In Defense of Affirmative Action in Employment Policy

Theodore William Allen


Getting down to cases

      Let us assume that the following two principles represent sound public policy: 1) Racial discrimination in regard to employment is a bad thing; and, 2) hiring and promotion should favor the best qualified applicant for the position.

      Consider, then, some typical cases in which workers are to be hired, promoted, laid off or demoted:

     Case 1) An employer is of the same "race" as his son-in-law and gives preference to the son-in-law, even though there is another applicant of a different "race" who is better qualified for the position under consideration.

      Would that constitute racial discrimination? If not, at what distance of affinity--kinship, friendship, private referral, personal obligation, school (or union) tie--would the same sort of favoritism become racial discrimination?

      A study done in 1983 regarding job discrimination against Catholics in Northern Ireland noted that the still prevailing practice of hiring workers on the informal recommendation of friends and family of the prospective employee "tend[s] to reproduce the existing patterns of job distributions" as between Protestants and Catholics. The study, therefore, urged the adoption of the principle that the employer "should not recruit new employees on the recommendation of existing employees."1 Would the insistence upon such a policy in the United States help to reduce racial discrimination in employment and to favor the best-qualified applicants? Would that not be an acceptable form of affirmative action?

     Case 2) Applicant A becomes aware that Applicant B is better qualified for the job both are seeking.

      Should applicant A persist in trying to beat out applicant B for the job? If Applicant A does persist, is he or she not promoting discrimination in favor of the less qualified applicant? In so doing, is not applicant A, as citizen A, deprived of moral grounds for attacking affirmative action as a violation of "merit" principle?

     Case 3) In recognition of the effect that historical experience may have had in delaying or otherwise interfering with U. S. veterans in starting or maintaining their job careers, they are given extra points on civil service exams, as a matter of public policy. May a non-veteran justifiably challenge this veterans-preference policy on the ground that he or she was in no way responsible for the interruption of a veteran's career by induction into the military; and indeed may not have yet been born when that involvement occurred?

      If a challenge to veterans' preference on such grounds is rejected as contrary to public policy based on historical reasons, what weight can be given to a challenge to a public policy of affirmative action designed to make up for historical impediments to career advancement by African-Americans on the grounds that the complainant was not even alive when such impediments were imposed. I cite just two such examples of how the basis for today's persisting pattern of racial preference in employment was put in place a century and more ago.

      In the years 1840-65, "whites" drove African-American wage-workers out of longshoring, tobacco manufacture, carting, table-waiting where African-Americans had been regularly employed since the founding of the Republic.2

      In the late nineteenth century, when ninety percent of the African-American population of the country lived in the South, the "Cotton Mill Campaign" established that region's flagship manufacturing industry with an employment policy that deliberately aimed "to keep that avenue open to the white man alone," because "the white mill workers ought to be saved from negro competition."3

     Case 4) A quota of jobs for African-Americans leads to a complaint that such a policy is racial discrimination against "whites," that it disregards the need to reward merit, and that there is no overriding public interest served by quotas.

      What is it but white racism to reserve the criticism of quotas, goals, and timetables favoring opportunity for African-Americans, while ignoring other "quotas" that are, or have been, far more widely imposed and practiced?

      Take the quotas in the United States Constitution. Prior to the Civil War, the slaveholding States had a quota of additional representation in Congress, proportioned to three-fifths of the number of African-Americans they held in bondage.4 That quota made it possible for the slaveholding states to dominate the United States government from the 1789 to 1860.

