Some things that we are able, or believe we are able, to recognize readily are nonetheless not easy to define. Most people, for example, use the word ‘play’ often and believe that they can recognize play readily, but would have to think hard to produce an adequate definition of play. I believe ‘racism’ is similar to ‘play’ in this way, as a term which is often used and which many believe they can recognize readily, but which nonetheless can be difficult to define. An attempt to define something, however, even if the attempt is difficult and produces an imperfect result, can help us understand the term in question, make our use of the term more precise, and can help to eliminate at least some problems or abuses in that use.
One might begin to think about racism by considering a provisional answer to the title question, ‘What is racism?’ Our provisional answer is: ‘Racism is an irrational hatred of, or negative prejudice concerning, or animus against, a person or group on account of race.’ How adequate is this definition?
It seems very difficult, if not impossible, to define ‘racism’ without using the prior category of ‘race’. The provisional definition just offered uses ‘race’. But ‘race’ itself is difficult to define. ‘Race’ has in fact been used very loosely to refer to such diverse things, among others, as nationhood (‘the British race’), religious belief (‘the Jewish race’), skin color (‘the white race’), language group (‘the Slavic race’), and more broadly still cultural background (‘the Latin race’ or ‘Hispanic race’).
These categories include distinguishing factors that overlap: a woman might, for example, be a Jewish Portuguese-speaker from Brazil all of whose ancestors came from west Africa. Some of these categories are matters or accidents affected by choice and human will, such as religious commitment. Speaking carefully, however, most people probably would define ‘race’ more in terms of genetic factors rather than cultural or personal accidents such as language, religion, citizenship, or political allegiance.
At this point, it seems clear that the subject is not simple. While it may seem obvious that a blond Finn belongs to one race and a black-haired Thai to another, the distinction can quickly be complicated: what of the child of a Finnish and a Thai parent? What characteristics are most important in defining race? I had a friend from Colombia once who asserted that Colombians had no racial prejudice or even racial consciousness. Sometime later this same friend indicated that there were in Colombia dozens of terms for various racial combinations (one half black, one quarter Indian, one quarter European = a particular term; one half Indian, one half European = a different term; etc.). I concluded, my friend to the contrary notwithstanding, that any nation with dozens of terms for various racial categories was not without deep racial consciousness.
In South Africa I have met people who were considered ‘white’ or ‘European’ by the apartheid regime but also were considerably darker skinned than other people who were considered ‘Coloured’ (that is, in apartheid reckoning, mixed-race). During World War II a late American friend of mine was in South Africa in the United States Navy. He was confused by signs indicating that some facilities were for ‘Europeans’: ‘I knew I wasn’t European,’ he said. ‘European’ is a term that names an accidental quality, namely geographical origin, which may or may not be appropriate for a native of North America. Likewise, ‘African’ is a term, of European origin, that embraces Arabic-speaking Libyans, Berbers in Morocco, Dinka pastoralists in South Sudan, Ashanti businessmen in Ghana, Afrikaans-speaking farmers in South Africa, and Hindi-speaking lawyers in east Africa. ‘African’ cannot really be a place-holder for race, given that huge diversity.
If the broad category of ‘race’ is complex, would it be helpful to break down the subject by concentrating on specific qualities or characteristics? If the quality is language, wealth, or education, we are back to consideration of accidental and personal qualities. Hungary has produced an unusual number of first-class mathematicians, for historical and cultural reasons that are mostly readily explicable. Few people, however, would think that a higher than average percentage of highly numerate Hungarians is a racial fact rather than a national accident.
But what if the quality or characteristic under consideration is more genetically based and intrinsic: say, height and lung capacity, and thus aptitude for long-distance running? When reduced to such qualities, it certainly is possible to make some broad generalizations about some groups. So, for example, people from the highlands of Kenya and Ethiopia are typically better at marathons than people from Cambodia or even from west Africa. While granting that such distinctions are largely genetic, how helpful are such generalizations in a broad discussion of race?
At this point, I think it is clear that the matter of ‘race’ is so complex as to require at least a partial retreat. ‘Racism’, as the term is commonly used now, is not a phenomenon strictly focused on ‘race’, whatever ‘race’ may be, but rather refers to a looser set of categories that may involve skin color but also involves language, socioeconomic status, and even self-understanding. I shared a house in college with a Jewish classmate whose younger adopted sister was of mixed race. That sister might certainly later in life have enjoyed advantages and suffered adverse discrimination due to her various backgrounds and qualities. To parse out all of the potential (or, later, the actual) advantages and disadvantages would be very challenging indeed. But certainly that child might later have experienced invidious racism and hostile discrimination as well as enjoyed many blessings from her family background, educational opportunities, and genetic makeup.
