Black and White Together:  A Reconsideration*



W. B. Allen


Harvey Mudd College

Claremont, California


Conference on Reassessing Civil Rights

of the Social Philosophy and Policy Center

Bowling Green State University


Principled discussions of civil rights became inherently less likely as a direct result of the observation by Earl Warren, in Brown v. Board of Education,[1] that, respecting freedmen, “Education of Negroes was almost non-existent, and practically all of the race were illiterate,” and in proportion as that observation increasingly became the foundation of common opinion on the subject.  Warren’s observation was not true in any meaningful or non-trivial sense.  Nevertheless, it served to perpetuate the myth of a backward people needing help to catch up instead of the truth of a people being held back.  That is the perspective—the disadvantaged group perspective—that ultimately infected all discussion of civil rights, even after the designation of so-called “disadvantaged groups” had been extended beyond American blacks.

To define civil rights we may well begin with what all mankind would likely recognize.  Thus the dictionary definition of “civil rights” stands:  “the rights that belong to all individuals in a nation or community touching property, marriage, and the like.”  In that definition the term “rights” may be further expanded to mean “legitimate claims,” following the definition of right as law—as “a claim or title or interest in anything whatever that is enforceable by law.”[2]  This definition applies with minimal distinction of regimes intruding and therefore without the host of recent complications in the United States that create the impression that civil rights have somewhat to do with pluralism.  Previously, the generic definition was thought to exhaust the meaning of the term in the United States.  Witness James Wilson’s pithy version of the early 1790s:

Under civil government, one is entitled not only to those rights which are natural; he is entitled to others which are acquired.  He is entitled to the honest administration of the government in general:  he is entitled, in particular, to the impartial administration of justice.  Those rights may be infringed; the infringements of them are crimes.[3]

When distinguishing between natural rights and acquired rights, the latter coming to be for the sake of the former, we readily discern a “fair play” formula that is well captured by an ethic of non-discrimination on the part of public officers.  In the context of racial tensions in the United States, the ethic of non-discrimination came to assume particularly the form of prohibiting to public officers recourse to race in the performance of their duties.

The ideal of holding race irrelevant in the “administration of government in general” and in the “administration of justice in particular” eventually spawned parallel concerns respecting gender, religion, ethnic background, age, and physical or mental handicap.  This proliferation did not fully occur, however, until after the equal opportunity standard implicit but fundamental in the ethic of non-discrimination had been challenged by an implicit equality of results standard.  That challenge appeared most openly and decisively in President Lyndon Johnson’s commencement address at Howard University 1965.  In that speech, Johnson maintained that the equality of opportunity defended only the year before in the landmark “1964 Civil Rights Act” was not enough.  Black people in particular, he maintained, required positive efforts on their behalf in order to enable their enjoyment of the rights otherwise enjoyed by full citizens.

In light of the definition offered by James Wilson, one could properly inquire what more than an ethic of non-discrimination is needed for any individual to enjoy the honest administration of government in general and the impartial administration of justice in particular.  That question would expose Johnson’s premise of equality of results and also an unannounced re-defining of civil rights to include an active role on the part of government to produce that equality of result.  That is by now far more nearly the operational definition of civil rights in the United States.

Before one makes that leap and simply dismisses Johnson as either a crank or a poor student of logic, however, one ought at least to explore a hint contained in the generic definition employed by Wilson that serves to explain how the segue from equality of opportunity to equality of results could understandably have occurred.  The hint in Wilson’s definition is his declaration that the “infringements” of these acquired rights are crimes.  Wilson does not seem to mean that these are statutorily declared crimes, although he proceeds to give examples of how American statutes provide for the punishment of these crimes.  He seems rather to mean, along with Justice Chase,[4] that once government is instituted to secure natural rights, there devolve not only acquired rights (the necessary means to secure natural rights) but substantive, enforceable claims against violations of these acquired rights whether acknowledged by government or not.  Naturally, full and proper acknowledgement is the vaccination against revolution.

Since the acquired rights are means, or practical activities and consequences, it follows that one ought to expect to see practical consequences flow from their being guaranteed or acknowledged.  The question for analysis, ultimately, is what those practical consequences should be.  In the context of a discussion of the contemporary definition of civil rights, I think it plain that President Johnson thought the practical consequences of protecting civil rights ought to be specifiable advances in educational and material attainments for black people.  To Wilson’s anticipation of the abuses of public officers, Johnson then adds the tacit caveat that private parties may no less, and perhaps still more, infringe these rights.[5]  Now, civil rights aim to improve the status of one or more private groups within a community relative to the status of a reference group, also within the community. Pursuing such a goal, President Johnson quite correctly could not see how the mere ethic of non-discrimination could produce it, since that ethic expressly left people precisely where they were, save for the added guarantee of an honest administration of government in general and an impartial administration of justice in particular.  Accordingly, he redefined civil rights in such a way as to render the term no longer applicable to “all individuals in a community.”

Where one expects a civil rights policy to augment the status and accomplishments of designated groups within a society, it necessarily follows that the policy’s practical objectives that will be pursued will vary in proportion as the relative status and accomplishments among all groups within the society vary.  It will further (if less obviously) follow that assessments of progress in civil rights will take on the zero-sum characteristics of unregulated competition for finite resources.  That is, more civil rights for some will mean less or fewer civil rights for others, to the precise extent that civil rights in general may properly be regarded as a matter of relative status within a finite universe.

By contrast, a principled discussion of civil rights would emphasize common terms of identity rather than difference among rights holders.  On such terms it would remain possible to observe that certain persons fail to enjoy the prescribed powers or privileges, but the specific reparation called for would be to restore common identity.  Ordinarily, the term of common identity would be citizenship—e.g., American.  A principled discussion of civil rights would be a discussion of the powers and privileges characteristic of American citizenship, such that we could not properly call someone an American citizen without simultaneously and rightfully confirming in him the self-same powers and privileges.

The disadvantage of the principled discussion of civil rights is that it discourages legislated amelioration of the circumstances of life (that is, life claims) for those citizens enjoying, but not fruitfully applying, the powers and privileges of citizenship.  One might argue the theoretical impossibility of persons enjoying, but not profiting by civil rights, as fairly elaborated.  That would be a secondary argument, however, resting on the adequacy of individual initiative.  Even if such an argument were theoretically correct, it could remain the case that theoretical propriety does not serve adequately to dispel unrealistic fears.[6]  Accordingly, the principled discussion of civil rights, the primary argument, must advance as well to the admonitory level at which it conjures literal obstacles to the enjoyment of civil rights—obstacles that derive expressly from attempts toward legislated amelioration of life claims.  In other words, an ethic of non-discrimination requires to be defended, not only as able to repair the past injuries of persons harmed by discrimination but also, as a positive and necessary requirement for anyone’s enjoyment of civil rights.[7]  Every exception to the principled argument must be seen as malign.

To speak of reassessing civil rights, therefore, is to raise the possibility of re-directing policy discussions with a decidedly practical turn of mind but nonetheless departing from a principled foundation.  The question is:  can discussions of civil rights in the United States ever transcend the inherent and debilitating dynamic of racialism (or cognate limitations)?  To answer this question with reference only to the resources customarily employed, one would have to respond, pessimistically, “No!”  In what follows, however, I will try to suggest an alternate course.  I will review only the most recent discussions—centering on newly introduced legislation—in order to elaborate the prospects for optimism.

On February 7, 1990 Senator Edward Kennedy introduced in the United States Senate the “1990 Civil Rights Act.”[8]  This resolution initiated the process of “correcting” or “overruling” a series of 1989 Supreme Court decisions that provoked stridently expressed and widespread fears of retrograde tendencies in civil rights.  Earlier anticipations of the Kennedy resolution had uniformly threatened to ignite debate over racial preferences—the consummate policy choice of those who employ the “disadvantaged group perspective” (that is, the view that civil rights provide protections exclusive to so-called disadvantaged minorities).  The Kennedy bill, however, sought to defuse this debate by means of two devices.  First, he included an express disclaimer:

Nothing in the amendments made by this Act shall be construed to affect court-ordered remedies, affirmative action, or conciliation agreements that are otherwise in accordance with law.

