Bergmann, Barbara. In Defense of Affirmative Action . New York: Basic Books, 1996: 42-7

There has been some progress: since the 1960s, African Americans and women have been gaining representation in some occupations to which they rarely had access in the past. However, the presence of blacks and whites, and men and women, in the same occupational groups does not mean that they all are on an equal footing when they are being considered for vacancies. Even when men and women have very similar jobs, they seldom are coworkers. Men who wait on tables generally work in expensive restaurants where the tips are high and no women are hired. Women tend to work in the cheaper restaurants, with no male colleagues.

Thanks to the work of the sociologist Donald Tornaskovic Devey, we have systematic statistical information about the extent of segregation by sex and race among coworkers in the same job for the same employer. In 1989, he asked a sample of North Carolina workers about the race and sex of their coworkers who did the same kind of work they did and had the same job title. He found that blacks and women were not excluded from any broad categories of work (see table 2.2). However, a considerable majority of people worked exclusively with people of their own sex. Jobs in which 100 percent of the incumbents were male or 100 percent were female together accounted for 70 percent of all jobs. In another 16 percent of all jobs, segregation by sex was not total but nearly so. Only 14 percent of the respondents held jobs in which males and females worked together in numbers approximating their share in the workforce.

The North Carolina survey found that segregation by race is less strict than segregation by sex. Still, a majority of the respondents (56 percent) worked in jobs that were totally segregated by race. Another 30 percent worked in jobs that were almost totally segregated. Only 15 percent of the jobs were shared by blacks and whites in numbers roughly proportional to their workforce presence.

The North Carolina survey results are broadly consistent with the findings of the Labor Department's Office of Federal Contract Compliance Programs (OFCCP), the federal agency that visits government contractors to inquire about their compliance with nondiscrimination requirements. The agency found that in 1994-95, 75 percent of the employers it checked were in substantial noncompliance....

People of color and women have, of course, been subordinated groups for a long time; as a result, differences in education, abilities, skills, and attitudes between them and white males have developed. These differences, however, are not so stark as to justify "on account of merit" the extreme degree of segregation that the North Carolina survey reveals. The least capable white available is not inevitably going to be better in a certain job than the most capable black available. The least capable man available will not always do better than the most capable woman available.

Since differences in the qualifications of women and men, and of blacks and whites, cannot account for the extreme degree of segregation the surveys turn up, we have to conclude that those who make job assignments are paying a great deal of attention to race and sex when they decide who is allowed to have which job. Qualifications such as education and ability are not ignored, but if a worker is female, not a lot of attention is paid to her suitability for jobs traditionally assigned to males with her education and ability. If a worker is African American, his or her suitability for a job traditionally assigned to similarly educated and skilled whites may not be noticed.

The purpose of affirmative action is to reduce segregation by race and sex in the workplace. Obviously, much of the segregation that affirmative action was designed to eliminate is still present. The argument that affirmative action programs have already accomplished so much that we no longer need any programs of this (or any other) type in the workplace cannot seriously be made by anyone who has examined the evidence of what is currently going on in the workplace.

Carl Gutierrez-Jones,

Department of English

University of California

Santa Barbara, CA 93106

E-mail: carlgj@humanitas.ucsb.edu