35. COMPROMISING THE UNCOMPROMISABLE: DISCRIMINATION

SHOULD PEOPLE BE FREE TO DISCRIMINATE AGAINST OTHERS?¹ At first glance, the only proper answer to this question would appear to be a resounding (and horrified), “No!” However, on further reflection, things are not quite so simple.

A few preliminary remarks. The liberals, who are most associated with this lofty reply in the public mind, by no means give total assent to it. Their protestations notwithstanding, they actually favor discrimination on racial² grounds—as long as it favors the underdog vis-à-vis the “overdog.” They might not choose to accept this description, but their advocacy of affirmative action, quotas, “goals,” set-asides, preferences, norming, and all such other programs amounts to discrimination—no more, no less. They of course do this for the “best” of reasons, or at least for reasons that seem good and sufficient to them, but the same might be said of anyone at all who acts (Mises 1966). In summary, “progressives” favor discrimination on behalf of downtrodden groups as a matter of principle, and, presumably, act in this manner in their own lives.³

What of the conservatives? Nowadays, they speak out in favor of something called “color blindness” (Bolick 1996; Eastland 1996). This amounts to a restoration of the Civil Rights Act of 1964, and Martin Luther King’s dream of a society in which decisions are made on the basis of people’s characters, not the color of their skins.⁴ This, too, sounds nice. What could be more fair or just? Treat all people based on their merits.

But this position is also philosophically flawed. First of all, merit is merely another characteristic on the basis of which people can discriminate against each other. Because merits of different types and varieties are statistically correlated with different groupings, including racial, preferences for people on this basis are not at all distinguishable, at least in effect, from choices made with respect to race or color. If true colorblindness is the goal, allowing discrimination on the basis of merit will hardly achieve it (Gottfredson 1987, 1988; Herrnstein and Murray 1994; Levin 1997; Rushton 1988; Seligman 1992; Sowell 1975).⁵

Secondly, neither liberals nor conservatives are really serious about this goal, at least for the personal if not the business arena. If colorblindness is truly virtuous, and for some reason should be required by law, then why confine this compulsion to one or another aspect of our lives? Why not compel it throughout? If a person must hire employees on the basis of character or merit or on the basis of anything but skin color and must apply this criterion to firing, investing, selling, buying, and so on, then why should this basic element of morality not be applied to private life as well? Why not prohibit racial discrimination in the choice of friends, dating partners, or marriage partners? The conservatives (at least nowadays) talk a good colorblindness line, but their adherence to it is only skin deep. Even the conservatives jettison the whole perspective as applied to their whole lives.^6,7

LIBERTARIAN COMPROMISE

What, then is the Libertarian Compromise? For this position, we must make a sharp and deep distinction between the public and the private sectors not only of the economy but of the society as well. The libertarian sees an unbridgeable philosophical chasm between the government, which is necessarily based on force and compulsion, and the private sphere, which consists of voluntary interaction.⁸

This being the case, the compromise must come in two sections, one devoted to each of these spheres. In the private sector, the rule is not race preference to the underdog of the progressives, nor is it the colorblindness of the conservatives. Instead, it is one of completely voluntary association, where each person is “free to choose.”⁹ All people are or should be free to make up their own minds about how they will choose their spouses, friends and neighbors, customers, employees, employers, and other business associates. The clear implication is that the Civil Rights Act of 1964 must be repealed, as it is inconsistent with the total freedom to choose one’s associates in all spheres of life.

Some people might recoil in horror from turning the clock on race relations back to the pre-1964 period. They would object that if a majority were free to discriminate against a minority, the latter would be greatly disadvantaged. That is, if, for example, whites, were to refuse to buy from, sell to, hire, work for, invest with, for example, blacks, the latter would be unemployed, homeless, and starving.

But this position is economically erroneous. All such scenarios fail to take into account the market’s fail-safe mechanism that helps those subjected to discrimination. Consider employment. If white racists rebuffed black workers, the first effect would indeed be unemployment or lower wages for the latter group. But this situation is only temporary, a mere first stage in the mental experiment we are now considering.¹⁰ For with lower wages or greater unemployment, some whites¹¹ would be sorely tempted to employ these blacks, because they can earn additional profits exploiting workers who are underpaid or idled.

