CHAPTER 6 On Civil Rights: Theories in Search of a Principle

As we have seen, there has been no more common distraction over “rights” than the tendency to fixate on rights to particular things, such as jobs or housing, while blocking from sight these underlying principles that mark the rightful and wrongful claims to these goods.

The main confusion that has befogged this subject may be clarified by the distinction between two different kinds of propositions:

(1) If Amazon moves its offices into the community, the community will prosper.

(2) It is wrong to hold people blameworthy or responsible for acts they were powerless to effect.

Proposition number 1 may be true, but then it may not be. It is an empirical prediction, at most a statement of probability. The desirable result it predicts is contingent on many other things. In contrast, the second proposition, as we have seen, states a truth that is logically necessary. There are no circumstances or contingencies in which it will fail to be true. And yet the litigation over civil rights, going back eighty years, has been cast along the lines of that first proposition, supported by a series of speculations, often quite tenuous and problematic, about material injuries that may result from racial discrimination.

I have made the case over the years that the surest anchoring ground for the laws on civil rights could be found in that second kind of proposition, one that supplies a rationale that would never fail. I have argued that behind the will or passion to discriminate on the basis of race is a species of “determinism”: the notion that race exerts a kind of deterministic control over the character and moral conduct of persons, so that if we know someone’s race, we can draw some plausible inferences about him. We can gauge whether he is, on balance, a good or bad man, whether his presence in the firm or the neighborhood would improve the business or the community or if that presence would have a degrading effect. In short, we would have ground for assigning benefits and disabilities to people on the basis of their race.

But if that were true, then no one could plausibly bear responsibility for his own acts, for everyone would qualify as a member of some known race. It might be said that the willingness to discriminate on the basis of race denies the moral autonomy or freedom that gives us our standing as “moral agents.” If we were not in control of our own acts, we would never deserve punishment at the hands of the law—and neither would we ever deserve praise. It is an axiom of the law, and indeed of moral judgment, that we cannot hold people responsible for acts they were powerless to effect. And so in all strictness it could be said that if discrimination on the basis of race were not wrong, then nothing literally could ever be “wrong,” for there would be no plausible standards of right and wrong to which persons could be held accountable. The whole language and logic of moral and legal judgment would be stripped of its meaning. The words “right” and “wrong” might imply a vague approval or disapproval, but they could not supply the grounds for casting judgments of right and wrong on anyone, either on others or oneself.

Or so I have argued over a good number of years—that this is what is deeply wrong in principle with racial discrimination.¹

I have thought that this understanding would rescue us from the bag of speculative theories and predictions that have been brought forth to explain the wrong of racial discrimination.

The moment came in Beauharnais v. Illinois (1952), a case dealing with the defaming of racial and ethnic groups.² Coming out of the war and the history of racial conflicts in Chicago, the Illinois statute made clear what it wished to forbid, and so it was charged that Beauharnais “did unlawfully… exhibit in public places lithographs, which publications portray depravity, criminality, unchastity or lack of virtue of citizens of Negro race and color and which exposes citizens of Illinois of the Negro race and color to contempt, derision, or obloquy….”

The offending publication was a leaflet cast in the form of an appeal to the mayor and city council of Chicago “to halt the further encroachment, harassment and invasion of white people, their property, neighborhoods and persons, by the Negro,” adding, “If persuasion and the need to prevent the white race from becoming mongrelized by the negro will not unite us, then the aggressions… rapes, robberies, knives, guns and marijuana of the negro, surely will.”

When Beauharnais was arrested, he invoked his “freedom of speech” under the First and Fourteenth Amendments. The judge drew from the script of Justice Holmes and told the jury that they would be warranted in convicting if they thought that “the article complained of was likely to produce a clear and present danger of a serious substantive evil that rises far above public inconvenience, annoyance or unrest.”³

The jury did convict, and the Supreme Court sustained the conviction in Beauharnais. Civil libertarians would argue that virtually all political speech may involve attacks on some groups, whether businessmen, pharmaceutical companies, or opponents of transgenderism. But the incitement of hatred toward racial, ethnic, and religious groups came with a different edge—and a vivid history.

Justice Jackson thought the judgment by and large right, and yet he was moved to dissent out of a serious concern for the details of procedure: In cases of libel or defamation, the “truth” of the speech is usually offered as a defense against the charges of libel. In this case, as Jackson said, the question of truth had not been tested or given an airing for the jury. Jackson was willing to overturn the conviction because that test of truth was not given to the jury. And yet he conceded that, in his own estimate, “This defendant [did not stand] even a remote chance of justifying what impresses me, as it did the trial court, as a reckless and vicious libel”: “A publication which diffuses its attack over unnamed and impersonal multitudes is likely to be harder to justify than one which concentrates its attack on named individuals, but the burden may properly be cast on an accused and punishment follow failure to carry it.”⁴

The burden was to be cast upon the accused even though, in Jackson’s judgment, he had no chance of carrying it. When a smart and seasoned jurist takes to the public stage to offer solecisms with a style of gravity, that should be reason enough for the accomplished man to look more closely into why he has twisted himself into that position. Why was Jackson so unshakably convinced that Beauharnais did not have the remotest chance of proving the truth of his charge that black people carry with them into the community a culture of “aggressions… rapes, robberies, knives, guns and marijuana”?