In the United States and much of the world, ‘racism’ is not usually treated or defined by itself or in isolation but rather tends to be grouped with a set of discriminations that are either outlawed or discouraged. ‘Racism’ is part of a list, which tends to include overlapping categories and to be very broad and encompassing. A government might prohibit for various purposes, ‘discrimination on the basis of race, skin color, national origin, language, religion, sex, sexual orientation, age, and appearance.’ If ‘race’ by itself is hard to define, most of what anti-racists wish to discourage is surely encompassed by ‘race, skin color, national origin, language’ and the rest.
At this point, consider an additional complication. ‘Discrimination’ can include positive as well as negative or invidious behavior or treatment. Our provisional definition, to deal with this complication, included two formal terms (‘irrational’ and ‘negative’). To ‘discriminate’ is a neutral act, and some discriminations are rational and desirable. In the United States, for example, screening for sickle cell anemia ‘discriminates’ between blacks and whites, because one group is susceptible to the disease and one is not. So also the Red Cross in its blood drives discriminates among potential donors and discriminates against groups at higher than average risk for HIV infection, such as sexually active male homosexuals, illegal intravenous drug users, and those who have paid for sex. Likewise, discrimination between males and females and the young and the old are common because of actual differences, which can be rationally explained, between the sexes or age groups. Such discriminations are often considered reasonable and desirable and directed towards positive goals.
In the United States and elsewhere more controversial forms of discrimination have been or are often justified as benevolent and as correcting or counterbalancing past discrimination. This kind of benevolent discrimination clearly is race-conscious and, in a large society, almost always paints with such a broad brush as to produce at least some debatable outcomes. Nigerian Americans are the best educated ethnic or national group in the United States. South Asian Indian Americans are the highest income ethnic or national group in the United States. Both Nigerians and Indians are ethnic and racial minority groups and, plausibly, can claim to have suffered historically from invidious discrimination as such. Despite that history, does it make sense to give advantage for purposes of university admissions, for example, to the child of two Nigerian physicians? or to the child of an Indian corporation president? But if being Asian or African, or having dark skin, or being a member of an historically disadvantaged group does not by itself justify special advantages, what does? Perhaps actual, personal socio-economic disadvantage is the best category for parceling out such benefits, if they are to be officially or publicly or governmentally distributed: but in that case, race is only a correlating factor and not the deciding or most important matter.
In any case, the proponents of ‘affirmative action’ or ‘economic empowerment’ for historically disadvantaged groups, both present themselves as supporting benevolent, positive, and desirable things, even if they also undeniably involve making distinctions on the basis of race, national origin, and similar categories which are often treated as publicly suspect.
Perhaps, then, the problem is not discrimination as such, but rather invidious or hateful discrimination. Our provisional definition included the formal elements of ‘irrational’ and ‘negative’. But such elements, and others such as ‘invidious’ and ‘hateful’, are subjective and are patient of differing interpretations. Even a casual reader of Gone with the Wind or other literature of its era will observe both implicit and also explicit justifications for racial discrimination that present such discrimination as positive and benevolent: a claim to white domination presented itself as paternal care for an immature people incapable (for the time being? permanently?) of self-government.
Such paternalistic racism often was, no doubt, a thin cloak for rather obvious, invidious, harmful discrimination and oppression. But even if the paternalism was subjectively sincere and benevolent in intention, how much weight should be given to a subjective self-justification and self-conscious intention in a discriminator? The addition of formally negative elements to a definition (‘irrational’, ‘hateful’, ‘harmful’, ‘invidious’, ‘malevolent’) simply pushes the definitional problem back one step. What is irrational and harmful discrimination, and why? And if we seek to exclude paternalistic forms of racism, how can we distinguish them from affirmative action and economic empowerment? Are not these simply more and less recent forms of the same kind of racist discrimination?
There are, then, both legitimate and illegitimate, reasonable and unreasonable, moral and immoral, forms of discrimination involving race and related categories. The difference cannot be reduced to the intention and subjectivity of the discriminator without admitting subjectively sincere, kindly intended racial discrimination even of the most systematic and damaging sort. But if the difference is not in the intention of the agent, it is difficult to say wherein it does lie.