Consider a two-stage numerical example. Initially, there is no racism, and white and black workers are all employed at $10 per hour (Table 1). We assume that marginal revenue product, marginal productivity, or just plain old productivity was also equal to $10 for both groups; thus we start out at a competitive equilibrium. All employers are of course seeking profits, but, because they rival for workers, wages are bid up to productivity levels and no profits are actually made.

Now, in the second stage, we introduce racism into the analysis. This means, if it means anything, that while the demand curve for white workers remains intact, that for blacks shifts to the left (Figure 1). This, in turn, implies that while white wages and employment levels remain as they were before, the number of blacks hired, and their compensation levels, falls.

Let us suppose that it decreases to $7 per hour (Table 2). But this, clearly, is untenable and cannot endure for long. If an employer hires a white, he earns zero profit;¹² in contrast, if he hires a black, he will net $3 additional per hour. Some employers, blinded by their race prejudice, will still insist on hiring only whites.¹³ However, in the competitive struggle, those who either have no race prejudice, or choose not to indulge in it, will hire the equally productive blacks¹⁴ at $3 per hour less. On this basis, they will be able to manufacture the product at lower cost. Then, they will either keep prices the same and pocket additional profits, or lower their prices. In either scenario, there can be only one result: The racists will be driven out of business.¹⁵

A similar analysis could be applied to every other arena of commercial endeavor. If blacks are not sold houses (or food, or clothing), the price they must pay will rise, making them better buyers from the point of view of lenders,¹⁶ others will jump into the breach. If blacks are not given loans, the interest they will be forced to pay will increase, rendering them better borrowers; others, blacks as well as whites, will be more than happy to “exploit” these blacks, by lending to them.¹⁷

But is this not unfair to blacks? Why should they have to endure the indignity of lower wages and unemployment (or higher prices for food, clothing shelter, loans, etc.), even if it is only temporary? One answer to this very reasonable challenge is to realize that the enemy is not the market, which is riding to the rescue of the downtrodden group (by first allowing it to suffer, and then, in effect, making this suffering the key to their economic salvation). Another perhaps better answer is that this scenario is a hypothetical construct, articulated in terms of two stages, separate in time, and mainly for heuristic purposes. That is, to clarify the process, we purposefully assume that there would be two stages; in the first, the position of blacks is worsened, to show that in the second they would be rescued. In actual point of fact, there are no such two stages. Any time the wages of blacks (or anyone else) dips below their productivity levels, even by a tiny amount, there are immediate profit incentives to hire them, which starts their wages up on an upward spiral back toward equality.¹⁸

Let us now summarize this part of our discussion. For the libertarian, there is a compromise between the conservative’s call for colorblindness and the liberals advocacy of equal representation of all groups in all possible niches of the economy. This compromise is, simply, that no one is forced to discriminate, but no one is prevented from so doing either. People are totally free to choose. There is complete freedom of association, and this applies to both the spheres of business and personal life. This is as a matter of right. As a matter of utility, the claim which emanates from this sector of the political-economic spectrum is that discrimination is impotent to do any real damage to its supposed victims. This is because any time any group is subjected to discrimination, its economic status makes its people more acceptable to others, and these others can bankrupt the discriminators.

PUBLIC SECTOR

Now let us move to the second stage of our analysis, and consider the libertarian compromise in the case of the public sector. For the libertarian anarchist, this compromise is a very simple one. Because there is no public sector at all, the question of whether it should or should not discriminate on any basis simply does not arise.

The response from the limited government libertarian is greatly simplified as well. For this position, the application of the question is very much reduced (although not completely eliminated); it no longer applies to virtually the entire economy, as at present, but instead is limited to police, armies, and courts. In this system, all people are forced to pay taxes for the upkeep and maintenance of the government; therefore, all of them should have an equal legal right to work for it in these three realms. Here, the libertarians borrow a leaf from conservatives: colorblindness and merit.¹⁹ This does not mean, however, that all racial, sexual, or other designated groups should be assigned jobs in proportion to their percentage of the population, as the liberals would have it.²⁰ Rather, as the purpose of government in this philosophy is to maintain civil order and protect person and property, then the people who can best achieve these goals, whatever their background, should be hired for them.

AN OBJECTION²¹

The implication of this paper is that the Civil Rights Act of 1964 was unjust,²² was enacted to correct a problem that was not really a problem, and was, therefore, unnecessary.