Is it not plain? Even if there is a higher incidence of crime in certain ethnic enclaves, we know that it would be deeply wrong to discriminate against a particular family because of what is known in the aggregate of the racial or ethnic group of which they happen to be members.

This point, so critical and yet so long overlooked, may spring out from an elementary problem: If we place people in different “groups,” where would we be warranted in drawing adverse judgments about the people who are placed accurately within these clusters? We might consider here the difference between: (1) arsonists, and (2) black people. No one falls into the class of arsonists unless he has engaged in the crime of setting fires for nefarious purposes. His fit with that label provides the sufficient ground for condemning and shunning him. But in contrast, it should be obvious that merely by knowing the color and race of any person we know nothing reliable and necessary about his moral reflexes—about any disposition to wrongdoing, as against a life of moral self-control or even saintliness.

These should be rather obvious points that one doesn’t need a graduate degree to grasp. But when we speak in this way, are we not imputing to all people a certain critical autonomy in reaching their own understanding of right and wrong apart from the ethic that may prevail in their racial or ethnic group? Are we not saying that it would be the gravest mistake to assume that the conduct of anyone is “determined” or controlled by his race or ethnic group, by forces outside his control? I would submit to a candid world, then, that if we look at the matter in this searching way, we will discover that the wrong at the root of racial discrimination is that it denies to black people their very standing as moral agents to bear responsibility for their own acts and receive the praise or blame that is theirs alone.

Now of course the ordinary man will not see a logic of “determinism” at work in the scheme of racial discrimination—any more than he is likely to mention the difference between things “contingent” or “categorical” even though the awareness of these differences is well lodged in his natural understanding. At every turn he is aware of things whose wrongness or harm will turn on matters of degree and moderation. And in contrast he has a sense of things so wrong that their wrongness will not be tempered by matters of degree or excused under different “circumstances.” The decisive and telling test can be seen whenever we find people simply willing to turn away, in a sweep, from all blacks or other racial or ethnic groups. (“Whites only.” “No Irish need apply.”) In this broad brush of aversion, we find no interest in considering the innocence or the redeeming goodness of anyone in the group. To discriminate in that undiscriminating way is to absorb, operationally, the premise that race essentially controls or “determines” anything of moral consequence that we would need to know about any person we are judging.

The rulings of our courts, refracting the laws on civil rights, have given us a train of misadventures and distortions, all springing from the fact that the key judgments could not find that anchoring ground of principle. They have been contrived, instead, to fit an array of “contingent” theories about the wrongs done in racial discrimination. The result has been a certain confusion in our public policy, as people in and out of politics have blurred the distinction between real principles and merely contingent theories, which may or may not be true. To recall the train of notable cases is to see the problem unfolded for us with almost comic sadness.

Those confusions had already been accumulating on the way to the famous decision in Brown v. Board of Education in 1954. But that case launched a new era, both by rejecting racial discrimination and by stirring the courts to break through the conventions that had limited and constrained the powers of judges. In Brown, the Court struck down a system of racial segregation in public schools mandated by the law.⁵ In the years following the case, there seemed to be certainty on all sides that the correct decision had been made. And yet from the first reactions to the case, even thoughtful liberal critics found it difficult to explain the reasoning that made that decision either defensible or comprehensible. For where had the Court found the wrong in racial segregation? The Court had said that the system of segregation would impart a sense of inferiority to black children, and that that sense of themselves was likely to have a demoralizing effect, impairing their capacity to learn.⁶ But no empirical evidence had been gathered to show that segregation had indeed produced any measurable decline in the performance of black children in the schools of Topeka, or in any of the other cities that had been collected in the case. The only empirical studies brought forth were from schools in Springfield, Massachusetts, and Little Rock, Arkansas. The sociologist Kenneth Clark had used a combination of dolls and coloring books. In a series of questions, Clark asked the children to identify the “nice doll,” or the color that little girls and boys ought to be, or the dolls that looked like themselves. The experiment elicited accounts of black children sobbing or trying to see themselves as a different color. These reactions seemed to reveal that black children had indeed absorbed a sense of themselves as part of a lower caste, ready to reject their blackness. But the embarrassment for the Court was that this tendency was more pronounced among students in Springfield, Massachusetts, at the racially integrated school. If there was something truly wrong causing black children to reject the color of their own skin, that wrong could not be imputed to segregated schools.⁷

And quite beyond that, the very setup of the problem invited the challenge: What if the children had been separated on the basis of race and their reading scores had gone up? Would the segregation have ceased to be wrong? Was the wrong here really contingent on the performance of the children in the schools, or was there something wrong with this segregation in principle, quite apart from whether the students did well or badly? And if so, what exactly was the operative principle? It would have made an evident and profound difference to explain that principle—to explain what made it wrong, of necessity, to draw moral inferences about the worth and deserts of people on the basis of attributes such as race.