      After the Civil War (by virtue of a provision of Amendment 14, but one that was in effect nullified by the Hayes-Tilden Deal of 1876), those same states were to have their Congressional quota reduced in proportion to the number of disfranchised African-Americans, thus diminishing the weight of the franchise of whites in those states.5

      Or, why is it that our "quotaphobes" can seem completely at peace with numerical quotas in our Constitution that absolutely disregard the question of merit for office, or deliberately negate the principle of one person one vote--quotas that are in full force to this very day? They are content that the United States Constitution in effect bars any two persons from the same state from serving together as President and Vice-President even if those two are the best qualified for those positions.6 They take no exception to the inequity of the Constitutional quota of two Senators per state,7 whereunder Wyoming, with a voting population of less than 200,000, gets two Senators, equal in national governing authority to the quota-limited two Senators from California, a state with a voting population more than 50 times that of Wyoming, thus diminishing the political voice of the California voter to a mere fraction of that of the Wyoming voter in this aspect of governmental affairs. (Substantial discussion of these matters may be left for another forum; they are cited here merely to draw attention to the "two-eyed-mule-one-eyed-argument" tendency of the opponents of affirmative action when it comes to the subject of "quotas.")

      More to the point of "racial preference," is the secret quota by virtue of which for nearly half a century, even by official government estimates, the chance of avoiding unemployment has been maintained at twice as great for "whites" as for "not-whites." When the Humphrey-Hawkins Bill passed in 1978, defining "full employment" as four per cent, the "White" unemployment rate was 4.5%, while the "Black and other" rate was 11.9%. In 1996, when "full employment" had been "achieved" by sheer redefinition as just under six per cent, the "White" rate was less than 5%, while the "Black" rate was over 10%.8 When a numerical ratio remains constant for nearly five decades, it is a quota; the failure of the opponents of affirmative action to acknowledge this instance of it shows the one-sidedness of their pretended concern with "doing away with quotas to avoid racial preference."

     Case 5) Tens of thousands of workers are to be employed by contractors in privatized service operations in large municipalities, and the contractors are bound to abide by principles of affirmative action, to assure African-Americans and Latinos a share of the jobs proportional to their presence in the labor pool, but the rule is denounced as a "quota" principle, by its very nature unfair to "whites," and a violation of the merit principle of employment.

      The "white ethnic" quota system, an integral aspect of big-city politics, driven by political job patronage, has prevailed in this country for more than a century. In their classic study, Beyond the Melting Pot, Nathan Glazer and Patrick Moynihan describe New York City, a typical case:

     Ethnic considerations have always been primary in New York City politics, where the three top spots of each party are regularly divided among a Jew, and Italian, and an Irishman....[and] the old Board of Education was regularly divided among three Jews, three Catholics and three Protestants.9

      Instead of denouncing such notorious practices, the authors call for more of the same, arguing that "the ethnic pattern offers the best chance for a humane and positive adaptation to group diversity."10 This facile and politically popular notion is not merely a justification of the historically white supremacist system of "ethnic preference," but also a proposal to guarantee that it be preserved as long as possible in the name of "humaneness." Certainly it should not be allowed to give shelter to opponents of affirmative action who resort to the "immigrant ladder" fiction of social mobility to sidetrack the question of racial discrimination against people, many of whom have family roots that go back to the American Revolution and beyond, and many other "not-whites" who became immigrants not because they crossed the border, but because the border crossed them in the course of "Manifest Destiny."


"Class-based preference" is no answer to racial discrimination

      Today, the "ethnic" ladder notion is amended by a theory of "class-based" rather than "race-based" government policy. A number of opponents of "racial preference" have advocated this approach. The argument is that "class preference" is color-blind and thus avoids unfairness to "whites," while, merely by the law of probabilities, "class preference" would at the same time "disproportionately benefit those who have suffered most under our nation's history of [racial or gender] discrimination...."11

