal Education Opportunity Act (EEOA) (7) require states to educate English learners, state law differs considerably on the rights of emergent bilingual students and the programmatic offerings available to them. The two states highlighted in this study--Arizona and New York--have divergent approaches to educating students learning English. Arizona has a uniform policy of Sheltered English Immersion (SEI), which places all students classified as English Learners in segregated classrooms for four hours a day with all instruction focused on English acquisition. (8) In contrast, New York law grants parents the right to select bilingual education for their children, the legacy of a legal battle waged by the Puerto Rican community in the 1970s that led to the so-called ASPIRA Consent Decree. (9) In recent years, accountability systems, which alternately reward schools when their students rapidly acquire English and penalize them when students fail to make adequate progress on state tests, have led to a precipitous drop in the number of bilingual programs. As a result, restrictive language policies guide both states' approaches to educating emergent bilingual students: in Arizona, an explicit policy, and in New York, an implicit one.

This Article explores how these restrictive language policies contribute to the overrepresentation of emergent bilingual students in special education. It further argues that in this legal and political context, the lack of adequate resources for educating emergent bilingual students limits educators' options when students require remediation. Because of the power of educator discretion in the special education identification process, deficit discourses about children, families, and bilingualism may lead teachers and evaluators to interpret a student's slow progress as resulting from an innate disability rather than conditions in the classroom. While race is rarely openly referenced in special education assessments, decisions are made in the context of highly contested debates over language, assimilation, and identity that are fraught with racialized meaning.

Indeed, outcomes in New York and Arizona are remarkably similar. This Article tests its hypothesis through an original analysis of data from the U.S. Department of Education's Office for Civil Rights (OCR). (10) While this data is publicly available online, it is underutilized by researchers and advocates. The analysis applies the statistical methodologies recommended by the IDEA Data Center in the federal Department of Education (DOE) (11) to show that emergent bilingual students are overrepresented in multiple high-incidence categories in both Arizona and New York, particularly in the SLI category. (12) In New York, for example, Latinx students are twice as likely to be identified as SLI. (13) In both states, Black and Latinx students are overrepresented in the Specific Learning Disability (SLD) and Intellectual Disability (ID) categories, while in Arizona, American Indian students are overrepresented in SLD and ID. (14) The OCR data also show that state thresholds for determining overrepresentation mask widespread disproportionate representation. (15) However, there is variation across school districts, suggesting that overrepresentation is not inevitable and can be addressed.

The extent of the overrepresentation of emergent bilingual students and Latinx students in special education requires immediate action. While legislators have taken steps to address disproportionality in special education and new regulations recently went into effect to standardize the methodology for identifying districts with overrepresentation, these changes are not sufficient to ensure transparency or accountability. Moreover, public enforcement has overly relied on procedural compliance, failing to address the underlying conditions that drive overrepresentation. As a result, stronger federal standards and enforcement, along with a renewed investment in transforming classroom practice, are needed to ensure that emergent bilingual students and students in special education receive an appropriate education.

This Article makes three unique contributions to legal scholarship on special education. First, the legal literature on emergent bilingual students and special education is extremely limited and does not discuss disproportionate representation. (16) This Article addresses this gap by braiding original data analysis together with interdisciplinary research on emergent bilingual students, restrictive language policies, and overrepresentation. Second, this is the first article to use the OCR data to study disproportionality: previous studies relied on data obtained through state governments, and as such the states in question were not explicitly named. This Article's findings demonstrate the vital importance of publicly available data, but they also identify some problems with the OCR data set that undermine its intended purpose of promoting transparency and accountability. Third, this Article demonstrates that even seemingly divergent language policies--in this case, a bilingual education ban in one state and explicit protections for emergent bilingual students in another--can produce similar patterns of overrepresentation. Ultimately, classroom practice matters and enforcement efforts must move beyond procedural compliance to incubate culturally responsive pedagogy.

While this Article includes data on American Indian, Black, and white students, readers might note that its analysis of racial data tends to focus on Latinx students. This is in part because Spanish-speaking students represent the majority of emergent bilingual students in New York and Arizona, and in part because this Article aims to address a gap in the literature: most studies of overrepresentation do not center on Latinx students. But while Spanish-speakers represent 75.6% (17) and 63.3% (18) of emergent bilingual students in Arizona and New York, respectively, emergent bilingual students are not exclusively Latinx. In New York in particular, the immigrant community is large and diverse. Statewide, 9.2% of emergent bilingual students speak Chinese, 4.3% speak Arabic, 2.7% speak Bengali, and 1.6% speak Russian. (19) There are also significant communities of Black migrants from Haiti and West Africa. (20) The Latinx label may mask intragroup...