That this was no trivial confusion was reflected in a seminar years later at the University of Chicago, in the mid-1960s. Philip Kurland, the revered professor of constitutional law, found himself earnestly asking how we got from Brown v. Board (racial segregation in schools) to the Palmer case, involving racial segregation in a public swimming pool.⁸ Kurland remarked that the holding in Brown rested on the claim of impairing the performance of children in schools. Was the claim now being made that if black children were denied the freedom to swim in a pool, their academic motivation would be chilled, along with their performance?

Once again, was the “wrong” remedied in the Brown case simply contingent on the effect of segregation on the motivations and performance of the children? If there had been something wrong in principle with that segregation, then Kurland had fallen into the familiar mistake of confusing principles with the instances in which they happen to be manifested. We may recall the classic experiment of a ball rolled down an inclined plane: as the angle of inclination becomes steeper, the rate of acceleration quickens for the ball rolling down the plane. Once we are clear on that principle, it doesn’t matter whether we try out the same experiment with wooden or plastic balls or red, blue, or white planes. The principle is virtually indifferent to the numberless variety of instances in which it may be manifested. And the same elementary point is engaged in that freighted matter of racial segregation: If we encountered a case of black people barred by local law from the use of public swimming pools or tennis courts, would it be necessary to articulate a new “constitutional right to swim” or a “constitutional right to play tennis”? Or does it make more elementary sense to say that we need no new rights here: we are simply dealing with the same principle, the principle that bars “racial discrimination,” manifested in a variety of instances?

But the confusion noted by Philip Kurland would have its most dramatic expression in a case that would eventually facilitate some of the most radical changes in our law and culture as it bore on the subject of marriage. The problem was revealed in the very opening lines of Chief Justice Warren’s opinion in 1967 when the Court faced the question of the laws in Virginia and other places that barred marriage across racial lines. “This case presents,” he said, “a constitutional question never addressed by this Court: whether a statutory scheme adopted by the State of Virginia to prevent marriages between persons solely on the basis of racial classifications violates the Equal Protection and Due Process Clauses of the Fourteenth Amendment.”⁹ A “question never addressed by this Court”? But what was the nature of the question the Court had never yet addressed? Was it about legal discrimination based on race? Or was it about marriage? I raised the question once of what the judges might have done if the case had involved a law in Virginia forbidding partnerships in business across racial lines.¹⁰ I imagined a case of two friends, black and white, who together purchased a delicatessen, and I called the case Zabar’s v. Virginia. Would the judges have said that they had never seen before a case involving delicatessens? And when they struck down the law, would they have articulated nothing less than a new constitutional “right to own a delicatessen”? That would have made as much sense as the judgment that the justices announced in Loving v. Virginia, that the case brought forth a new constitutional right to marry.¹¹ As the justices acknowledged, that so-called right could be restricted at many points, as the law forbade incestuous couplings and the marriage of children.¹² The drift of the justices into new confusions about marriage was yet another sign of the fact that they had not been clear on the principle in the case. For if they had, they would have recognized that this case could have been resolved without saying a thing about marriage—just as the case of the partners in business would have been decided without saying anything about a right to own a delicatessen. The case could have been resolved on the grounds that we had merely another instance of racial discrimination, of something wrong in principle, wrong then categorically, under all conditions and instances. Instead, with this careless flinging about of labels, the Court created a right to marriage that it did not strictly mean, for the justices would take occasion over the years to keep pointing out that marriage could rightly be restricted on many plausible grounds.¹³ But the “right to marry” would become woven into the precedents of the Court until it finally set the ground for that right to be claimed by couples of the same sex.¹⁴

When we are alerted to the problem in this way, it becomes even more revealing to pose anew the original question: What did the justices have to say finally about the wrong of racial discrimination? Where did the wrongness lie? In dealing with that question, central to the case, Chief Justice Warren stated as his leading point that “this Court has consistently repudiated” distinctions based on race. That is, he did not explain the ground of wrongness in these discriminations; he merely reported that he and his colleagues on the Court had rejected those discriminations in a line of cases. But clearly the Court had not “consistently repudiated” those racial distinctions. The Court had upheld those distinctions, after all, in Plessy v. Ferguson (1896),¹⁵ and the justices had by and large sustained those distinctions until a slight turn beginning in the 1930s. What Warren meant was that the Court had rejected that discrimination over the preceding twenty or thirty years. In a rough estimate, Warren was talking about a cohort of about twenty-nine judges sitting on the Supreme Court when just about all of those twenty-nine were willing to reject racial discrimination most of the time. That was quite different from saying that the discrimination was categorically, in principle, wrong—and indeed Warren noted that only two judges had come close, of late, to making that claim. All that Justices Stewart and White had said was that they themselves could not “conceive of a valid legislative purpose… which makes the color of a person’s skin the test of whether his conduct is a criminal offense.”¹⁶ That is to say, all we were offered here was a report on the sensations of Justices Stewart and Douglas—a report of what they could see or conceive. That was quite far from explaining how racial discrimination is wrong in principle—wrong categorically, in all instances, regardless of its effects in any case and regardless of what judges are able to “conceive” or “see.” Neither Justices Stewart nor Douglas ever articulated a categorical proposition, defining something morally wrong under all conditions. And what Stewart and Douglas had left unexplained, no one else over the years has bothered to supply.