I will assess the legal value of this disclaimer below, but here we may at least note its relevance in the policy context:  insofar as any of the Court decisions may be conceived to establish the principle of color-blindness or common identity, this bill does not aim to challenge it!  Secondly and far more importantly, however, the bill does not at all touch the one case that raised the greatest outcry—namely, Richmond v. Croson.[9]  There the Supreme Court disallowed a plan by the City of Richmond that reserved a set portion of city construction contracts for firms owned by minorities.  That case involved racial preferences directly—“business set-asides”—and to the extent that the erroneous view prevailed, of the Court having invalidated such preferences on the merits, one would have expected an attempt at congressional remedy (especially since the Court expressly opened that door).  Now the interpretation through most of 1989, both of advocates and opponents of racial preferences, was precisely that the Court had struck down such preferences at least in municipal and state governments. It was, therefore, a great surprise that Congress did not challenge Croson, and I submit that the reason was precisely the desire to avoid, to the extent possible, a debate over the question, “Whose civil rights count?”

To place these matters in proper perspective, I need to return to an argument I made previously, concerning the case, Runyon v. McCrary,[10] which was invoked as precedent in one of the 1989 decisions, Patterson v. McLean Credit Union.[11]  The principles involved in the reconsideration of Runyon point most dramatically to the reassessment underway in the country, which yet remains only tacit.




Forty years ago, the Supreme Court aimed to redeem the Reconstruction era’s civil rights promises of eighty years before.  In Shelley v. Kraemer,[12] a case involving an attempt to enforce a racially restrictive covenant in a residential real estate transaction, the Court interpreted one of several closely related legal provisions with a brief, straightforward simplicity that disguised its  far reaching implications.  The objective was to refurbish that old-fashioned Anglo-American eagle, liberty of contract.  That standard was hoisted in the middle of a modern era which had seemed anything but friendly to liberties so defined.  In this case the Court struck down racially restrictive covenants largely on the basis of the 14th Amendment’s guarantee to individuals of a liberty of contract. Since that auspicious beginning, however, the principle has fluttered listlessly.


            The Shelley decision lay there, its far-reaching implications largely neglected.  Meanwhile, the language of “equal protection” vaunted to the forefront through the school segregation cases; a heavy emphasis was placed on the comparative conditions of groups.  This tendency reached its zenith in 1954, when the Court declared, in Brown v. Board of Education,[13] that an “equal protection analysis” rendered a “due process” analysis superfluous.  The focus on the Fourteenth Amendment’s equal protection clause suggested not only a focus on groups but also a backing away from the enforcement of individual claims such as liberty of contract. That is the reason, perhaps, that so little attention was paid in Brown v. Board of Education to the question of whether the plaintiff, Linda Brown, actually acquired the right to attend the “white” school in Topeka, Kansas, while so much more attention was paid to the question of whether black children in general were segregated or integrated.


            Twenty years after Shelley, however, the Court returned to the liberty of contract, or at least near to it, in Jones v. Alfred H. Mayer.[14]  There, an old statute proscribing discrimination on the basis of race or color in the sale of real estate received new life.  Unlike Shelley, though, Jones was content to found the strength of this right of contract on Congress’s evident authority to enforce the provisions of the Fourteenth Amendment.  Thus, the broad language of §1982 of the original “1866 Civil Rights Act” (recodified in 1871) was not treated to the same broad development that had struck down restrictive covenants.  Nevertheless, insofar as the Reconstruction Congress intended even in §1982 such results as Shelley accomplished, then Jones kept alive the prospect for a fulsome reading of the right of contract as “fundamental to a scheme of ordered liberty.”  All that was lacking was a Court reading of the right as resident not only in the Fourteenth Amendment but in the Constitution itself, and as subject to Congress’ enforcement authority either piecemeal (as in the case of real estate) or altogether.


Another ten years, after Jones, the Court began the process which promised finally to realize all these implications—in a case involving private education, of all things.  That is, the Court was offered, and accepted, the opportunity to revisit not only Shelley but also certain peculiarities infecting Brown.  This opportunity was presented in Runyon v. McCrary.  It is important here to notice how the Court dealt with this opportunity, in light of Shelley, and what remains to be done, in light of the hesitations of Runyon.


The decision in Shelley turned on narrow factual considerations but had broad effects.  The leading prior precedent was a case which was defective in its fact basis.  The Court maintained in Corrigan v. Buckley[15] that restrictive covenants challenged in the District of Columbia survived because “there was no showing that the covenants, which were simply agreements between private property owners, were invalid.”  Whereas in Shelley, when restrictive covenants were again challenged, the issue had ripened to “raise the question of the validity, not of the private agreements as such, but of the judicial enforcement of those agreements.”


That consideration explains the Shelley Court’s determination not to discuss the statute, §1982, where a test of reasonableness would apply and raise the question of merits directly.  Instead they focused on the 14th Amendment, where it sufficed to find a right and apply it.  Here the right of contract/property is treated correlatively to the freedom of religion, where the Court is indifferent to the doctrinal content.  They cited the Amendment’s framers:


Equality in the enjoyment of property rights was regarded by the framers of that Amendment as an essential pre-condition to the realization of other basic civil rights and liberties which the Amendment was intended to guarantee.[16]


That appraisal provided the foundation for the Court’s interpretation of the 14th Amendment as proscribing discriminatory state action touching “the right to acquire, enjoy, own and dispose of property,” even without a statute.

In Shelley actions to acquire or dispose of real property—fundamentally private agreements—run afoul of the Fourteenth Amendment only if “imposed by state statute or local ordinance.”  It follows, too, that “participation of the state courts in the enforcement” of the discrimination so defined is equally state action.  Indeed, for the Court the situation ceased even to be the mere protection of ex-slaves, though their fate was the specific motivation for the legislation.  In order to benefit the ex-slaves it is the right itself that requires to be secured; for it benefits them not only as they exercise the right directly but also in proportion as its direct exercise is guaranteed to every other individual.

The precise question before the Court in both the Buchanan and Harmon cases involved the rights of white sellers to dispose of their properties free from restrictions as to potential purchasers based on considerations of race or color.[17]

In short, these protections were fully reciprocal, no less guaranteed to persons of color seeking to acquire property than to white sellers.  Otherwise, the right could be defeated by merely inverting the restriction.

With this bridgework laid toward its eventual refusal to enforce restrictive covenants, the 1947 Shelley Court was then in a position to give due credit to the “state action” language of the 1883 Civil Rights Cases.[18]  It held, accordingly, that the first section of the Fourteenth Amendment inhibits “only such actions as may fairly be said to be that of the states.”  It establishes no “shield against merely private conduct, however discriminatory or wrongful.”  While the “purposes of those agreements are effectuated” by voluntary means, there is no violation.

It would be a mistake to imagine that the Court invented the distinctions employed here. They were already the bases of the Civil Rights Cases, properly read.  More importantly, these principles were elaborated in Ex Parte Virginia:  “A state acts by its legislative, its executive, or its judicial authorities.  It can act in no other way.”[19]  Here, we find the Court here accepting a theoretical construct as determinative of its reading of the law, for the distinction between the three branches of government is not itself a legal distinction.  Contemporary disputes over what to make of administrative agencies, regulatory agencies, and other hybrid bodies make practically clear that the idea that the state “can act no other way” is rather a theoretical than a juridical concept.  What is important, however, is the connection between the argument from the separation of powers and the argument from limited government.  The Court, by using such an argument, confirmed that it sought—in the language of state action—not a positive but an intrinsic restraint on the power of the state (here meant generically).  The Shelley Court looked no less for generic restraints and therefore reflected, when citing the Civil Rights Cases, that the language of that case—namely, “laws, customs, or judicial or executive proceedings”—recurred no fewer than eighteen times within the opinion.