But an immediate objection arises: Can we not think of this legislation as a form of reparations? After all, suppose you are a member of a minority race and a majority comes along and captures you, forcing you to work as a slave. This system lasts for hundreds of years. Then, happily, you and your people are freed. After that, you are allowed to trade and exchange freely, but virtually all of the means of production have been taken up by others; more specifically, the land ownership rights are almost totally owned by the majority. Why should you consider this allocation of property rights a legitimate starting out point for trade and exchange? Would justice not consist of the majority’s giving up some of its wealth, property, and privileges, by force if need be, so that the minority could reclaim at least some of the fair market value of what had been confiscated from them?

The starting out point, then, is crucial for the analysis of markets. If the minority were brought back to the status quo ante,²³ then, perhaps, the argument might go, there would be no need for the Civil Rights Act of 1964. But, as this has not occurred, that legislation is justified, at least as a second-best policy.

The liberal-egalitarians would deny there is any need for reparations. They think that income should be equalized among all people in any case. So, there is no warrant for transferring funds to them based on considerations of the past. Suppose there were two groups of poor people, one victimized unfairly in the past, the other not. The consistent egalitarian would not wish to elevate the former over the latter; rather, he would want to bring up the income of both groups, so that it attained the level achieved by everyone else. The conservatives too would deny there is any need for reparations. Some would argue that none of the people alive in 1964, and even more certainly not at the time of this writing, were actually slaves. They are only the descendants of such people. Therefore, restitution should only be given to those who were direct victims of injustice; as there are no such people alive any longer, compensation for slavery is not justified. Others correctly see the demand for restitution as a demand for justice. However, remarkably, they either deny that there is any such thing as justice, or maintain that “the search for justice will destroy the world.”²⁴

In contrast, the libertarian position is that amends of the income transfer variety are most certainly justified. For example, I advocated just this in a case analogous to slavery, stating (Block 1985, p. 496):

According to the [John Lockean] theory of property rights, the peasants who tilled the soil are the rightful owners of the land. The conquistadores who conquered them stole this land. Their descendants, many of the large land owners in South and Central America, thus hold unjust title to their land.²⁵

Note that not every type of reparation is justified under libertarianism. For restitution to be proper, it must meet certain criteria. Foremost, the land or other valuable consideration must be taken away from the actual thieves or from their descendants,²⁶ not from innocent people, or all people including the innocent. Second, the burden of proof always rests on those who would overturn extant property titles. As the lawyers say, possession is nine-tenths of the law. If redress is due, the claimant must offer proof, not mere allegation.

Unhappily for the objection we are now considering, these criteria do not apply to the Civil Rights Act of 1964 nor to any welfare scheme that transfers funds, on net balance, from whites to blacks in the modern era,²⁷ because forcing present day whites to integrate with blacks, or to in any other way give up their right of free association, is to punish innocent people for the evil deeds of their grandparents. But in the libertarian legal code, a person can be punished only for his own rights violations. In any case, there are many whites now in the country whose grandparents were in Europe or in Asia and had absolutely nothing to do with slavery.

In sharp contrast, however, there is a strong case that can be made for the transfer of some property now owned by whites to some blacks. Had justice been attained in 1865, the slave owners would have been punished²⁸ for the crime of slave owning.²⁹ Certainly, their property would have been seized from them and given over to their ex-slaves, as (partial) compensation for the years of misery inflicted on them. The slave owners, unfortunately, are beyond the reach of present day justice. However, their lands, plantations, other holdings, never should have been given to their children, and, through further inheritance, to those who hold these properties at the turn of the 21st century. In justice, they should be given back, provided that specific present day blacks can prove they are the descendants of slaves who worked on precisely located land. With historical records, including notations on family Bibles, this should not prove impossible. Of course, it would have been somewhat easier to offer such proofs in 1964, because records tend to become lost as time passes.³⁰

Levin (1982, p. 85) addresses the dichotomy these two options present, saying, “The difference between the two policies is the difference between restoring a robbery victim’s property to him, and hunting up the descendants of robbery victims and giving them goods at the expense of people who robbed no one.” Levin “has no quarrel with the former, many quarrels with the latter.”³¹ But we can see that he must necessarily have a quarrel with the libertarian view of reparations. The descendants of the white slaveowners robbed no one. According to Levin, it is therefore unjust to find the descendants of slaves and give them the plantations and other property on which their great-grandparents worked. However, even though the present white holders of these plantations themselves robbed no one, they never should have been given these lands in the first place, and, in justice, ought to be made to give them up in favor of the great-grandchildren who would have inherited them, had their slave ancestors been given them in 1865. These people might not have robbed anyone, but they are in effect (innocent) holders of stolen property, and must in justice be forced to disgorge them. But this holds true if and only if their rightful owners can be determined, and the burden of proof rests with the latter.³²

_____________________

American Journal of Economics and Sociology 57, no. 2 (April, 1998): 223–37.