When people were moved to say something deeper about the wrong of racial segregation, the most attractive temptation was to look back to Justice Harlan’s dissent in Plessy v. Ferguson. The laws of Louisiana had mandated the separation of the races in railway carriages. Homer Plessy, a rather light-skinned man of color, had been assigned to a car reserved for black people. Justice Harlan insisted that to take the side of Homer Plessy in this case was not to impose a scheme of social equality any more than the civic service of blacks and whites on the same jury marked a presumption of social equality. Harlan declared, in memorable, resonating lines, that the Constitution was “color blind,” that “there is no caste here.” He argued that “in [the] view of the Constitution, in the eye of the law, there is in this country no superior, dominant, ruling class of citizens.”¹⁷ And what else was to be inferred from this separation of races in public places but that “colored citizens are so inferior and degraded that they cannot be allowed to sit in public coaches occupied by white citizens.”¹⁸

But again, putting the question severely, where in principle lay the wrongness in regarding a race of people as “degraded”? Was it something more than a matter of personal preference? Did it have something to do with regarding a whole class of citizens, defined by race, as degraded, without knowing anything about the character of the individual persons who made up the group? Harlan had insisted that it was wrong to stamp citizens as members of a superior or subordinate class based simply on their race (“there is in this country no superior, dominant, ruling class of citizens”). But then was it inadmissible, in the same way, to create superior and subordinate classes of “persons” based on race in gauging their fitness to be given the standing of “citizens” in this country? On that point, Harlan revealed the flaw that called his whole argument into question. In the sweep of his opinion, he remarked that “there is a race so different from our own that we do not permit those belonging to it to become citizens of the United States. Persons belonging to it are, with few exceptions, absolutely excluded from our country.”¹⁹ Harlan made it clear that he was referring to “the Chinese race.” His rhetorical point was this: even “a Chinaman,” he said, “can ride in the same passenger coach with white citizens of the United States, while citizens of the black race in Louisiana, many of whom, perhaps, risked their lives for the preservation of the Union, who are entitled, by law, to participate in the political control of the State and nation,” were treated as unfitted, as a class, for the company of white people.

Harlan’s sentiments on blacks and race were rightly animated, but it must be said that he never did explain the wrong in principle here. He never really explained why it was wrong to draw moral inferences about people on the basis of race, to judge them more or less fitting for acceptance or rejection, for honors or disparagement, in settings private or public. The want of that argument may explain why Harlan’s dissent offered no particular help in the later arguments about affirmative action. It may also account for why it offered no guidance when the justices, in the 1940s and ’50s, were searching for some way of explaining whether segregation on the basis of race was not only wrong in relation to this or that particular case, involving schools or lunch counters or swimming pools, but whether there was something in principle wrong with discrimination based on race, wherever that wrong was manifested. As Akhil Amar has reminded us, the Court did not overrule Plessy in Brown v. Board, because Plessy involved segregation in transportation and the cast of the Court’s opinion in Brown confined its holding to racial segregation in education.²⁰

The judges never quite broke out of the scheme of identifying the wrong in any case as contingent—contingent, that is, on the injuries that could be produced in, say, schools or the assignment of jobs. There would be a string of cases involving black students at universities forced to sit by themselves in classrooms, libraries, and dining halls.²¹ Just to describe the segregated arrangements is to have the wrong of the situation spring out. But when it came to homing in on the legal wrong at issue, the Court fell into the speculation that the black students would lose the chance to be recognized and respected by white students and perhaps miss connections that could be highly useful for their careers.

But simply bringing students together offered no guarantee that a student would gain recognition of his merits, or indeed that he would have merits to compel admiration. And whether it is good in any case to gain acceptance depends on the character of those people whose acceptance is being sought. Instead of invoking a clear, applicable principle, this argument draws a hazy and dubious series of inferences about what might happen. The argument here resembles those against barring entry to certain private clubs on the basis of race or gender on the ground that it prevents people from meeting and making connections that may be critical to their careers. That is all, of course, speculative and problematic: to be in the club, to be in the university or the law school, is no guarantee that one will make friends and connections that lead to success in the vocation.