The Court’s professed belief is that the right of contract is but the obverse of the right of property.  To the Court, the “difference between” enforcing the restrictive covenants or not doing so “is the difference to petitioners between being denied rights of property available to other[s]… and being accorded full enjoyment of those rights on an equal footing”—that is, an equal opportunity.  This conclusion may appear to be based on theoretical appreciation of the dynamics of free markets.  It is necessary to insist, however, on the facts which generated the Shelley case in the first instance—namely, the factual operation of the market to perfect contracts between whites and blacks and which could only be resisted by means of impermissible legal impediments!  We are talking of cases in which black folk had not only acquired but actually moved into residences designated for permanent segregation.  Accordingly, even in the absence of statutes affirmatively aimed to perfect the right, there clearly existed principles and legal avenues sufficient to eliminate residential segregation and, one may well believe, every significant discriminatory contractual relation.

Against the retort that equal protection requires nothing more than that whites be no less liable to exclusion than blacks, the Court’s rejoinder was decisive:

The rights created by the first section of the Fourteenth Amendment are, by its terms, guaranteed to the individual. The rights established are personal rights . . .  Equal protection of the laws is not achieved through indiscriminate imposition of inequalities.[20]

I hope that I do not have to point out, here as throughout this discussion, what powerful language this is from the perspective of the defense of liberty.  Again, then, to shut the Court to attempts to validate discriminatory private contracts is not a denial of equal protection, for the Constitution empowers no individual “to demand action by the State which results in the denial of equal protection of the laws to other individuals.”

The Court has been emphatic:  it was not the statute but the Fourteenth Amendment that vested property interests and a freedom of contract in this special, robust way, and which no statute could violate, not even residually through enforcing private agreements (the fate of private affirmative action plans would be implicated here).  Because of this reasoning, it is clear that the right protected is far broader than the right to purchase real property.  This brings us to Runyon.

All possible contracts are covered, rendering discriminatory private contracts effective only insofar as they need not rely on the power of the state for their validity.  In Jones v. Alfred H. Mayer the Court had refused to enforce contracts requiring racial discrimination in the sale of housing.  They reacted similarly in handling state anti-miscegenation statutes, though in that case they misapprehended the principle involved.[21]  The true principle, of course, had been enunciated when the Court, in 1954, needed to find a way out of its infelicitous separation of equal protection from due process because it was confronted with public school segregation not in states subject to the Fourteenth Amendment (as in Brown v. Board of Education) but in the District of Columbia.  The nation’s capital is not a state and therefore not subject to the Fourteenth Amendment proscription, “nor shall any state . . . deny to any person within its jurisdiction the equal protection of the laws.”  To eliminate school segregation in the nation’s capital the Court resurrected the notion of due process with some fancy footwork:

Liberty under law extends to the full range of conduct which the individual is free to pursue, and it cannot be restricted except for a proper governmental objective.[22]

The lingering question from Runyon, resurrected in Patterson v. McLean Credit Union,[23] was precisely how to realize the promise of liberty thus judicially defined—how to extend it beyond housing contracts and to the “full range of conduct which the individual is free to pursue.”[24]  As we shall see, I believe that the Court allowed this opportunity to slip by.




In light of this discussion, it is now appropriate to review all of the 1989 Court decisions, before looking at the “1990 Civil Rights Act” in detail.  There finally surfaced in the 1989 spring Court term something resembling the long-awaited conservative majority—a gang of five.  It is not always a self-consistent gang, and sometimes its internal tensions are more interesting (and more threatening) than differences with the standard dissenters.  That is certainly true of tensions between O’Connor and Scalia (with perhaps Kennedy) over the question of how color-blind the Constitution is.  But, in any event, we can point to five major decisions carried by the gang of five in the 1989 term, and in which each one of the five took a turn speaking for the Court:  O’Connor in Richmond v. Croson, White in Wards Cove v. Atonio,[25] Scalia in Lorance v. A.T.&T.,[26] Rehnquist in Martin v. Wilks,[27] and Kennedy in Patterson v. McLean Credit Union.

The initial reactions to Richmond v. Croson implied, when they did not overtly assert, that the Court meant to eliminate race as a policy foundation in our society.  While Justice Scalia most eloquently demonstrated that potential result in concurring with Justice O’Connor’s decision to retain “strict scrutiny” even in benign discrimination cases by states and municipalities, surely the best evidence of what the future may hold is found in Justice Stevens’s concurrence. There, he elaborated a rule that, by the time of Martin v. Wilks, he completely repudiated.

Dealing with the City of Richmond’s reservation of 30 percent of public works contracts for minority-owned businesses (racial set-asides) as a remedy for generalized but not documented prior discriminations, Stevens held that “it is only habit, rather than evidence or analysis, that makes it seem acceptable to assume that every white contractor covered by the ordinance shares in that guilt.”  Further, Stevens added, to impose “a common burden on such a disparate class merely because each member of the class is of the same race stems from reliance on a stereotype rather than fact or reason.”  These remarks called into question the policy of assigning benefits to non-victims at the expense of innocent persons. Nevertheless, that is the very policy Stevens inconsistently affirmed in his later Wilks dissent:  “Just as white employees in the past were innocent beneficiaries of illegal discriminatory practices, so is it inevitable that some of the same white employees will be innocent victims who must share some of the burdens resulting from the redress of the past wrongs.”  He thus resisted the Court’s argument, in Wilks, that white employees had an equal right to file a civil rights action in Birmingham, alleging impermissible discrimination against themselves even though the city was acting under a consent decree to cure past employment discrimination that had excluded blacks from the city workforce.  In Wilks, the Stevens who before thought it intuitively wrong to blame folk for what they had not done, now argues that policy requires the injustice.  His wavering is characteristic of the general wavering that has characterized policymaking on these subjects, and that does not seem to have been relieved by the 1989 decisions.

Objections to the Wilks decision are perhaps hardest of all to understand.  For, in that case, the Court did not overturn the affirmative action plan.  It held merely that white persons had as much right to sue as other persons, and were required to make no stronger a prima facie case than were black persons in order to have their claim taken seriously.  The objections seem to describe this even-handed result as somehow offensive, despite clear indications that, once the white employees’ suit has been heard, it is very likely to be rejected.  Indeed Stevens’ dissent specifically objects to the “never-ending stream of litigation and potential liability.”  The objections mean, then, that differential burdens of proof, based on race, should apply to legal actions challenging discrimination—what my erstwhile colleague, Commissioner Robert Destro (United States Commission on Civil Rights), once aptly labeled, a “jurisprudence of minorities.”

Justice Thurgood Marshall has accepted the differential standard.  In Croson he quoted his Wygant[28] dissent to the effect that “agreement upon a means for applying the equal protection clause to an affirmative-action program has eluded this Court every time the issue has come before us.”  He thought that the nation’s experience with bigotry was more than sufficient to justify the set-aside program in the City of Richmond; it therefore eliminated the need for the ordinary prudence of congressional and court oversight called for by Sec. 5 of the Fourteenth Amendment.  In this decision (Croson), in which the so-called “Reagan Court” actually applied the theories of a Carter Administration Assistant Attorney General for Civil Rights,[29] Marshall suspected the Court majority of being unfriendly to blacks only because it sought to hold blacks and whites to an impartial administration of justice.  It becomes increasingly clear that Justice Marshall sincerely believes (what he had already expressed in the Bakke case long ago[30]) that we can articulate no standard whereby blacks and whites today shall be treated in the same manner.

It is a dubious claim that the task of repairing the historical injuries of slavery and discrimination more than justifies burdening innocent persons even if they did (and by no means did all) profit from the past.  Proponents of this view often try to reduce it to the idea of a sacrifice in the interest of the common good—or a greater good—the same, say, as if someone had to accept the unpleasant fate of a noisy roadway situated next the home that he had, perhaps, purchased for its quiet.  Immediately, of course, the examples do not differ at all, so long as a citizen were no less eligible for the one sacrifice as the other.  But that condition cannot be satisfied.  For, while all persons may be liable, without regard to race, to suffer the inconvenience of the roadway, only some, designated by race, must suffer the unfairness of racial preferences.

