Cracks in the Bell Curve: How Education Matters

Abstract

This article rebuts three major presumptions undergirding the arguments of The Bell Curve: first, that intelligence is largely inherited, fixed, and distributed unequally across groups; second, that it is represented by a single measure of cognitive ability (the g factor) that is predictive of life success; and third, that it is not substantially affected by education, health care, or other environmental factors. The author focuses especially on the substantial body of research demonstrating that (a) education makes a profound difference in attainment; (b) educational opportunities are more unequally distributed in this society than nearly any other; and (c) when students have equal access to high-quality curriculum, teachers, and school resources, disparities in achievement narrow sharply. INTRODUCTION In his history of 18th-century colonial education, Lawrence Cremin (1970) writes: For all of its openness, provincial America, like all societies, distributed its educational resources unevenly, and to some groups, particularly those Indians and Afro-Americans who were enslaved and even those who were not, it was for all intents and purposes closed. . . .For the slaves, there were few books, few libraries, [and] few schools. . . the doors of wisdom were not only not open, they were shut tight and designed to remain that way. . . .[B]y the end of the colonial period, there was a well-developed ideology of race inferiority to justify that situation and ensure that it would stand firm against all the heady rhetoric of the Revolution. (pp. 411-412) Ideologies such as those Cremin describes, developed to justify slavery and honed in the eugenics movement at the turn of the 20th century, have festered for decades. Recently, they have erupted anew in a contemporary representation of pseudoscience called The Bell Curve: Intelligence and Class Structure in American Life (Herrnstein & Murray, 1994), which, along with a resurgence of racialist thinking, has been received with a remarkable presumption of credibility. A major failing of The Bell Curve's analysis is the evidence it uses--and that which it ignores--regarding the distribution of cognitive abilities across racial and ethnic groups in the United States. Quite simply, Herrnstein and Murray rely on tests that are not good measures of intelligence, and they ignore research demonstrating that the racial achievement gap in performance on exactly those measures has been closing in recent years. Moreover, they seem almost wholly ignorant of the last 20 years of research on cognition, intelligence, and performance, on the effects of education on performance, and on the inequalities in educational opportunity that exist and affect performance. For example, on the Scholastic Aptitude Test, one of the tests Herrnstein and Murray inappropriately use as a measure of intelligence, the scores of African American students climbed 54 points between 1976 and 1994 while those of White students remained stable. This progress is directly related to school desegregation and school finance reform, twin efforts to equalize access to education initiated 30 years ago. Yet, The Bell Curve fails to address the critical influence of education on achievement. It is this failure that the present article addresses. Herrnstein and Murray's arguments rest primarily on three sets of presumptions about the nature and meaning of intelligence: first, that it is largely inherited, fixed, and distributed unequally across groups; second, that it is represented by a single measure of cognitive ability (the g factor) that is predictive of life success; and third, that it is not substantially affected by education, health care, or other environmental factors. The first two sets of presumptions about the nature of intelligence and the meaning of IQ scores have already been roundly rebutted in the months following The Bell Curve's publication, and the data to dispute them had been widely available long before then. …