But when it comes to offering a chain of predictions, hardly anything compares with the chain of speculations that the Court relied on in sustaining something as momentous as the Civil Right Act of 1964. That landmark act barred discrimination based on race in “public accommodations”: certain private businesses open to transactions with the public. The act ran into a constitutional tangle. The Fourteenth Amendment, which provides for the equal protection of the laws, dealt only with policies imposed by the laws. That was strikingly different from conventions of racial separation that were adopted by people as a matter of private right in managing their own homes or businesses or clubs. The most familiar legal justification for the federal government to regulate private businesses was found in the Commerce Clause. But the jurisprudence there had long been built on a pyramid of fictions, often implausible and at times even comic. (It used to be said that if a window washer, high in a skyscraper, could see a railroad track, and that track carried trains that crossed state lines, the federal government could regulate the wages of that window washer!) The reasoning offered by the Court was no less implausible as it unfolded its argument in the first case to test the Civil Rights Act of 1964, the case of Katzenbach v. McClung (1964).²² The case involved Ollie’s Barbecue in Birmingham, Alabama, a notable local establishment that refused to admit black people to its main dining room. Mr. Justice Clark, writing for the Court, sought to bring the case under the Commerce Clause in this way:

The record is replete with testimony of the burdens placed on interstate commerce by racial discrimination in restaurants. A comparison of per capita spending by Negroes in restaurants, theaters, and like establishments indicated less spending, after discounting income differences, in areas where discrimination is widely practiced. This condition… was especially aggravated in the South…. This diminutive spending springing from a refusal to serve Negroes and their total loss as customers has, regardless of the absence of direct evidence, a close connection to interstate commerce. The fewer customers a restaurant enjoys, the less food it sells, and consequently the less it buys. S.Rep. No. 872, 88th Cong., 2d Sess., at 19; Senate Commerce Committee Hearings at 207. In addition, the Attorney General testified that this type of discrimination imposed “an artificial restriction on the market,” and interfered with the flow of merchandise.²³

In other words, the argument moved in this way: If black people faced the prospect of discrimination in public inns and restaurants, they might be discouraged from traveling among the states. If blacks were discouraged from traveling, the effect would be to reduce the total volume of trade available to restaurants, inns, and other places of public accommodation. That shortfall of trade would, in turn, reduce the orders that these businesses would place with other businesses (for example, for meat, linens, and silverware). The effect would be to depress even further the level of trade and the general standard of living.

In the eyes of the law, then, the problem of discrimination in places of public accommodation did not lie in any injustice that was done to black people, but in the hindrance it might cause to the interstate transport of meat. As a colleague of mine once pointed out, the problem stated here by the Court might have been resolved quite as well if the racists in the country had simply made up the shortfall by eating more meat!

This confusion between things wrong in principle and things merely contingently wrong was brought out in the most striking way in one of the classic cases in the 1930s, Missouri ex rel. Gaines v. Canada (1938).²⁴ In the 1930s the state of Missouri refused to admit blacks to the law school supported by the state. Instead of establishing a separate law school for blacks, the state offered to pay “reasonable” tuition fees for any of its black citizens who gained admission to law schools in adjacent states where segregation was not practiced. The Supreme Court struck down this arrangement. The fact that the state supported a legal education for blacks as well as whites was not enough. Black students would not have the chance to meet lawyers of their own age who might become important to know in practicing law in Missouri. In furnishing a law school for whites only, the state was establishing privileges or benefits for white students that it was denying to blacks solely on account of their race. The conjecture was that the policy was barring black students from the possibilities of serious material benefits in the future. But skipped over by the Court was an indelicate question: What if the law schools in neighboring states were in fact superior to the law school at the University of Missouri? The embarrassing point here was that black students could in fact be “forced,” in this arrangement, to attend a better law school than the one they were barred from attending in Missouri. That this was indeed the actual experience in a number of cases was suggested in an interview given a while back by Cecil Partee, a prominent black politician in Cook County, Illinois. Partee recalled that he had graduated in 1938 at the top of his class at Tennessee State University. As a native of Arkansas, he applied to the law school of the University of Arkansas, but Arkansas had a policy similar to that of the state of Missouri. The state offered to pay Partee’s tuition at another school. Partee happened to be admitted to the law schools at the University of Chicago and Northwestern University, both notably superior to the law school in Arkansas. Partee ended up choosing Northwestern. He later commented, “I laughed all the way to Chicago.”²⁵

Cecil Partee did not suffer a material injury as a result of being excluded on the basis of race from the law school of the state, but he was nevertheless wronged. That is to say, he was treated unjustly, treated according to an unjust principle. Partee found the situation laughable. It is one of the ironies in the record of the law that the victims themselves can be quite unaware at times that they are adopting the very principle they think they are resisting. That curious want of noticing was on display, of all places, in the classic case Plessy v. Ferguson, in the argument made by Homer Plessy, the plaintiff in the case.