These excuses for racial injustice would be more patently clear if we placed them in juxtaposition with policies and practices everyone experiences and understands.  But what is better understood than the progressive income tax?  American society has accepted that people differently situated will pay taxes in different ratios, hence unequally.  This is deemed in the common interest and thought by most to be fair.  The reason, in brief, is that the distinction producing different treatment under the law is adventitious rather than invidious—it could theoretically happen to anyone.  Thus, they who are situated so as to pay more are making a contribution—a sacrifice—for the common good.

Suppose, however, that the progressive tax scale were calculated rather by race than by level of income—call it the affirmative action tax.  Historically, the differences would be statistically imperceptible.  Whether one said white folk or rich folk had to pay more, one would still be collecting largely from the same people.  Thus, the same practical result would follow.  Does the same moral result follow?  Does it not matter that at least some of the white folk are actually poor, while at least some of the black folk are actually rich?  A progressive tax based on actual income affects people equally, as to race, while the other is unequal and unfair.  The question is not whether but how sacrifices are made in the common interest.  Under a “jurisprudence of minorities” sacrifices are made unevenly—and based on invidious distinctions which deny by imputation the existence of a common good.

Let us review the Court’s decisions in order to see whether they are as far reaching as many civil rights advocates claimed—or are rather, as the Bush Administration initially maintained, merely technical.  In Croson, Justice O’Connor for the Court maintained that the judicial standard of strict scrutiny must apply no less surely to cities run by blacks than to cities run by rednecks.  Further, the Court might eventually contemplate loosening the judicial theory of “discrete and insular minorities” as a basis for law and policy.[31]  When the Court first adopted that theory fifty-three years ago, it was on the assumption that ordinary majoritarian political institutions would not protect minorities.  The new question is, how far such judicial guardianship is still required where the evidence of policy is that minorities do prevail and, indeed, sometimes even constitute the political majority?

One example of the problem that we face has arisen in Alabama and elsewhere, where white citizens have begun to file suits under the 1965 Voting Rights Act.  As hard as it is to believe, some observers, including the New York Times, have actually paused to wonder whether federal voting rights laws cover white voters?  Although a city such as Birmingham has a black majority, and its white minority sees itself as frozen out of local office by an at-large system, some civil rights advocates—as they call themselves—actually believe the lawsuits and tools used by blacks should not be extended to whites who now find themselves in the position about which blacks have long complained.  When I was a schoolchild we were taught a ditty that aimed to inoculate us against this very prospect:  “Oh how a minority, upon becoming a majority, hates a minority!”

The decision in Wards Cove addressed the question of the extent to which minorities may rely upon statistical disparities in order to prove discrimination in a work place.  The Court reasoned that plaintiffs who use statistics to build their prima facie case must nevertheless take the further step of proving that the disparities resulted from illegal bias. The absence of minority group members in skilled jobs (in an Alaska fish cannery) would not constitute evidence of discrimination, the Court maintained, if there were also an absence of qualified minority applicants for reasons beyond the employer’s control.

In Wards Cove, Justice White spoke for the majority and maintained that statistical comparisons had to meet a comparability test before they could establish prima facie evidence of discrimination.  Thus, a comparison between the basketball team and the graduating medical interns who treated the team’s injuries would serve no purpose for measuring discrimination, whether as between their immediate employers or in society in general.  In rendering this opinion the Court carried out the implications of its earlier decision in Watson v. Fort Worth Bank and Trust (1988), in which the Court widened the role of subjective factors in “disparate impact” or statistical cases but at the cost of bringing them closer to “disparate treatment” or individual cases in terms of rules of evidence.[32]  Thus, the Court maintained that, where discrimination is charged, a discriminatory practice must be identified.  To that extent the Court narrowed prior practice in interpreting §703 of Title VII of the “1964 Civil Rights Act”.[33]

A simple question was posed in Lorance, namely, whether employees have 300 days from the time a new seniority system affects them to file a complaint under Title VII?  The Court answered, “No.”  In fact, the legislation contemplated starting the statute of limitations clock from the implementation of the new work rules and not from the experienced effect of the rules. Justice Scalia expressed the majority’s view that Title VII of the “1964 Civil Rights Act” meant to create a preference for administrative regulation over litigation.  The rather strict limits on jurisdiction were designed to foster a greater dependence upon agencies such as the Equal Employment Opportunity Commission than on direct litigation.  Accordingly, not even “protected classes”—in this case, women—could be exempted from the effect of the statute.  Scalia effectively admitted that the statute produced an unfair result, and thereby invited Congress to reconsider the policy of discouraging litigation and regular compensatory and punitive damages.[34]  It will be up to Congress definitively to make Title VII compatible with American principles.  In the absence of such reconsideration, however, courts must give effect to the will of Congress.  Objectors to the decision did not focus on the policy option; they wondered instead whether women did not deserve better from the law.  The question is, better than whom?

In Martin V. Wilks ,Chief Justice Rehnquist held that the rights guaranteed by Title VII are personal, individual rights and not group rights.  The case asked whether white men who were not involved in litigation leading to a court-approved affirmative action plan providing preferences for minorities and women in the City of Birmingham could subsequently attack the plan as a violation of equal protection or at least of Title VII of the “1964 Civil Rights Act.”  Rehnquist, for the Court, held that no consent decree between an employer and a group of its employees could bind another group of employees not represented in the settlement.  Accordingly, individual white employees could indeed file suit, even against an existing consent decree which resulted from a prior suit filed by black citizens.  Objectors maintain that this creates a kind of multiple jeopardy for black gains by allowing deals struck with public (or presumably private) employers to be challenged on reverse discrimination grounds.  The majority opinion, however, emphasized that such suits would be subject to the identical rules of evidence required in other discrimination suits.  Accordingly, if nothing impermissible were accomplished in the consent decree or its implementation, no gains could be threatened.

Finally, in Patterson v. McLean Credit Union, Justice Kennedy upheld for the majority the decision in Runyon v. McCrary, namely that a Reconstruction era civil rights act applied to private acts of discrimination—but only insofar as the discrimination took the form of interfering with the right to make and/or enforce contracts.  Accordingly, racial harassment on the job (during the life of the contract) was not covered, although it is covered for most employees by Title VII of the “1964 Civil Rights Act.”[35]  Patterson had inquired whether the “1866 Civil Rights Act” might prohibit on-the-job racial harassment and thereby vindicate Patterson, who claimed that her employer had systematically harassed and belittled her.  The Court ruled, however, that the Reconstruction era statute only regulated the “making and enforcement” of contracts—not the conduct of employers and employees after formation of a contract even if that conduct is discriminatory.  Initially the Court had intimated that it might reconsider whether the 1866 statute applied to any private discrimination at all, thus reviewing unbidden its 1976 Runyon decision.  In the end, however, it upheld Runyon, while narrowing the focus of the 1866 statute as explained.

In order to understand how the demand for congressional action applied in each of these cases, we need to review one other case, Johnson v. Transportation Agency of Santa Clara County.[36]  When that case was decided in 1987 it was hailed by the same civil rights advocates who subsequently called for overturning the Court.  They counted it a victory that a more qualified white male had lost his suit for a position that had been awarded to a less qualified female.

What did Johnson say about the critical questions of rules of evidence and the status of white males as plaintiffs in discrimination suits?   Justice Brennan spoke for the majority, saying:  “the petitioner bears the burden of establishing” discrimination.  He quoted a different majority in Wygant to sustain his point (when the question was, What must a white male do in order to show that he was the victim of discrimination?):  “’The ultimate burden remains with the employees to demonstrate the unconstitutionality of an affirmative-action program,’ and we see no basis for a different rule regarding a plan’s alleged violation of Title VII.”

Brennan next laid out a precise order of burdens: 

Once a plaintiff establishes a prima facie case that race or sex has been taken into account in an employment decision, the burden shifts to the employer to articulate a nondiscriminatory rationale for its decision.  The existence of an affirmative action plan provides such a rationale.  If such a plan is articulated as the basis for the employer’s decision, the burden shifts to the plaintiff to prove that the employer’s justification is pretextual and the plan is invalid. . .  That does not mean, however, . . . that reliance on an affirmative action plan is to be treated as an affirmative defense requiring the employer to carry the burden of proving the validity of the plan.  The burden of proving its invalidity remains on the plaintiff. [Emphasis added.]