¹ Discrimination occurs on the basis of race, gender, sexual preference, national origin, handicap status, baldness, citizenship, beauty, intelligence, height, weight, sense of humor, left or right handedness, and dozens of other criteria. The law singles out the first five of these for special protection, leaving the “victims” or “undesirables” in the latter categories to fend for themselves. Presumably, public choice considerations (Buchanan and Tullock 1971, 1980; Gwartney and Wagner 1988), not any intrinsic characteristics of these groups, can account for this selection bias.

² Although we mention race alone in the text, this could stand in for any characteristic on the basis of which certain people are distinguished from others and are reacted to in different ways.

³ Or not. Many liberals champion racial integration yet live in white-only neighborhoods themselves; others favor public schools but (e.g., the Clintons) send their children to exclusive private schools.

⁴ Conservatives, of course, did not always pursue this policy. Once upon a time, they were in favor of discrimination—racial preferences—for the overdog, on a compulsory basis, for example, Jim Crow legislation. (At this time the liberals favored colorblindness.) Then the conservatives went through a period of advocating separate but equal separation. Only lately have the conservatives adopted the liberal democratic policies of the 1960s.

⁵ To be fair to the conservatives, they would be philosophically satisfied if, by adhering strictly to merit, racial disparities were to arise on a statistical basis. The explication is that conservatives see colorblindness as an end in itself, not merely as a means (e.g., to a statistically representative society), as do the liberals.

⁶ Were colorblindness applied to personal life, then the state, before granting a marriage license, would have to query citizens as to whether they had dated people from other races. If not, why not? If blacks are 15 percent of the population, and Jews are 3 percent, then each should intermarry with all others at about that percentage. If not, racism (e.g., non-colorblindness) is presumed to be the cause. (This is true at least if we follow the same pattern in personal life as we do for business practices). Surely, no rational commentator reflecting on this issue for one moment can deny that non-colorblindness is rife in personal relationships. If the civil rights law of 1964 is morally required, and no relevant difference between commerce and personal relations can be adduced, then it should be used to stamp out these racist patterns. If this is not a powerful reductio of the whole idea, then there is no such thing as a reductio.

⁷ Another “feel-good” notion is that of equality of opportunity rather than outcome. But what, precisely, does this mean? If all it means it that the government must not interfere with the free choices of individuals, it is entirely vacuous. More exactly, it means, then, no more than free enterprise, and we already have a perfectly good phrase to express that idea. On the other hand, if it means more than this, if it is applied to race or sex discrimination, it is a positive evil. This can be shown by subjecting it to the personal life test. To wit, equality of opportunity must mean that, for example, heterosexuals must date others of their own gender (we’ve got to give these people an equal opportunity to marry us, after all). Similarly, homosexuals must be forced—this is the law, after all—to date members of the opposite sex. As well, whites and blacks must be forced to date one another. None of these three groups need marry each other (that requirement can be deduced from a law prohibiting discrimination in marriage, not from “equality of opportunity, not of outcome”), which would be to compel equality of outcome, not merely opportunity. Nevertheless, it is entirely noxious, at least for the libertarian, to force people to do anything against their wills (apart from requiring them to keep their hands to themselves, of course).

⁸ From this fact, the anarcho-libertarians (Benson 1989, 1990; Friedman 1989; Hoppe 1989, 1992; Rothbard 1973, 1982) draw the conclusion that government must be banned totally, root and branch. In contrast, the minarchist libertarians (Epstein 1985; Nozick 1974) deduce that “government is best which governs least.” Arguing from the premise that a certain minimal amount of coercion is necessary for the functioning of society, this latter group maintains that if force cannot be eliminated, at least the state should be limited to the basic tasks of protection of person and property, for which government, courts, armies, and police are needed, and nothing much else.