Plessy fell into the same groove of convention that had been at work in the notorious cases of racial libel. In Germany in the 1930s, good Aryans could sue for damages if they had been identified in published reports as Jews. And in the American South, white people could sue if they were identified falsely as “Negroes” or blacks. In their trials, the defense would try to show that the accuser did indeed have features that looked “negroid,” and in Germany the defense would try to show that the accuser did bear features that appeared quite “Jewish.” These empirical offerings were brought forth without the doctors’ of the law noticing that the very premises of the law contained a vicious falsehood: namely, that if certain persons were known to be blacks or Jews, decent people would naturally draw adverse moral inferences, and of course seek to avoid their company.

And that is exactly the groove of argument that Homer Plessy fell into when he and his lawyers made their case about the wrong he had suffered. Plessy was assigned to the car reserved for colored people not because anthropologists had carefully studied his physiognomy and speech, but because one of the lead conductors for the train had to make a judgment call and thought that Plessy, even as light-skinned as he was, looked like a colored man. And indeed, Plessy and his supporters were counting on the fact that he would be seen that way, as they sought to test the law with this lawsuit. Still, in pressing that case, Plessy and his lawyers were willing to invoke anything in the arsenal of the law, and so they reached for this familiar lever: Plessy was willing to argue that by placing him in the carriage marked for black people, the management of the railroad had deprived him of his property in being a white man!²⁶ The action deprived him, that is, of the advantages of being known as white, and opened him to all of the material disadvantages of being identified as black. All of the confusions that would arise in later years might have been foretold in the confusions of the principal figures in Plessy v. Ferguson. Even John Marshall Harlan in his great dissent could not quite give an account of the principle that established the deep wrongness of the racial discrimination. And the chief victim in the case invoked on his own behalf the same flawed principle that stood behind the laws he was resisting.

One of the enduring oddities of moral judgments is that people may be observed doing exactly the same things, even though their acts are motivated by notably different principles. Consider, in this vein, two restaurant owners in a liberal academic town—perhaps my own Amherst, Massachusetts.²⁷ Or it could be Ann Arbor, Michigan, or Berkeley, California. And let’s suppose that the owners of both establishments have settled on this policy for their restaurants: they will not discriminate among their customers on the basis of race.

Imagine that Restaurant Owner A operates on this maxim: it is wise to accord the rules of your establishment with the ethos, the moral outlook, that prevails in any place. In liberal Amherst (or Berkeley) it would be quite bad for business if word got around that a restaurant refused to accept black customers.

In contrast, let’s suppose that Restaurant Owner B holds to this anchoring maxim: it is incoherent to draw any inferences about the goodness or badness of people as persons or customers on the basis of race, as though race controls and determines the moral character of persons. Therefore he will make no discriminations among his customers on the basis of race. The two owners are conducting themselves in the same way, but with underlying maxims that are radically different.

Now let us imagine that both owners are somehow transported to South Africa in the 1970s, when apartheid was still in effect. What happens now to the conduct of these two owners and their establishments?

Restaurant Owner A does a flip. His animating maxim is to accord the rules of his establishment with the dominant ethos wherever it is. A policy of racial exclusion would have been bad business in Amherst, but it may be the means of commanding the right clientele in Johannesburg.

Restaurant Owner B’s policy remains the same. For his position has been anchored in a proposition that will not have altered with the shift in locale. Whether we are in Johannesburg, South Africa, or Amherst, Massachusetts, it is incoherent to claim that we can draw moral inferences about people on the basis of race, as though race essentially determines or controls the conduct of every person.

Restaurant Owner B may quickly come to see that he will have trouble surviving in business if he continues to hold to this underlying principle. But nothing in the change of locale alters in any way the validity of the principle guiding his action. Whether we are east or west, the language of moral judgment would lose its meaning and coherence if we worked on the premise that none of us was in control of his own acts, that we are not properly the subjects of blame or praise because our acts are “determined” by forces outside our control.

Restaurant Owner B was governing himself according to a maxim that springs directly from the logic of morality itself. It springs, that is, from a principle that is true of necessity—a principle that, as Alexander Hamilton said, contains its own “internal evidence, which… commands the assent of the mind.” His act was grounded in the laws of reason that will be true in all places. And therefore, as Kant would say, it is fit to be installed as a universal rule. Its validity will not be affected even by the melancholy results of the owner’s finding his business failing. The goodness or rightness of his policy is not contingent upon the success of his restaurant. His policy is right or good in itself—it is, we can say, categorically right and good.

Lincoln once remarked to a young man aspiring to be a lawyer that if he really worried that he could not be honest and be a lawyer, it was preferable that he remain honest rather than be a lawyer. This bit of advice wisely and soberly offered to lawyers would not become utopian if offered to the owners of restaurants—or to anyone else.