This is the identical order of burdens that is laid out in Wards Cove and relied upon in Wilks, and it was about its application to minorities that civil rights advocates complained.  The symmetry is inescapable, the difference no less obvious.  Brennan, who dissented in Wards Cove and Wilks, nevertheless applied the same order of burdens to white male plaintiffs.  But when it comes to minorities in similar cases, something called a “manifest imbalance” will suffice; for the purpose is to “remedy underrepresentation” for disadvantaged groups rather than to ensure justice for all citizens of a common tradition.

Following this rule, the “1990 Civil Rights Act” established different rules of evidence and different standards of justice, by race and gender, for the enforcement of civil rights laws.  It recognized distinctions that amount to legal differences.  In contemplating the future to which we thus resign ourselves we could do worse than to recall the reason that George Washington so constantly prayed for the “same justice for rich and poor.”  He seemed instinctively to realize that, if society once separates the moral and legal interests of rich and poor, it will render them far less disposed to cooperate with one another.

There are other aspects of these cases that might encourage us as well as point the way toward acceptable reforms.  The opportunity to press demands for revision in Court opinions as well as comprehensive legislation is great indeed, on account of the evident acceptability our political system accords to expressions of disagreement with the Supreme Court and widespread demands for revision.  In every Congress for the past twelve years or so, Congress has acted to overturn a Supreme Court decision in civil rights and related areas.[37]  It is therefore now commonly accepted that the American people should expect their representatives to correct judicial errors, and that they do not need to settle for the judicial bottom line.

Why should we seek revisions in Court decisions?  Look anew at the five cases we are considering.  The Croson majority still maintains the legitimacy of set-asides.  Justice O’Connor expressed the problem squarely:  “we confront once again the tension between the 14th Amendment’s guarantee of equal treatment [under law] to all citizens, and the use of race-based measures to ameliorate the effects of past discrimination on . . . minority groups in our society.”[38]  O’Connor resolved that tension in favor of policy instead of principle, while I would insist that policies without firm principles are merely arbitrary.

O’Connor raised, without settling, the question of the fate of the ethic of non-discrimination.  Justice Kennedy, in Patterson, stepped right through that open door, albeit cautiously.  He introduced in a key role in a majority opinion of the Supreme Court for the very first time the language from Harlan’s 1896 dissent in the “separate but equal” case.  “The law regards man as man, and takes no account of his color when his civil rights as guaranteed by the supreme law of the land are involved.”[39]

Justice Kennedy’s caution is shown by his stopping one phrase short of the controverted language, the “Constitution is color-blind.”  His boldness shines, however, in his willingness to use the citation the Court has shied away from for two generations. He indicates thereby a judgment that there exists today a firm national policy and sense of justice that supports prohibiting racial segregation and discrimination.  On that basis, the Court upheld Runyon.  (If the Court had not unwisely limited the ruling too severely, it would already be serving as a beachhead for color-blindness in our laws.  Instead, we have new legislation that serves to move us away from rather than toward that goal.)

Some of Justice Kennedy’s argument on behalf of color-blindness, however, is rather wish than reality.  There is no clear consensus that “race-based measures” are no less discriminatory than old Jim Crow.  Accordingly, there remains a need to build the consensus for which Justice Kennedy longs, and to do so with sufficient momentum to resolve positively Justice O’Connor’s tension.

A future challenge, after Croson, should focus on set-asides on the national level—to go after Fullilove.[40]  (I never did understand why Congress did not simply prefer bounties to employers or contractors for hiring members of minorities, anyway!  They would be no more perversely race conscious and would have the advantage of enlisting the self-interest, rather than the fears, of the persons who stand most to be adversely affected.)  Out of the Wards Cove decision there remains a need to structure rules of evidence in discrimination cases so as to connect charges of discrimination with concrete practices rather than random effects and chance events (See note 6 above).  It is a rather puerile argument that would legislate the conclusions of science by declaring statistics to prove causation.  From Lorance we learn that an injustice enforced by the Court was actually legislated by Congress—which summons us to expect from Congress recognition and a remedy of the defect.  It does not justify an unfair statute that it be applied equally.  Where the dearest personal rights of Americans are involved, Congress had no business to discourage litigation in the first place (which amounts to protecting wrong-doing).  Out of Wilks we are encouraged to think that Asians (who are no less likely than whites to intervene in consent decrees after the fact) will benefit from the affirmation that all Americans are protected from discrimination by race—even where affirmative action is involved.  Our task is to assure that they also understand the compelling attractiveness of this proposition, as they seek to break down barriers to elite universities and to break through the all too real “glass ceiling” in industry and government.

From Patterson, nothing can dwarf the significance of the establishment of Harlan’s color-blind language, which ought to be the language with which we teach our offspring to speak.  We require, moreover, to make good the contract guarantee opened up by that case and its predecessors.  The promise is more far-reaching than has been grasped or than the Court has elaborated thus far.  Patterson maintains that conduct within the contractual relation is expected to be regulated by private efforts that are, in turn, protected through guaranteed access to enforcement procedures.  Unless this process be side-tracked by new legislation, it offers the potential to demonstrate to every disadvantaged person in America just how far ahead this will leave him.

These cases have opened the prospect of a future in which the prevalence of the discussion of right and wrong over the discussion of policy will be a prelude to completely refashioned policies.  In order to satisfy demands that something be done about the legacy of discrimination, we must articulate clearly the principles on the basis of which anything we do will be defensible.  The idea of formulating public policy—or legal precedent—in a vacuum, can only lead in the end to unintentional justifications of despotism.  If standards of right and wrong are not the foundation of our efforts, then actions indifferently right and wrong (which means mostly wrong, because accomplished by force) must follow.

A proper policy goal would be to eliminate once and for all routine references to race and gender in surveys, plans, projections, and other official accounts of private and public workforces—wherever the logic of the profession itself does not impose such categorizations.  The fact that such usage is pervasive and deeply rooted describes the nature of the task before us—the first thing that must fall, accordingly, is the very concept of group representation or, more precisely, “protected groups.”  All Americans must be protected by freedom, or, in the end, none will be.

Summarizing, based on a principled approach, I have called for a number of legislative or judicial initiatives since 1987.  In summary, these reforms are:

1.      Complete the reaffirmation that the Constitution is color-blind.  This is the work of the Court and well within the reach of the Court’s resources.[41]  At a stroke this would deprive Congress of the resort to race and thus foster greater creativity in dealing with questions of civil rights.

2.      Complete the elaboration of the nexus between economic liberty and civil rights, making the right of contract a more robust source of protection.  This, too, falls initially to the Court to effect.[42]

3.      Extend recourse to tortious litigation as the principal means to defend individuals against the impermissible deprivation of legitimate powers and privileges, and, therefore,

4.      Replace legislated obstacles to litigation with recognition of the principle that individual injuries, whatever other reparatives may be encouraged, ought always to be compensable by means of such litigation.  Congress alone can accomplish these tasks.

5.      Re-codify the jerry-rigged structure of civil rights laws with a coherent, self-consistent code, enunciating principles by which any citizen may discover his salvation in the protections of the law.  This falls no less to Congress than the foregoing.


Such reforms as these are straightforward.  Nevertheless, it is safe to say that they are unlikely to be accomplished without some degree of reassessment of our general approach to civil rights.  For example, in 1988 Congress enacted the Fair Housing Act Amendments, in which they gave increased space to elements of the third and fourth proposals above.  Congress, though, simply added them onto new administrative obstacles, including a new layer of administrative judges.  Congress also obscured the definition of an injury (i.e., denial of contract) by seeking to specify classes of injuries to deal with the non-accommodation of the handicapped.  On balance, therefore, the 1988 Act was not an advance.

