Mises Daily Articles
What Is Prejudice?
Comedian Chris Rock has a routine where he says,
You know the world is going crazy when the best rapper is a white guy, the best golfer is a black guy, the tallest guy in the NBA is Chinese, the Swiss hold the America's Cup, France is accusing the US of arrogance, Germany doesn't want to go to war, and the three most powerful men in America are named "Bush," "Dick," and "Colon." Need I say more?
Amusing yes, but is it prejudice? Presumably Mr. Rock is expecting black rappers, white golfers, arrogant Frenchmen, and little Chinese people. He is, in short, expecting those things that we have come to know as stereotypes.
Is it possible, perhaps, that this expectation is hindering his rational judgment? Is he prejudging the kind of rappers and golfers he should expect, and if so, what do we mean by this allegation? Should we be offended?
What about some other examples? Are we being prejudiced if we expect Irishmen to be binge drinkers, or expect Native Americans to love gambling? What if we think that women are bad at parallel parking, or that men are bad at cooking? Are we prejudging them?
Experience, Prejudice, and Group Judgment
According to some eminent remarks on the subject, prejudice is a product of our experiences. In a broadcast in December of 1955, journalist Edward R. Murrow told his audience that "everyone is a prisoner of his own experiences. No one can eliminate prejudices — just recognize them." British essayist Charles Lamb held a similar view, complaining of his "earth-bound" experience and describing himself thusly:
a bundle of prejudices — made up of likings and dislikings — the veriest thrall to sympathies, apathies, and antipathies.
Another quotation, this time from an unknown author:
We are each burdened with prejudice; against the poor or the rich, the smart or the slow, the gaunt or the obese. It is natural to develop prejudices. It is noble to rise above them.
According to all of these views, prejudice is something that we develop from our experiences in life, and which hinders our ability to make rational judgments. We are prisoners of our earth-bound and finite experience, and it is this experience that thwarts our attempts to judge others rationally, without any preconceptions. Though there is disagreement in these views as to our ability to transcend our prejudices, what all of these quotes share is more striking: a common disparagement of life experience and its subsequent effects on our judgment.
But what is prejudice? As is apparent from the etymology of the word, it is the prejudgment of something — it is a judgment occurring before it should. What does it occur before? Well, presumably, a prejudiced judgment occurs before all the relevant and available facts are in. If the fellow sitting up at the big bench with the horsehair wig sends some poor rube to the gallows before his lawyer has even had a chance to give an opening statement, then this is prejudgment of the issue, and we might rightly cry afoul — prejudice! This judge is demonstrating prejudice because he has leapt to judgment before he should have; in this case, before considering relevant information that was available to him, but that he did not care to consider.
We hear allegations of prejudice quite often. The allegation is applied liberally when people make judgments about others on the basis of group memberships. If this entails demographic considerations like sex or race, then the chorus of denunciation is likely to be great indeed — it is prejudice, stereotyping, narrow-mindedness and bigotry! (Indeed, these are often taken to be synonyms.)
Such judgments are often characterized as prejudices because they supposedly ignore the individual merits of those being judged. To judge someone by their race or sex is allegedly to ignore their inner goodness (or badness) and leap to conclusions, before we know what kind of person they are. The problem with this allegation is that individual merits are often unknown to us, and we are frequently put in positions where we require judgments about people we have never met or interacted with and know very little about. Sometimes these are snap judgments, required before we have any real opportunity to get to know those whom we judge. Other times they are judgments that are protracted, but nonetheless involve considerations of merits that we cannot assess directly in the circumstances.
In these kinds of cases, it is often perfectly legitimate to use group memberships (even characteristics like sex, race, age, etc.) as predictors for other characteristics of direct interest that are not available to us for observation — i.e., to use rational inference and discrimination. Doing so is not a derogation of the principle of judging people on their merits, so long as it involves a bona fide attempt to infer these individual characteristics from all the information that is available at the time. (In fact, it is the refusal to consider such proxy characteristics that constitutes the failure to properly assess the individual merits of the person using all the information available.)
Perhaps I'm being a bit sneaky here, and using a "persuasive definition," since I have said that prejudice refers to a judgment made before all the relevant and available facts are in. Perhaps it should only be the former, and I am just sneaking this second requirement into the concept, to load the word in favor of my argument. In other words, perhaps the concept of prejudice actually refers to a judgment made before all the relevant facts are in, whether or not these facts are available to the person making the judgment. Assessing someone on their individual merits therefore means assessing them on all their individual merits, known or not.
Now, if this is the case, then it must mean that a person making a decision without all the relevant facts, even the unavailable ones, is prejudging the issue — he is "prejudiced" because he fails to take into account information that he cannot possibly take into account. But, surely, to require such — and to disparage its absence — is to hold people to an impossible standard — to require omniscience, and complain of the limited knowledge of mere mortals!