The force of the principle here is that we have in hand a necessary truth that furnishes one of the anchors of our moral and legal judgment. There is no case, there will be no set of circumstances, in which it could fail to be true. When producers need to cast actors to play the roles of Jackie Robinson or Martin Luther King, they of course must turn away white actors and choose only from among black men for the roles. But that is strikingly different from a case of drawing moral inferences about people—either rewarding or punishing them—solely on the basis of race, as though race could exert a force that reliably “determines” actions or character. And if it is truly wrong categorically to visit penalties or punishments on people solely on account of their race, then it is wrong to do that even some of the time, just occasionally. If it is wrong to kill people on the basis of their race, it is wrong to do it even if we do it “only at the margins”—perhaps just “taking race into account” as part of an ensemble of other attributes. Race may be only one ingredient that fuels the rage of people with a murderous bent, but if that is the criterion that the would-be murderer acts on, decisively, then the full wrongness of the act is undiminished. A wrong in principle is quite indifferent to matters of degree. And so if racial discrimination is so deeply wrong, wrong of necessity, it must surely be indefensible to say that we are taking race into account, as one among a rich host of diverse criteria, in deciding admissions to colleges and graduate programs. If the decision does pivot, at the margin, on race, then that is indeed the decisive point, and the deep wrong cannot be disguised by the stylish dance. At the end of the day, if applicants are given the gift of admission to Harvard or Yale on the basis of race, then it follows, as night follows day, that other candidates are being turned away, decisively, on the grounds of their race.

That was the point that made the critical difference, at least at the threshold, for the late Justice Lewis Powell in the famous case of Bakke v. University of California, Davis (1978).²⁸ Allan Bakke, a middle-aged white man, was turned away from the UC Davis medical school even though he presented with “board scores” that were outstanding by any commonsense measure. In the standard tests measuring aptitude on matters “verbal” and “quantitative,” Bakke placed in the 96th and 94th percentile. In contrast, the applicants who had been set aside for special admission to a cluster of places reserved for ethnic and racial minorities showed an average score at the 46th and 24th percentiles. The “regular admittees,” accepted without any racial or ethnic preferences, had average scores in the 81st and 75th percentiles.²⁹ In other words, Bakke’s performance had placed him well above the run of students thought good enough to be accepted, even without a racial tilt to the scale. Justice Powell and his colleagues reacted to the grossness of the numbers, which bespoke a “quota.” They recognized that the advantages of admission had been assigned unequivocally and decisively on the basis of race. Even if it were true that blacks and Hispanics had suffered disadvantages that had accumulated over the years, it was not clear that Allan Bakke had borne any responsibility for them. In making this strong case, at least in the first phase of his opinion for the Court, Powell came to the very threshold of explaining the deep wrong, in principle, of racial discrimination. He remarked that the Court had “never approved a classification that aids persons perceived as members of relatively victimized groups at the expense of other innocent individuals in the absence of judicial, legislative, or administrative findings of constitutional or statutory violations” [emphasis added]. He recognized that if a preference had to be given now to one race or ethnic group over another, there would be a need to explain just why the miseries suffered in the past by certain groups could be converted into a ground for favoring one group over others now, in a racial lottery. The Irish had fled the famine in their country and suffered humiliating discrimination in employment. So too had the Chinese and Japanese. Poles and Jews had been driven from Europe and the killing fields of the Nazis. And that is to say nothing of the “boat people” and refugees fleeing later from Vietnam and Cambodia. Powell set forth the problem aptly:

The white “majority” itself is composed of various minority groups, most of which can lay claim to a history of prior discrimination at the hands of the State and private individuals. Not all of these groups can receive preferential treatment and corresponding judicial tolerance of distinctions drawn in terms of race and nationality, for then the only “majority” left would be a new minority of white Anglo-Saxon Protestants. There is no principled basis for deciding which groups would merit “heightened judicial solicitude” and which would not. Courts would be asked to evaluate the extent of the prejudice and consequent harm suffered by various minority groups. Those whose societal injury is thought to exceed some arbitrary level of tolerability then would be entitled to preferential classifications at the expense of individuals belonging to other groups. [Emphasis added.]³⁰

Might the matter just be left to local majorities voting in, say, San Francisco or Los Angeles? Might they be invited to decide which racial or ethnic group should receive this largesse in the scheme of racial and ethnic favorites? That would surely set one group against another in racial conflict. On the matter of academic admissions, Gallup surveys have shown over the years that whites and Hispanics have been firmly on the side saying that “applicants should be admitted solely on the basis of merit even if that results in few minority students being admitted.” In the same surveys, blacks have been emphatically tilted to the side saying that “an applicant’s racial and ethnic background should be considered to help promote diversity on college campuses, even if that means admitting some minority students who otherwise would not be admitted.” But by 2016, 50 percent of the black respondents came down on the side of “merit,” edging out for the first time those who would give some weighting to race in admissions. But all of this depends, of course, on the vagaries of the sampling, and also on the ups and downs in our politics. In the aftermath of the killing of George Floyd in Minneapolis in 2020 and the conflagration it set off, even soothsayers will lose their certainty in offering their readings of the data. Still, that may not throw off an attitude settling in among black people in favor of a policy of “neutrality,” rather than letting governments decide which will be the favored races in any city or state.³¹