Two arguments are frequently opposed to reliance upon tortious processes as the preferred response to impermissible discrimination.  The first argument asserts the obstacle of excessive costs both to initiate litigation and, more importantly, in the realm of prospective awards or settlements.  This argument is spurious, as I will show, and stands rather as an expression of temporizing hesitance than of logical propriety.  The second argument is the warning that dependence on compensatory and punitive damages to deter discrimination will surely inspire employers or contractors to rely on the practice (as opposed to a legislated policy) of racial or gender preferences as the surest defense against the risks of litigation and liability.  This argument is serious and by no means disposed of by mere logical analysis.  Nevertheless, I will show that its chief error lies in its statement of the terms of analysis.

The cost argument separates into one argument about the cost of litigation for plaintiffs (who are supposedly too poor to pursue their cases) and another argument about so-called social costs.  Respecting the former argument, it is manifest that the availability of adequate counsel is directly dependent on the prospects for real financial gain in this realm no less than in the realm of personal injury litigation.  The cost of initiating litigation, therefore, is a barrier only to frivolous litigation.  As it turns out, however, in order to make the litigation worthwhile for plaintiffs, it is necessary greatly to increase the liability exposure of defendants.  Just as cost explosions in other of areas liability coverage have traumatized the United States, it is feared that a like result will occur in discrimination cases.  It is often maintained that society could not bear the cost of large settlements in discrimination cases.

In 1985 I participated in a conference on affirmative action, where I posed the following question to an author of the “1964 Civil Rights Act”:  “Just why did you design the Act in such a way as to discourage private litigation with attendant compensatory and punitive damages?”  The response was curt but complete:  “Are you kidding?  That would bankrupt the society!”  I maintained then, and I do so no less now, that I could not comprehend how a transfer of whatever magnitude between two members of a single society could bankrupt that same society.  My interlocutor’s premise must rather have been that the perpetrators of discrimination and the victims of discrimination in fact constituted two distinct societies.  Be that as it may, the social cost argument against reliance upon private tortious litigation with large awards or settlements as a deterrent to discrimination is plainly spurious.[43]

What, then, must one make of the propensity of likely defendants in such private litigation to shield themselves from exposure through the practice of racial preferences?   This would unfortunately be a necessary consequence of such an approach, as predictable as the raising of automobile insurance rates when states impose mandatory insurance laws, provided there were any available shield adapted to the purpose.  The assumption that employers or contractors could hire black people by preference and thus shield themselves, for example, is entirely vitiated by the premise that only “minorities and women” are protected by these legal procedures.  The moment one provides, however, against racial or gender discrimination altogether, there is in fact no such preference that can survive—for what is a racial or gender preference, after all, but implementation of an impermissible discrimination?  Thus, to protect oneself by slighting white people or males would be to little effect, where white people or males had the same access to these legal procedures as all other folk.[44]

At all events, in proportion as a society relies upon tortious litigation to deter impermissible discriminations, and despite any possible private schemes of preference that evolve, to that degree it will be found unnecessary to retain vast administrative structures to supervise the integration of society.  Further, legislated policies of affirmative action would be contra-indicated (as undercutting, deliberately perhaps, peoples’ opportunities to vindicate their own claims).




How, then, does the “1990 Civil Rights Act” compare with this agenda?  Some of its aspects are praiseworthy.  It avoided to reaffirm business set-asides as sanctioned remedies and also expressly disclaimed any intent to strengthen affirmative action or racial preference programs.  In addition, I would note the following worthy elements of the proposal:

1.      The act broadened compensatory and punitive recovery rights in cases of intentional discrimination.  This has been long overdue.


2.      The act established that the statute of limitations for certain civil rights violations in employment would effectively toll from the moment the individual is sensibly injured rather than from the point that a possible injury may be first projected or hypothesized.  This was a sensible response to Justice Scalia’s observation in Lorance that Congress itself needed to correct the tendency in the “1964 Civil Rights Act” to discourage litigation.  It is still only a small step, but a most important one.


These are significant accomplishments and, standing alone, they would easily constitute significant steps toward a fulsome reassessment of civil rights in the United States.  The very fact, however, that these steps were misconstrued as “correcting” Supreme Court decisions suggests that not all is well with the act.

There are other provisions in the 1990 Act, among them a requirement that “disparate impact cases” be resolved under a burden of proof requirement that enables the plaintiff to satisfy his obligation with mere statistical results.  In this section, a distinction is drawn between those employment practices that may disproportionately affect a particular group (however defined) and a particular discriminatory practice.  The latter, disparate treatment, was specifically excluded as a necessary condition of proof in these statistical cases.  Thus, without ever mentioning a particular group, Senator Kennedy made it directly unlawful that certain groups should participate in certain activities at a rate inconsistent with their proportion of the population in general or some relevant population in particular.  It need be said, however, that the Atonio decision made it crystal clear that the important question was precisely the question of the relevant comparison population base.  To the extent that the proposal did not address that question, and it did not, its utility is highly doubtful.  Nevertheless, it took sides in the dispute as to what the burden of proof should be for plaintiffs differently situated.

This is perhaps a good place to insist that the statute’s disclaimer regarding affirmative action is either empty or disingenuous.  Here is the language again:

Nothing in the Amendments made by this Act shall be construed to affect court-ordered remedies, affirmative action, or conciliation agreements that are otherwise in accordance with law.

It is evident that the qualifier, “otherwise in accordance with law,” may be quite significant indeed.  A challenge to a voluntary affirmative action plan, such as Paul Johnson’s 1987 challenge in Johnson v. Transportation Agency, would surely turn on a determination whether it were “otherwise in accordance with law.”  The important question, though, would be whether Mr. Johnson had to prove the plan inconsistent with law or whether he needed merely show that it affected him in a manner that revealed a disparate impact upon the group to which he belonged, thus shifting the burden of proof to the agency implementing the plan.  Since the burden of proof language in the statute failed to address this situation, it failed altogether to resolve the problem we face.

The language of the bill itself was unclear.  But Senator Kennedy’s testimony, and that of virtually every other Senator who addressed the matter on the day of introduction is very clear indeed.  Those senators unfailingly designated the exclusion of “minorities and women” from employment as the specific ill to be addressed.  They claimed, further, that this correction would restore the rule from Griggs v. Duke Power Co. (1971)[45]  that expressly tied the concept of “disparate impact” to the exclusion of “minorities and women.”  Such a reading is of course merely nominal; that is, nothing in the conceptualization of a disparate impact requires restricting its operation to minorities and women.  Thus, the decision that “minorities and women” are disparately “impacted”, while others are not, is entirely arbitrary.  The opening testimony in behalf of the “1990 Civil Rights Act” confirms that the statute aimed to codify that arbitrary standard, and apparently with the conscious intent to do so at the expense of Paul Johnson.  The actual process that results is meaningfully mischaracterized in the official “Summary” accompanying the text of the resolution in the Congressional Record:  “The Civil Rights Act of 1990” restores the Griggs rule by providing that, once a person proves that an employment practice has a disparate impact, the employer must justify the practice by showing that it is based on business necessity.”[46]  The plaintiff does not “prove” the connection between an employment practice and a statistical disparity.  That would have been the rule the Court had already imposed!  The plaintiff merely “shows” the disparity, and then the employer must “prove” and not merely “show” the practice to be substantially required by business necessity.  It should be noted as well that the formal “Summary” also uses the exclusive formulation, “women and minorities,” reflection upon which leads to the logically necessary conclusion that “white males” are excluded from the protections of the law.  This is a fatal flaw of the resolution.

Related to the foregoing flaw is the attempt by Senator Kennedy to overturn a Court decision in a minor case (Price Waterhouse v. Hopkins [1989]), in which the Court held that bias or racial or gender animus alone does not disqualify a business decision appropriately founded on independent considerations.[47]  The law, as passed, now insists that the appearance of prejudice in any decision fatally infects the decision.  Nevertheless, the act limits recovery for such an injury to damages and not to the position itself, if the plaintiff were indeed unqualified for the position!  In short, Senator Kennedy here aimed a blow not merely at private conduct but at the “crime” of impermissible opinion!