⁹ The title of Friedman’s 1981 work.

¹⁰ For an economic analysis of discrimination, see the works of Becker (1957); Block (1982a, 1982b, 1992); Block and Williams (1981); Demsetz (1965); Higgs (1977); and Sowell (1975, 1981a, 1981b, 1994).

¹¹ Also, blacks with savings who were on the margin of being an employer or an employee, plus those blacks who were already employers.

¹² We assume the employer can just stay in business, even with zero profit, because, by assumption, the employer still can do just as well as the next best alternative, say, working for someone else, and investing his or her capital in another business.

²³ We assume away as a complication that does not affect the main line of analysis the possibility that customers or other employees object to being served by, or working next to, blacks. On this see Becker’s (1957) work.

¹⁴ If blacks are not equally productive, then that will account for their lower wages, not discriminatory behavior on the part of their employers, as we have been assuming.

¹⁵ This is according to the old saw that there is a sucker born every minute. This may well apply to racists too, that there is a racist born every minute. If so, racism will always exist, even though the market continually weeds them out. This is one answer to the question of why racism still exists if these beneficial effects of the market are really operational. The other reply is that full free enterprise—laissez-faire capitalism—does not now apply. Instead, the government interferes in all sorts of ways, the effect of which is harming blacks, for example, the minimum wage law, unionism, laws compelling union rates of pay, and so on. On this see the work of Williams (1982).

¹⁶ We here implicitly assume that the costs of servicing these two groups (pilferage rates, etc.) are equal. If they are not, then this factor, not racism, would account for the longevity of differential prices.

¹⁷ We here implicitly assume that the likelihood of paying back the debt is invariant between white and black. If it is not, then this factor, not racism, would account for the longevity of differential rates of interest.

¹⁸ Why, then, do wages of blacks lag behind those of whites, and females behind males? It is because, contrary to our assumption in the text, productivities are not at all equal. In the case of racial differences, several theories account for the gap—the history of slavery, the Jim Crow law and other incidences of statist repression, and lower IQ (Herrnstein and Murray 1994, where they control for this variable and find that the pay gap has all but disappeared). In the case of differences between the sexes, empirical research suggests the differences are due to differences in marital status: Marriage lowers female productivity in the market and raises that of males. See the work of Block (1982a), Block and Williams (1981), Levin (1984, 1987), and Sowell (1975), which show that when marriage is factored out of the equation and the wages of never-married males and females are compared, there is virtually no wage gap.

¹⁹ The only exception is if skin color or gender are relevant to doing the job. For example, if the police want to infiltrate, for example, a black gang or the Mafia, then they must use, say, blacks or Italians. Obviously, to prevent female prostitution, female decoys would have to be used. (However, because prostitution would be legal in the libertarian society, this is a moot point.)

²⁰ The liberals claim that a logical implication of total non-discrimination would be a proportional share of all jobs, housing locations, income, wealth, etc., to all racial, ethnic, sexual, etc., groups. That is, if blacks are 15 percent of the population and women are 51 percent of the population, then blacks should comprise 15 percent and women should comprise 51 percent of all professors, doctors, merchants, actors, and so on. For a critique of this view, see the works of Sowell (1975, 1981a, 1994). Interestingly enough, the liberals never apply this theory to high-status and high-paying occupations in which blacks are statistically overrepresented (e.g., the National Basketball Association, the National Football League, etc.).

²¹ I wish to thank Laurence Moss for forcibly bringing to my attention the importance of addressing this issue, and, as well, for the form of this objection, which is based, almost verbatim, on his words.

²² Support for this contention is given by Levin (1982, p. 85), who analyzes the situation of “giving someone a job he was denied because he was discriminated against” as follows:

This is current policy, as reflected in the Civil Rights Act’s ban on racial discrimination in employment. … It is, however, arguable that employers do not owe anyone jobs. An employer, on this argument, is choosing whom to give his job to, and is entitled to choose on criteria that most people would find objectionable. His right to liberty overrides his obligation not to indulge prejudice. If so, the Civil Rights Act curtails freedom impermissibly.

²³ This is impossible, strictly speaking. But if they were at least compensated by a good faith effort.