      However well-intentioned this proposal may be for disposing of the contentious question of "race" by folding it into "class," analysis makes its fatal fallacy apparent. "Class preference" is the whole point of capitalist society; why strive to accumulate class-defining personal or corporate wealth if it does not carry with it corresponding political, economic and social privileges or "preferences"? What is proposed here is not merely apportioning the tax burden according to ability to pay, but altering social rank by the promotion of the those in the lower ranks relative to the those in the higher ranks. It is hardly to be expected that the strident opponents of even progressive taxation and government, among whom opposition to affirmative action is almost universal, will be persuaded to take this supposedly alternate route to racial equality. Nor can they be expected to enlist as supporters of "reverse class discrimination" whereby those in each quintile of wealth-holders take precedence over the hard-earned(?) "class preference" of the one above, on the basis of inverse relative rankings, as determined by layers of bureaucrats, with respect to parental wealth, income and occupation; the quality of educational opportunities; and the stability of family structure.12 Let it be noted in passing that, contrary to the assumption made by its advocates, progress along this line would diminish the impact on racial discrimination because it would reduce the proportion of African-Americans in the favored categories, even as the general competition for the benefits of "reverse class discrimination" was intensified.

      Proponents of this scheme of reverse class discrimination suggest that, despite such opposition, a politically favorable constellation could be achieved by quintile-splicing until over half the population were included among the favored class.13 This strategy bears the same Achilles heel that has foredoomed leveller programs of the past, the white blindspot, the denial of the special character of racial oppression of African-Americans.

      Those in the United States to whom it has been given historically to decide such matters have found it expedient to have "class preference" modified by "white-race preference." They have thereby established and maintained a form of oppression distinct from class oppression, namely, racial oppression. The hallmark, the informing principle of this "peculiar institution" is not the social preference of "whites" in a given socio-economic quintile over African-Americans in a lower quintile, but over African-Americans of the same or higher socio-economic quintile. It is precisely this prevailing social anomaly that advocates of "class-based preference" refuse to take into account.14

      Whenever the politics of wealth redistribution--whether "land to the tillers," "populism," "New Deal," or "War on Poverty"--has been proposed, and, in some cases, actually attempted, the "race card" has proved to be trumps, played as "Negro domination," "Dixiecratism," and outcries against "welfare queens," and "poverty pimps."

      The current advocates of the "reverse class discrimination" strategy of social reform refuse to acknowledge the fact and the nature of racial oppression of African-Americans. It is precisely because of this attempt to avoid the issue of "race" that the "reverse class discrimination" strategy is doomed to the fate of earlier movements for fundamental social reform in this country. The multi-quintile alliance would be inescapably doomed soon after our radio talk show hosts discovered the first instances in which African-Americans have been given preference over "white" applicants in a higher socio-economic quintile. Race-based affirmative action is an essential consciousness-raising pre-conditon for making it possible to assure social reform in the general interest of the lower socio-economic quintiles.


Affirmative action is not a barrier to, but a necessary condition for selecting the best qualified applicant

      It is to be hoped that such arguments will serve to put in perspective the complaint that affirmative action brings about "racial preference" for "not-whites," and thus unfairly discriminates against better qualified "white" competitors. Being of such very recent vintage, such "White" concern for the centuries-old phenomenon of racial preference is rendered suspect. The same may be said of the sudden espousal of the principle of "racial equality of opportunity," merely for the partisan purpose of opposing it to "racial equality of result." That common argument necessarily rests on the assumption of, or at least conveys the suggestion of, "white" racial superiority.

      As a matter of American public policy broadly considered, affirmative action--obstructing racial discrimination against African-Americans and other "not-whites," and gender discrimination against women--is not a barrier to assuring that the best qualified person will be employed, but rather a necessary condition for achieving that result. It should not be discouraged, but made ever more effective. It is far less likely to result in relative merit deficiency than preferential hiring by family connection, veterans status, or by white "ethnic" category, or by "old boy" connections.


Affirmative Action as a Civil Right

      But, even when the opponents of affirmative action lose every argument on the merit-and-fairness issue, they shift the burden of proof. In this way they avoid having to deny the facts of "patterns" of discrimination, historical or contemporary; they simply deny that such patterns are necessarily relevant, except where individual complainants individually can prove deliberate discrimination by the defendant.