In yet another area, the supposed overturning of the Patterson decision, the statute in fact acknowledges the correctness of the Court’s opinion—as far as it went.  As Senator Kennedy put it, the “Court nullified the only Federal antidiscrimination law applicable to the 11 million workers in the 3.7 million firms with fewer than 15 employees.”[48]  In other words, Senator Kennedy conceded Justice Kennedy’s observation that the plaintiff in that case had access to the “1964 Civil Rights Act,” within the terms of that Act.  Since Patterson did not involve a non-Title VII employee, and courts generally decide the cases before them, its impact on non-Title VII employees rather reveals the limitations in Congress’ prior enactments than any defect in the proceedings of the Court.  The broadening of the protection of contracts in this provision is unremarkable, but it may someday contribute to strengthen the principle itself.

On balance the “1990 Civil Rights Act” modestly proposed to improve United States practices in the area of civil rights; it offers no encouragement that we may anticipate a general and productive reassessment.  The idea that Congress may serve as a protector of citizens’ civil rights remains elusive.

In the immediate aftermath of the Supreme Court’s 1989 spring term, numerous voices urged Congress to do something to rectify the supposedly retrograde direction of the Court. The informed observer, though, might have detected in these demands the emergence of a renewed demand to provide exceptions to otherwise general laws on behalf of so-called “protected groups.”




There is a different course heretofore eschewed; it may be summed up rapidly.  The tone was set by a recent study from the Heartland Institute, which expressly recognized that the decision in Richmond v. Croson, far from dismantling set-asides, only limited the licenses of states and municipalities to impose them.[49]  The point, they held, is that the means themselves neither produce the desired end nor are just.  “The notion that it is an outcome that is either just or unjust, rather than an individual’s actions,” they wrote, “leads to conclusions that contradict many of our common notions of justice.  For example, the rule seems to say that we may discriminate against a black person yesterday, pay a different black person for the injury today, and call it even.  Under this principle, one black person is pretty much the same as another.”[50]  The point is precisely to move our polity toward the day when “one black, one woman, one minority” is not pretty much the same as another—when counting noses is no longer an objective of public policy.


* Published in Reassessing Civil Rights, edited by Ellen Frankel Paul, Fred D. Miller, Jr., and Jeffrey Paul (Cambridge, MA: Blackwell Publishers for the Social Philosophy and Policy Center, Bowling Green State University, 1991), pp. 172-195.

[1] 347 U. S. 490.

[2] Funk & Wagnall’s New Standard Dictionary of the English Language, 1946.

[3] James Wilson, Works, ed. by James DeWitt Andrews (Chicago:  Callaghan and Company, 1896), vol. 2, p. 426.


[4] Calder v. Bull, 3 Dall. 386.

[5] One may imagine that the impartial administration of justice is sufficient to every such hypothetical.  In that sense, Johnson’s refinement of the definition would amount to a rejection of Wilson’s anticipation of honest government.  It stands on a theory of democracy that denies the claim to establish a “government of laws.”


[6]  For example, consider an experiment with a collection of marbles, differently colored in given proportions but otherwise identical, rolled out in a single cast.  The marbles as a whole will distribute themselves in a normal (bell-shaped) curve.  The different groups of colored marbles, however, will not be evenly distributed across the distribution.  The probability that they would do so on any given cast is remote.  Nevertheless, over an infinite series of such casts (with a two dimensional analysis, that is), they would  approximate such a normal curve.  This experiment suggests that the assumption that human beings, in groups alpha through omega, will distribute themselves across activities or fields a through z in a random pattern approximating for each group a normal distribution within each activity for each single iteration, absent discrimination, is wholly unwarranted.  Nevertheless, this very assumption is the theoretical foundation of all civil rights policy in our time.  A different but cognate observation is made by Thomas Sowell in his essay, “By the Numbers,” Policy Review, Winter 1982 [reprinted as Hoover Institution Reprints No. 49].  Sowell focuses here, as elsewhere, on the problem of indeterminacy; that is, our general inability to assess adequate causal evidence to determine the particular outcomes in given social distributions.  My point is rather a methodological consideration, for which purpose it must be assumed—as the marble example permits—that determination is complete.  Given that, it still remains that random distributions cannot be expected, one by one, to recreate general population patterns.

[7] Compare n 24 & n 42 below.

[8] Congressional Record - Senate, February 7, 1990, S-1018ff.

[9] City of Richmond v. Croson, 57 U.S.L.W. 4132.

[10]  427 U.S. 160 (1976); compare W. B. Allen, “Let’s Re-Do Runyon:  Questions to Guide Justice White,” Rutgers Law Review, vol. 41, no. 3, p. 893

[11] Patterson v. McLean Credit Union, 109 S.Ct. 2363 (1989).

[12] 334 U. S. 1, 68 S. Ct. 836, 92 L. Ed. 1161 (1948).  “These cases from (Missouri and Michigan) were successful challenges to judicial enforcement of the once widely used practice of restrictive covenants—agreements among property owners to exclude persons of designated races.  In the Missouri case, for example, a 1911 agreement signed by 30 out of 39 property owners in the area restricted occupancy for 50 years to persons of  ‘the Caucasian race’ and excluded ‘people of the Negro or Mongolian race.’  The petitioners in these cases were blacks who had purchased houses from white owners despite the racially restrictive covenants.  Respondents, owners of other properties subject to the terms of the covenants, sued to enjoin black purchasers from taking possession of the property and to divest them of title.  The state courts granted the relief.”  The United States Supreme Court reversed.  Gerald Gunther, Individual Rights in Constitutional Law, 4th Edition (Mineola, New York:  The Foundation Press, Inc., 1986), 543.

[13] 347 U. S. 483 (1954).

[14] 392 U. S. 409 (1968).

[15] 271 U. S. 323-331 (1926).

[16] Shelley, 334 U. S. 10, emphasis added.

[17] Shelley, 334 U. S. 11, 13, 12;  Buchanan v. Warley, 245 U. S. 60, 79 (1917) and Harmon v. Tyler, 273 U. S. 668 (1927).

[18] 109 U. S. 3.

[19] 100 U. S. 339, 347 (1880); Shelley, p. 14 emphasis supplied.

[20] Shelley, 334 U. S. 22; the first section of the Fourteenth amendment reads:  “All persons born or naturalized in the United States and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside.  No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws.”

[21] Compare Loving v. Virginia, 388 U. S. 1,2 (1967), in which the Court mistakenly claimed that this “constitutional question” had never before been presented to the Court.  Only the specific fact situation, disguised as statute-relevant, had not been previously addressed.

[22] Bolling v. Sharpe, 347 U. S. 497 (1954).

[23] In the 1989 term the Court gave itself a renewed opportunity to exploit Shelley’s full potential by binding over for reargument this apparently innocuous case, precisely for the purpose of reconsidering its Runyon opinion.

[24] See the fuller discussion of a path the Court might have followed in my essay, “Let’s Re-Do Runyon:  Questions to Guide Justice White.”

[25] 109 S.Ct. 2115 (1989).

[26] 109 S.Ct. 2261 (1989).

[27] 109 S.Ct. 2180 (1989).

[28] Wygant v. Jackson Board of Education, 106 S.Ct. 1842 (1986).

[29] Drew S. Days, III, “Fullilove,” Yale Law Journal, vol. 96, no. 3 (1987) pp. 453-85.  This very interesting essay makes the point that the Court had been too gullible when, in 1971, it accepted Days’ brief on behalf of the congressionally ordered racial set-asides defended in Fullilove.  His theory was that ultimately such blanket approval would lead to a general undermining of affirmative action by inviting abuses in states and municipalities.  Reminding his readers of the strenuous effort it took to get the Court to hold states and municipalities to a standard of “strict scrutiny” in the first place, Days reasoned that it would be a long-range advantage to retain that perspective, even in dealing with supposedly benign discriminations on the part of states and municipalities.  Finally, he warned, the failure to do so would surely mean that an eventual abuse in states and municipalities would erode confidence in such affirmative action measures altogether, thereby removing them as well from the panoply of federal remedies.

