INTRODUCTION This article focuses on one of the more detrimental outcomes of the complex web of social forces that mitigate against African American males in school and society: their all-too-frequent placement in special education programs. Special education was mandated by law in 1975 with the passage of the Education for all Handicapped Children Act (EHA), now renamed the Individuals with Disabilities Education Act (IDEA). The purpose of the EHA was to provide specialized services to students who, by virtue of disability, could not profit from regular educational curriculum and instruction. The law was intended to regulate and extend to all children, regardless of handicap, the provision of special education services that already existed in various forms across the country. By the time it was enacted, however, evidence of misuse of this umbrella construct was already visible, as reflected in the EHA's requirement that assessment for special education purposes be nonbiased and conducted by a multidisciplinary team. Despite widespread concern, this requirement has proven to be extremely difficult to implement. Racial, gender, cultural and linguistic biases remain integral aspects of the special education process, particularly for African American males. In this article, we argue that the entire process is seriously biased against African American male students, from their first experiences in regular education through their disproportionate referral to, assessment for, and placement in special education programs. We begin with a brief overview of the history of this phenomenon and conclude with recommendations reflecting our belief that the inappropriate designation of educational disability has been created by schools and must be dismantled by schools. The disproportionate placement of students of a particular group in special education programs means that the group is represented in such programs in a greater percentage than their percentage in the school population as a whole. The issue is the relativity of placement, not absolute numbers. In assessing disproportion, then, one must first note the percentage of a given group in the population as a whole and then compare it to the percentage of that group in the special education program. Chinn and Hughes (1987) define disproportion as plus or minus 10% of the percentage that would be expected on the basis of the school-age population. For example, if African Americans account for 16% of the U.S. public school enrollment (U.S. Department of Education, Office for Civil Rights, 1994), then one would expect the special education enrollment to fall within a range of plus or minus 1.6% of the total enrollment; hence, any special education placement percentage outside of the range from 14.4% to 17.6% would be considered disproportionate. Another important aspect of this phenomenon is that historically it has been concerned mostly with what are referred to as the categories of disability-that is, the milder disability categories whose diagnoses are based essentially on clinical judgment rather than verifiable biological criteria. The centrality of the role of judgment makes it a potential source of bias. The disability categories considered to be most susceptible to bias are Educable Mental Retardation (EMR), Serious Emotional Disturbance (SED), Specific Learning Disability (SLD), and Speech Impairment (SI). Another category susceptible to bias that reflects more severe disability is Trainable Mental Retardation (TMR), which historically has been used to refer to moderate rather than mild mental retardation. For several decades, the U.S. Department of Education's (USDOE) Office for Civil Rights (OCR) has been charged with the responsibility for monitoring disproportion in all these categories.1 While the present article focuses on issues related to these judgment categories, identification of a disproportionate number of African American males has also been noted in recent years in the categories of deafness, visual impairment, and other health impairments (USDOE, Office of Special Education Programs [OSEP], 1992b). …