      This turn in the argument contains a challenge to even wider issues of public policy than just affirmative action. It assaults the basic principle of civil rights law as represented by Brown v. Board of Education, the Voting Rights Act of 1965, and the rules for preventing racial prejudice in the constitution of jury panels. According to that principle, the underrepresentation of African-Americans, for example, in the make-up of school populations, voting populations, and jury pools, constitutes prima-facie evidence of racial discrimination, and places the burden of proof on the accused supervisory entities in such instances. Thus the enemies of affirmative action remind us that affirmative action is an integral part of the general cause of civil rights; to retreat on this issue is to unravel the fabric of the hard-won gains of decades of struggle against racial and gender discrimination. Therefore, let this be our resolve: Not one step backward!


Return to Allen Interview



1 R. J. Cormack and R. D. Osborne, eds., Religion, Education and Employment: Aspects of Equal Opportunity in Northern Ireland (Belfast, 1983), pp. 160, 218.

2 See Albon Man, "Labor Competition and the New York Draft Riots of 1863," Journal of Negro History, October 1951, pp. 375-405.

3 Broadus Mitchell, The Rise of the Cotton Mills in the South, pp. 27-28, 136-37. Reports of the U. S. Industrial Commission (1901), vol. 14, p. lii.

4 Article 1, Section 2, paragraph 3.

5 Article 12, paragraph 3.

6 Amendment XII. The electors of each state are barred from voting for natives of that state for both positions.

7 Article 1, Section 3, paragraph 1.

8 See Tables on unemploymet rates in successive editions of the annual Economic Report of the President (Government Printing Office, Washington, D. C.); and U.S. Bureau of Labor Statistics, Monthly Labor Review, February 1997, p. 53.

9 Nathan Glazer and Daniel Patrick Moynihan, Beyond the Melting Pot: The Negroes, Puerto Ricans, Jews, Italians, and Irish of New York City, second edition (Cambridge, Massachusetts, 1970), pp. xxvii-xxviii.

10 Ibid, p. xxiv.

11 Richard D. Kahlenberg, The Remedy: Class, Race, and Affirmative Action (New York, 1996), p. xii. This approach, it is said, has the incidental benefit that, "because blacks are disproportionately poor, they would [[under "class- based" preferences]] still continue to reap a disproportionate share of the benefits, but without engendering resentment or feelings of injustice within the white community" (The Remedy, jacket blurb). In the discussion between opponents of affirmative action and President Clinton and Vice-President Gore on 19 December 1997, Linda Chavez makes a congruent argument ("Excerpts From Round Table with Opponents of Racial Preferences," New York Times, 22 December 1997). However, another participant, Abigail Thernstrom, author with her husband Stephan Thernstrom of America in Black and White: One Nation Indivisible, deplores even this sort of concession to remedies for social inequities, as an example of "victim status creep" (Abigail Thernstrom, "A Class Backwards Idea: Why Affirmative Action for the Needy Won't Work," Washingnton Post, 11 June 1995, C1, C2, cited by Kahleberg, p. 87).

12 See Kahlenberg, The Remedy, pp. 123-36,

13 Kahlenberg, ibid., pp. 139-43, 202-3.

14 This fault, it seems to me, informs Kahlenberg's entire book. But see particularly: pp. xii ("Class preferences are color-blind."); p. 45 "[A]cknowledging that the legacy of past discrimination [does not mean] that the most advantaged black applicant is more disadvantaged than the poorest white."); p. 261, n. 108 (Kahlenberg favors temporary resort to protect "proportionality of job opportunities," but "[t]o address broader past societal discrimination, class-based preferences would provide the remedy.")


go to this issue's index

go to the current issue's index




Contents copyright © 1998 by Theodore W. Allen.

Format copyright © 1998 by Cultural Logic, ISSN 1097-3087, Volume 1, Number 2, Spring 1998.