[30] Regents of Univ. of California v. Bakke, 438 U.S. 265 (1978).

[31] Originally adopted by the Court in 1938 in United States v. Carolene Products Co. [304 U.S. 144], this theory maintains that the Court is required to show special solicitude for the rights of minority inasmuch as one could not expect majoritarian political institutions to forward their interests.  The Court promised, accordingly, that while it would show decreasing alertness to public abuses of economic liberties, it would exhibit increasing vigilance to sniff out “prejudice against discrete and insular minorities.”

[32] 108 S.Ct. 277 (1988).  The chief contribution of this plurality decision was to add to the list of objective factors, such as standardized tests, that might have a disproportionately negative effect upon minorities, the kinds of subjective practices that had been thought theretofore to be safe from “disparate impact” review and to that extent less likely to figure in class-based litigation.  A general conceptual model for distinguishing “disparate impact” and “disparate treatment” would be to regard the former as that theory whereby some methodology is generally regarded as producing an impermissible discrimination, without raising any question of the intentions of the employer or responsible party.  Disparate treatment cases, on the other hand, almost require the additional step of alleging of discriminatory intent; such intent must at a minimum constitute a strong inference.  Each type of litigation may in fact involve statistical demonstrations and also affect classes rather than individuals alone, but only recently has it seemed viable to raise such practices as interviews and other subjective practices in disparate treatment cases.

[33] Relevant precedent for disparate impact analysis is found primarily in the following cases:  Griggs v. Duke Power Company; Albemarle Paper Co. v. Moody, 422 U. S. 405 (1975); Dothard v. Rawlinson, 433 U. S. 321 (1977); New York City Transit Authority v. Beazer, 440 U. S. 568 (1979); and Connecticut v. Teal, 457 U. S. 440 (1982).

[34] A district court judge in Birmingham has challenged the prevailing understanding, as well as the wisdom, of discouraging litigation and truncating constitutional guarantees, such as the right to jury trial.  Beesley v. Hartford Fire Insurance Co. may begin a period of reappraisal.

[35] For instance, for employees who work for employers having fifteen or more employees.

[36] Johnson v. Transportation Agency, Santa Clara Cty., Cal., 107 S.Ct. 1442 (1987).  Johnson and the woman, Joyce, competed for a road dispatcher position with the transportation agency.  The position was awarded to Joyce despite Johnson having survived the evaluation with a preferred recommendation.  The County, accounting for its decision when challenged by Johnson in a lawsuit, responded that Joyce was preferred on affirmative action grounds.

[37] The “Civil Rights Restoration Act of 1988” is only the most obvious example, rewriting as it did the statutory interpretation of the Court’s decision in Grove City v. Bell, 465 U.S. 555 (1984).  The Court had maintained that Title IX of the “1964 Civil Rights Act” applied only to the specific college program that discriminated on the basis of gender, rather than to the entire college.

[38] Richmond v. Croson.

[39] Plessy v. Ferguson, 163 U.S. 537 (1896).

[40] Fullilove v. Klutznick, 448 U.S. 448 (1980).  A premature and too little thought through attempt to revise this landmark decision was attempted in the spring of 1990 in Metro Broadcasting v. FCC, 58 U.S.L.W. 5053.  The question tested was whether federal racial preferences in the awarding of broadcast licenses do not impermissibly discriminate against whites.  In one of his final majority opinions, Justice William O. Brennan wrote for a 5-4 majority that “It is of overriding significance in these cases that the FCC’s minority-ownership programs have been specifically approved—indeed mandated—by Congress.”  The good news of this decision was that it found Justices O’Connor and Kennedy united in declaring that the United States has “one Constitution, providing a single guarantee of equal protection . . . to all citizens.”

[41] Compare my discussion in the essay, “A New Birth of Freedom:  Fulfillment or Derailment,” especially pages 79-80 and 85-86, in Slavery and Its Consequences:  The Constitution, Equality, and Race, eds., Robert A. Goldwin and Art Kaufman (American Enterprise Institute, 1988).

[42] Compare note 24 above.  The argument that a robust contract scheme would license discrimination gives too little credence to the notion that certain contracts are ruled beyond the pale from the beginning.  Yet, this is not a novel theory.  We do not enforce contracts for murder; we punish such contracts even in the case that nothing more than the contract has been executed.  The willingness to distinguish rightful from impermissible contracts gives even greater scope to the regulation of conduct by rightful contracts.  Persons will naturally divide their activities into those they are unwilling to undertake without an enforceable contract (and therefore on a non-discriminatory basis) and those regarding which they do not find a contract reassuring.  Surely, it will pose no burden for society to establish that same line of division respecting discrimination, regarding as impermissible only that discrimination that cannot attain its end without relying on the common force of society for the purpose.

[43] The United States Commission on Civil Rights has correctly described the fallacy of that argument in its “Report of the United States Commission on Civil Rights on The Civil Rights Act of 1990,” July, 1990.  The authors show that “Settlement amounts represent a cost to defendants and a benefit to plaintiffs.  On net, they represent neither a cost nor a [financial] benefit to society.”  p. 71n.

[44] One of the mysteries in the debate over racial preference is the clear decision by political conservatives in general to argue against preferences, not by reason of a defense of their own rights but rather on the spurious and paternalistic grounds of the harm racial preferences cause for the “disadvantaged.”  There seems to be an unspoken and awkward embarrassment that inhibits white males above all from simply declaring, “I got my rights.”  Indeed, I made this observation quite tellingly in a bastion of political conservatism in Washington, D.C. in 1989.  Recommending that one pose the rights of white males rather than the spectre of quotas as the real issue in dispute, the response I drew consisted largely of personal abuse directed at me (in absentia) by a Republican Congressman who had drafted legislation inconsistent with the goal I had espoused, followed by the general publication and distribution of that abuse by the conservative think-tank involved, although it had not published my statement to which the response was directed.  The message seemed clear to me:  the strategy of avoiding the issue in the debate about civil rights had the highest blessings.  I cannot help but believe, however, that that strategy is doomed to the failure that had greeted the American Revolution if the Founders, instead of saying “The tax hurts!” had insisted instead on “India’s need for trade!”

[45] 401 U.S. 424 (1971).

[46] Congressional Record, p. 1021.  Emphasis added.

[47] 109 S.Ct. 1775 (1989).

[48] Since Title VII of the “1964 Civil Rights Act” provides protection only for employees of businesses with fifteen or more employees, all other employees must rely on the Reconstruction era statutes for comparable protections.

[49] This argument was set forth most persuasively by a gathering of legal and constitutional scholars, summoned to Cambridge, Massachusetts by Drew S. Days, III and other defenders of affirmative action.  On March 30, 1989, this conference produced a “Constitutional Scholars’ Statement on Affirmative Action After City of Richmond v. Croson,” in which the participants announced, upon reflection, that although “some have recently argued that race-conscious remedies by local and state governments should be regarded as conflicting with the Constitution, [a]s long-time students of constitutional law, we regard this assessment as wrong.”  What followed was a resounding defense of racial set-asides but also of federal government supervision of local and state recourse to that remedy (as provided in the article by Days).  The signatories to this remarkable document were:  Judith C. Areen, Philip C. Bobbitt, Paul Brest, Denise Carty-Bennia, Jesse Choper, Peggy C. Davis, Drew S. Days III, Walter E. Dellinger III, Norman Dorsen, Christopher F. Edley, Jr., Yale Kamisar, Patricia A. King, Frank J. Michelman, Susan W. Prager, John E. Sexton, Laurence H. Tribe, James Vorenberg, Lee C. Bollinger, Barbara A. Black, Guido Calabresi, John Hart Ely, Herma Hill Kay, Gerald P. Lopez, Eleanor Holmes Norton, Robert M. O’Neil, Dean Rusk, Geoffrey R. Stone, Cass R. Sunstein.

[50] Diane Bast et. al., “Disadvantaged Business Set-Aside Programs:  An Evaluation,”  Heartland Policy Study, No. 26, June 29, 1989 (Chicago:  The Heartland Institute).


Back to Top