²⁴ This was Milton Friedman’s recitation of a favorite saying of his “old teacher” Frank Knight. Friedman (1985, p. 498) interpreted Knight as saying: “Justice is in the eye of the beholder,” and went on to assert on his own account in 1982 that

there are no really objective standards of justice. And there’s no way other than force, ultimately, of mediating different claims of justice. It’s a search for justice that animates Khomeini’s Iran today. … So, I believe that it is very dangerous to base any judgment of social policy upon the objective of searching for justice.

²⁵ My response (Block 1985, p. 498) to Friedman was to

take the latifundi in the Third World where, as far as I am concerned, the historical facts show that it was the conquistadores who, at one time, took over the land, kicked the peasants off, or allowed the peasants to stay there, but claimed ownership of it. And, on the other hand, you have a bunch of peasants who had, according to the Lockean theory, been the true owners of it. Now, if you say that there is no such thing as justice, and we must couch everything in terms of what is, then clearly the people who are working as peons there have no right to take over the land which I contend should really belong to them.

²⁶ David Friedman (1985, pp. 501–02, 505) makes the following response to me:

The point where I would want to agree with my father [Milton Friedman] … against Walter Block or at least make an argument on [the side of the latter], goes back to the latifundi. It seems to me entirely possible . . . that as a matter of abstract justice, if I were a judge in a court, I would agree that the peasants were in the right. But second, that it would be better for the world, including the peasants, if they forget about the past. Fighting over their claims to justice will get people killed. There is no particular reason to think that the most just people will win the war. Fighting tends to create unjust situations. So, it seems to me quite plausible to argue both that here is an abstract principle of justice, which in principle could be applied; and that as a matter of practical, social reality we accept what is and work from there.

If by “justice” we are concerned with initial ownership of things, I think in the long run that isn’t enormously important. In the U.S. at the moment, if you gave the country back to the Indians, in some fair way where you didn’t give them the buildings that are built on it, but just the land; and divided it fairly evenly among the Indians, it would not noticeably affect the distribution of income in the U.S. It wouldn’t much affect how well off I am, and so forth.

Starting with either a “just” or an “unjust” distribution of property, in a generation or two you end up in not very different circumstances, except in very extreme cases.

²⁷ Here is my response to David Friedman (Block 1985, pp. 505–06):

Milton and I were distinguishing ourselves on a “normative” question. David replied in a “positive” vein, with which I happen to be in full agreement. That is, I agree that the Indians or the natives would probably be better off if instead of worrying about their lost endowments of property, they were just concerned with creating a libertarian society from hence forward.

²⁸ My point is still worth making:

Those people owned that property. … Here we are supposed to be defenders of property rights; and yet, grant me the facts of the case, massive theft has taken place; and we’re giving … a positive statement with which I happen to agree: that this stolen property is economically unimportant. But I think it’s very inadequate to (make) that positive statement. What we have to make is the normative statement, too. Both. We have to say, “Yes, in justice, that property belongs to you.” Namely, the free market advocate is not just in favor of the status quo, where blatant theft has taken place.

²⁹ I conclude from all of this that the starting point might be crucial to the descendants of black slaves, or it might not. That is merely an empirical issue. The far more important point is that they have a right to start out as fairly as possible. Fighting, in any case, is not really an issue here. In answering this question, I am implicitly considering a historical state of affairs alternative to the one which brought us the Civil Rights Act of 1964. In it, reparations based on the libertarian code would have been made instead.

³⁰ If my great-grandfather stole something from yours, and then I inherited it, my claim to it is no more justified than his. Because his claim was improper, so is mine. If you sue me for the property, then libertarian justice would require that I turn it over to you. I cannot be imprisoned, because I did not do the initial stealing. I am, in this case, akin to the innocent holder of stolen goods; I cannot keep them, but I am not guilty of theft.

³¹ Levin (1977) rejects compensation to blacks from whites on grounds very different from those of our discussion.

³² The implication of the libertarian theory of reparations, thus, is that the further back in history the initial injustice occurred, the more difficult it is to justify such acts, because in addition to the fact that the passage of time causes records to deteriorate, the further back one goes, the less likely it was that written records were kept. For both these reasons, restitution based on theft from Japanese-Americans during World War II would be far easier to make than restitution for crimes that occurred 2000 years ago in the Middle East. Similarly, it would be easier to give justice to American blacks than to Native Americans.