In fact, this is exactly what those with this view of prejudice require, as when Charles Lamb (in the above quotation) disparages his own likings and disliking as mere prejudices, on the basis that he is "earth-bound and fettered to the scene of [his] activities." But what more can one ask of poor Charles? That he transcend his own experience, knowledge, and perceptual capabilities — forming likings and disliking on the basis of things he does not have an opportunity to know or observe? If so, then this is pure mysticism and cannot form the basis of a legitimate critique of judgment and the decision-making processes.
The False Identification of Prejudice
If we reject this mystical and impossible standard, and instead accept that prejudice only properly refers to judgments formed without consideration of the available information, then it is easy to see that much of what is liberally described as prejudice in these heady days of "progressive" thought is actually the exact opposite. What is described as prejudice is often nothing more than belief that is formed on the basis of the sum total of a person's actual experience and learning — that is, formed on the basis of observation of reality.
If one asserts, for example, that women are hopeless at parallel parking (intending this as a general statement, rather than a strict one-to-one relationship), then this is likely to be a judgment based on observation. Similarly, if one asserts that Irishmen are a bunch of drunks (again, intending this as a general statement) then this is likely to be a judgment based on observation. These are both empirical propositions, and whether each is true or false is a matter to be resolved by an appeal to observation; perhaps to anecdotal experience, and comparison of one's own experiences with others, but ideally to some more reliable method, such as meticulous empirical study of the proposition. (Of course, information of this kind is often not available.) In either case, if the judgment comes when it is needed, is based on observation, and uses all the information that is available at the time, then the charge of prejudice is misplaced.
Though the concept of prejudice is a statement of prejudgment, it is quite often the case that allegations of prejudice involve no attempt to assess the process by which the relevant judgment is made, and the evidence that is used. Indeed, allegations of prejudice are often mere bald assertions. There is no allegation that a person has failed to take account of some piece of information available to them, and the accuser merely presumes that the judgment made could not possibly be a judgment formed on the basis of evidence, since they themselves disagree with it. It is merely presumed a priori that certain empirical propositions cannot possibly be true, since they would offend the sensibilities of the person alleging prejudice. Indeed, it is quite common to see beliefs described as prejudices even though they are actually true, and reflect some demonstrably empirical relationship in reality. Here, ironically, it is not the disparaged belief that proceeds in willful ignorance of the available facts; it is the allegation of prejudice. It is the judgment of prejudice that is prejudiced!
Do you want to be able to identify actual prejudice without being prejudiced yourself? Then consider asking the following questions: How was this belief formed? Is this actually a prejudice, or is it a proper empirical judgment? What do the observable facts say about the issue? What facts seem to have been taken into consideration in forming the belief? What, if any, facts were ignored? Did the person with this belief have access to facts that they willfully ignored, or were some facts simply unavailable when making the judgment?
These are surely legitimate issues if we wish to distinguish irrational prejudgment of an issue from legitimate judgment with all the available facts. To say that prejudice develops from experience, as in the above quotations, is to admit that the beliefs being referred to actually are based on evidence from reality (even if they are ultimately false). To disparage experience in this way is virtually an admission that the kinds of beliefs being disparaged are not prejudices at all, and are being disparaged precisely because of this fact. Such judgments are disparaged, not because they are made with a lack of information about the facts, but because they are made with factual information.
Examples of this kind of muddled thinking about prejudice abound, but it is most apparent in people's attitudes toward judging stereotypes. If one is ever caught musing that Native Americans like to gamble, that the Irish are frequent binge drinkers, or that an awful lot of black guys seem to like basketball, there is a good chance that this will lead to an allegation of prejudice. Yet all of these allegations have some support in empirical observation, and it is quite possible for a person to form such views on the basis of empirical observation and inductive reasoning. Indeed, it is quite arguable that these are true empirical statements, so long as they are framed as statements of correlation, and not causal statements of racial determinism, or strict one-to-one relationships. However, in the face of accusations of prejudice it is usually futile to cite empirical evidence on these points, because the charge is not really a complaint that you are prejudging the issue — but a complaint that you are not!
In fact, most (if not all) stereotypes are formed and dispelled on the basis of people's actual experience, aggregated to account for the experiences of large numbers of people. Stereotypes arise when large numbers of people notice some correlation between human characteristics and behaviors and mention this to one another in conversation. The observation eventually enters the lexicon of common experience and the stereotype becomes widely known. When characteristics and behaviors change over time, leading some previous correlation to cease or reverse, the stereotype becomes outdated. As with its initial formation, the dissolution of the stereotype occurs when large numbers of people notice that the alleged correlation is no longer present, and mention this to one another in conversation. Eventually this new experience means that the credibility of the stereotype wanes, and it is discarded. Hence, stereotypes change over time, distilling the collective experience of people in a particular culture and time period. They are generally true, and even when they are false, it is usually the case that they were previously true at some point in the fairly recent past, prior to their disposal. Moreover, regardless of whether a given stereotype is true or false, it is rarely a prejudgment ignoring the available facts. Most often it is formed and held on the basis of personal and second-hand empirical observation.