In Bakke, Justice Powell seemed set to pronounce on the wrongness of racial preferences under all conditions, no matter who ends up with the spoils. But then he suddenly did a turn and fell into the familiar lingo of judges. The racial discrimination was grave enough, he said, to trigger the test of “strict scrutiny.” Strict scrutiny supposedly stands for the most demanding standard offered by the Court, and yet by its own terms it turns away from recognizing the depth of a wrong that is truly “categorical.” For it tells us that certain acts are so presumptively wrong that they require a demanding set of reasons to justify them. But in all strictness, there is no way for a categorical wrong to be justified—no coherent way of explaining why, say, it could ever be justified to punish people for acts they are powerless to perform. Or, in the case of “racial preferences,” to punish people for the acts done, or forborne, by persons with whom they had no connection.

For Powell, raising the flag of “strict scrutiny” was the sign of moving into the second phase of his opinion, where he would lose the colleagues who had joined him in striking down racial quotas and clearing the way for Allan Bakke’s admission to medical school. In this second part of his opinion, Powell would give new meaning to the word “diversity.” He would trigger a dynamic that would reconfigure the administrations of our leading colleges and universities. Race could not supply a ground, he said, for favoring one race at the expense of the other—unless it was done at the margins. Race might be “taken into account,” but as one among a host of “diverse” criteria that could weigh in a candidate’s favor. With this decision, “diversity” became the mantra that gave a new veneer of justification to racial preferences. Beneath that veneer, the system of racial preferences would become ever more entrenched, enforced by a growing cadre of administrators bearing labels such as “diversity coordinator.”

But just as there was no principled ground for choosing one race over another in the carnival of racial claims, there was no principled ground for choosing, among the array of “diverse” groups, the features that should make some candidates rise to the top in the crowd of applicants. Yes, the school could need a quarterback and a flute player, but how would one weigh an aspiring actor in the dramatic arts against a black child coming from a dangerous neighborhood? Faced with such a wide array of choices, could there be any quibbling with a director of admissions who finally confessed that he finds the race of the applicant far more compelling as a ground of judgment than anything else? The elaborate contrivance of “diversity” could be easily seen through as the most transparent fig leaf. But it would go on covering a scheme of racial preferences with moral pretensions ever more elevated.

And so it was understood by the eight colleagues who sat with Powell on the bench. Four of his liberal colleagues would not join with him in the first part of his opinion, as he came close to making the strong case in principle against racial preferences. But those colleagues who could not join him in rejecting racial preferences quickly moved into support for Powell in the second part of his opinion. For there he was offering the kind of fetching rationalization that would rescue racial preferences not only for this day in court, but into the next millennium. In this distracting dance of an opinion, the colleagues who were willing to join Powell in the first part of the argument now swung into opposition to the second part of his opinion. The result was one of those truly rare cases with one judge writing for the Court, along with eight justices in dissent. Powell was the only member of this accomplished Court who supported both parts of the opinion. That made him precisely the swing vote who could determine the outcome. Eight colleagues had registered their own judgment that the opinion in Bakke simply could not hold together as a coherent opinion.

For the judges to come up with a scheme of “taking race into account” in dispensing benefits and disabilities, either they must have failed to grasp what is deeply wrong in racial discrimination or else they had become enthralled with the kinds of theories that can be served up as the occasion requires by clever people. As Jefferson said, the ploughman is more apt than the professor to get the moral questions right because he will not be dazzled and distracted in the same way by “artificial rules” or theories. Theories are the products of the educated classes. The judges and lawyers have produced a legion of rationalizations while ordinary people, uninstructed in those theories, seem to have recognized all along that there is something not really right about the government’s rationing out benefits to some and disabilities to others solely on the basis of their race. Lincoln remarked in another one of his simple and enduring lines, “As I would not be a slave, so I would not be a master.”³² Since he rejected slavery in principle, his rejection was utterly indifferent as to whether he stood on the advantaged or the disadvantaged side of that relation. If racial preferences are wrong in principle, their wrongness will be quite detached from the question of whether we happen to be the beneficiaries or the victims of the policy. And the acquiescence of the victims no more attenuates the wrong than does sheepishness on the part of the beneficiaries.

In that vein, the lingering last line must come from the dissent in the case of Parents Involved in Community Schools (2007) by the redoubtable Judge Carlos Bea. A community with educated and liberal families had sought to apply its collective genius in a strenuous way to arrange the assignment of children to schools by race: with all of the intricate schemes of ranking and fine-tuning, race would still be the tiebreaker, the decisive test in assigning children to schools. The liberal Ninth Circuit was content to sustain this scheme, but the Supreme Court would come down on the side of Judge Bea. Chief Justice Roberts wrote for the Court, and he offered the most telling compliment to Judge Bea by ending his opinion with an echo of the words that Bea had set down as the culminating line of his own: “The way to end racial discrimination is to stop discriminating by race.”³³ That may be, after all, the last word that need be spoken on the subject.