I had a friend who was a head teacher for first grade students in a public school. Jan had three teachers with whom she worked, one of whom she considered incompetent. The incompetent teacher was tenured. Many of the students were from poor, disadvantaged backgrounds, and those disadvantaged students in turn were disproportionately African American. Jan assigned the academically strongest first grade students to the incompetent teacher, with the expectation that those students would do fairly well if only because of family advantages. Jan, who thought she was the most competent of the teachers, took the most disadvantaged students into her own class and had, in consequence, a disproportionately large number of African American students. Jan assigned to the other two teachers the middling students. Jan clearly did discriminate, and she did so in part on the basis of race or factors that correlated strongly with race. Jan’s intention was to help children, particularly the most disadvantaged. Was that racism? Should Jan be condemned or lauded?
At this point we might conclude: 1. that ‘racism’ by itself is very difficult to define, because ‘race’ is very difficult to define; but that, 2. that particular problem is made less acute than it otherwise would be if we consider race as part of a broad bundle of categories and factors. We also might conclude that, 3. while we are reluctant to rely on subjective attitudes in a definition of racism, because racism can in fact present itself as, and perhaps even actually can subjectively be, benevolent, nonetheless it is virtually impossible to avoid reliance on subjective matters of intention and self-understanding.
These conclusions do not solve the definitional difficulties surrounding the issue of racism. The second conclusion above, however, does show that those difficulties can in part be finessed and made less acute or practically significant. Are there other such finesses available?
One possible approach is to adopt a libertarian and individualistic approach to the public policy questions that quickly arise when race is involved. That is, for public purposes we might refuse to acknowledge racial, and many other kinds of, groups as such. Individuals would only be treated as themselves, without reference to their background and group memberships or identities. This is the ‘color-blind’ approach asserted on billions of pieces of public stationary which, after passage of the U.S. Civil Rights Act, asserted that ‘The University of _______ hires without regard to race, color, sex, religion, national origin etc.’. That the assertion was almost always blatantly violated in fact by the institutions in question only proves that those institutions were lying, not that the approach they claimed was or is wrong or impossible as a real guide to behavior.
The color-blind approach clearly will produce some results that are unfair, in the sense that individuals from backgrounds of, for example, material affluence, greater educational opportunities, and more stable family background will have undeniable advantages. If we begin by assuming that the world has much evil in it and that there are many unfair things in life, the question is surely how best to minimize the effects of the evil and to overcome as much as possible unfairness?
Any approach to the problems posed by such initial inequalities will involve controversial and debatable matters of public policy, and different people of goodwill will come to different conclusions. But it is plausible to suggest that the best and quickest way to overcome the negative effects of past and invidious discrimination, including racist discrimination, is to permit market mechanisms to work.
For example, suppose that businesses historically have discriminated against brunettes and that the law permitted, or even mandated, such discrimination. Suppose, again, that a society comes to conclude that such discrimination is immoral and wicked and then asks itself how best to overcome the effects of past discrimination.
One might seek to overcome the effects of past discrimination by legal fiat and then by positive discrimination to counterbalance historical discrimination and, as it were, to bring the scales back into better balance. This affirmative action has the disadvantage of itself being hair-color conscious and of extending consciousness of and sensitivity to the category in question. This approach also may well stir up resentment among the non-brunette and extend an element of self-doubt in brunettes. It might not be effective without very heavy-handed legal intrusion into private arrangements, and even with such intrusions might prove dishearteningly slow and ineffective.
An alternative approach in regard to future behavior would rely on market forces to overcome the effects of past discrimination quickly. The state might outlaw state-enforced discrimination and otherwise let markets overcome the problem. Businesses that discriminate against a large category of potential customers will suffer a relative and competitive disadvantage as compared to businesses that willingly do business with all. A tire manufacturer, for example, that refuses to sell tires to brunettes who own tire dealerships sets up for itself an irrational, economically harmful restriction. Likewise, a shop that refuses the patronage of brunettes will suffer for its refusal. Such market considerations probably will more quickly and efficiently punish irrational discrimination than governmental fiat will. Furthermore, the market is relatively impersonal and will not create a sense of grievance and resentment in those who in a changed world may view themselves as the victims of a new kind of discrimination.
Such policy observations are certainly debatable. What seems much more clear, however, is that ‘racism’ is a complex and difficult term. At the least when ‘racism’ is spoken of, it is reasonable to ask the user to define and justify the term and his usage thereof. Without such definition, one must suspect that the term is used as a kind of word magic and does not really clarify much.