In his column "Proposition 209: Wrong for CA and Us" [Nov. 15], Scott Jacobs criticized the recent passage of Proposition 209 in California, which will prohibit preferential treatment on the basis of "race, sex, color, ethnicity, or national origin in the operation of public employment, public education, or public contracting."
Although I appreciate his spirit in endorsing the pro-affirmative action position, I was compelled to write this column to evince a number of problems in his argument. My intention in undertaking this task, despite the fact that I essentially agree with his position, is to prevent the conservatives from further attacking the liberals as idealists without much intellectual rigor. I hope to present at the end of this column a better argument in support of affirmative action.
First, Jacobs often seems unable to distinguish the attributes of the socioeconomic status from those of the race. Nowhere in Proposition 209 is there a statement that bars the government from helping the socioeconomically disadvantaged individuals; yet, he suggests that one feasible consequence of the bill is elimination of "scholarships that allow disadvantaged students to attend Dartmouth."
Second, Jacobs informs the readers that his life as a white man so far has been free of the infringement that Caucasian males reputedly experience, according to the criticizers of affirmative action. Any person who has some knowledge of statistics would explain that there is a serious sampling problem with his empirical evidence. To invoke just one counterexample to a general trend is absurd.
Third, the inability to characterize correctly the counterfactual also inhibits sound reasoning in Jacobs' column. Observing that 82 percent of the first-year class at Dartmouth are Caucasians, he claims that it is ridiculous for them to cry out that their opportunities have been hampered. One has to consider the strong possibility that white students would have constituted an even larger proportion of the population at the College, if it were not for affirmative action.
Having shown the problems of Jacobs' argument, however, I intend to present an alternative model that shows the necessity of affirmative action. Consider a world composed of two types of people, black and white, whose population ratio in every institution is two to eight. In essence, this is a perfect world in which there are no differences in initial opportunity endowments between the two races. Let us imagine that every individual of a firm in such a world is involved with the decision-making to hire employees. The firm, according to this model, always selects one employee after reviewing two applicants at the same time, and immediately after being employed, the new member of the firm has a voting power in the selection process of applicants seeking positions after him.
Now I write the crucial assumption of this model: whites feel more comfortable working with whites, while blacks feel more comfortable working with blacks. This leads to the consequence that when two candidates of the same qualifications, but of difference races, are reviewed, everyone votes in favor of the one representing his or her race.
The implication of the model is as follows. If, by chance, the first applicants to be employed are white, then the chance of employment by another white person increases slightly. As a result, while the ratio of retirees remains eight to two between whites and blacks, the racial mix of the firm quickly degenerates into such that every employee is white. The pressure in the direction of the other extreme is also plausible. If enough blacks are hired in blocks by chance, the firm could end up entirely black.
The reason for this transformation is that the initial racial mix is an unstable equilibrium. And the significance of this model is that, even in a world where the socioeconomic differences between the two races have disappeared, it might be necessary to maintain affirmative action to ensure the ideal racial mix in various institutions.
One must always question the assumptions when one encounters a model. In this case it is important to ask if the assumption that members of a particular race tend to prefer most interaction with other members of that race is a valid one. My personal experience tells me that the assumption is strong; I do not think it is merely an accident that the vast majority of my friends are Asians, for instance. And I am not actively seeking Asian friends, in particular. Look around yourself. Do you tend to find yourself often with people of your ethnic background? I think so.
Perhaps my conclusion is a disturbing one. But it follows naturally from human nature.