The crucial point here is that whether or not these kinds of assertions are prejudices is determined by whether or not they are based on proper empirical observation. The common hostility to stereotypes (even when they are true) stems from the hostility to actual human experience and inductive inference. To describe such judgments as prejudice is to completely invert the meaning of the concept.
To people who take this muddled view of prejudice, the position of the infant child is the epistemological ideal, whereas the position of the mature adult who can draw upon a wealth of experience is an epistemological quagmire from which there is no escape. Many people are quite explicit about this fact, and they hail children as allegedly clear thinkers, unclouded by the prejudices of adults. Under this view, the infant child is new and free from prejudice, whereas the old man is embittered by a slew of irrational prejudices formed from his life experiences. Hence, when playwright Lillian Hellman says that "nobody outside of a baby carriage or a judge's chamber believes in an unprejudiced point of view" she is implicitly idealizing the baby, holding it in the same esteem as the judicial process of the court of law. This alone should be enough to cause serious questions about this attitude to prejudice. The ideal held aloft as the standard of rationality is the infant child — i.e., the least experienced, least wise, least knowledgeable person possible! To the prejudice proclaimers of the world, this wailing, babbling little moron is the ideal standard of rational thinking, not despite his lack of knowledge and experience, but because of it.
Actual prejudice is a species of irrationality in which judgment is formed without consideration of all the relevant and available facts. "Availability" here may refer to what is actually known and has already been observed or learned by the decision maker when forming the judgment, or it may include other pieces of information that can be acquired with some further reasonable effort. Which is the case will depend on the context of the judgment and the importance of the judgment, which will determine whether or not it is reasonable to expect additional efforts to acquire more information, and, if so, how much effort should be expected.
In assessing whether or not a judgment is premature, it is also necessary to note that one often needs to make judgments pro tempore, before the opportunity to acquire comprehensive information about a subject. In this context, rationality requires that such judgments be formed on the basis of the facts available at the time, and that these interim judgments be supplanted when more information is available and taken into account in the decision-making process. In this case, the interim judgment is contextually legitimate in its timing; though it is prior to a fuller consideration using more information, it nonetheless comes at the correct time. It is a prejudgment only in the sense that it comes prior to a later judgment, using more information. It is not a prejudiced judgment in the normative sense complained of here. It comes when it is supposed to come, and it uses the information that is available.
If we were to ignore the issue of availability of information, and require all relevant facts to be taken into account, then we would never make decisions under uncertainty, and the entire enterprise of inductive inference would be prohibited. Under such a directive, man would be incapable of functioning. His standard of rational judgment would require omniscience, which he does not have, and would regard anything short of the omniscient identification of truth as a flawed decision-making procedure. Moreover, he would look upon human judgment and decision not just as flawed but as wicked iniquitous behavior. He would paralyze his own mind and inductive faculties for fear of being "prejudiced."
A proper understanding of the nature of prejudice requires a realistic expectation as to the information that is available to a person forming a judgment, and an a priori openness to the possibility that empirical propositions may be true or false. It is not at all inconsistent to make observations about group behavior and characteristics, or use stereotypes, while at the same time eschewing genuine instances of irrational prejudgment. To do so merely requires that we evaluate, when we require a judgment of something, what information we can get about it at the time and what we can infer from this information.
The Consequences of False Ideas about Prejudice
The concept of prejudice is one of the most misused and misunderstood concepts in the world today. Indeed, when one hears an allegation of "prejudice," it is quite often used to describe exactly the opposite of what it actually means. It is most often applied as an allegation against the use of stereotypes, despite the fact that these are actually the distilled wisdom of the empirical judgments of millions of people, and are applied in situations where the stereotype represents a valid inductive inference.
In case the foregoing argument is of interest, it may also be interesting to note the common reaction of people who hear it. I have, on several occasions, been privy to a faulty accusation of prejudice, of the kind discussed, and I have occasionally taken it upon myself to politely explain that the allegation seems to refer to the exact opposite of a prejudice. One might expect that such a revelation would be surprising to the accuser, and, if they agreed with the argument behind it, this might lead them to some reconsideration of their views. However, in my experience, people using the concept of prejudice in this way seem quite unperturbed by the fact that they are using it to describe its polar opposite. If this is pointed out to them, their reaction is usually to regard this correction as horribly pedantic and trivial. They react as if someone is playing a clever semantic trick on them. "Well sure, I guess that technically that's what prejudice actually means, but still!"
This attitude stems in large part from the view that stereotypes and other kinds of judgments viewed as prejudicial are certainly bad, even if the appellation "prejudice" does not correctly describe them. Thus, when informed of their misuse of the concept in deriding someone else's judgment, the attitude of many people is that the judgment they have disparaged is definitely worthy of indictment, even if the indictment itself needs to change. Even if the relevant facts have all been considered, assertions that lots of Native Americans like to gamble or that lots of Irish guys like to get drunk offend the feelings of people who are reflexively hostile to empirical statements about groups of people. Prejudice is sometimes the preferred indictment, but the actual complaint is usually the opposite — that these are a priori wicked statements that cannot be true, and that the appeal to empirical evidence on such matters is itself wrong and sinister.
But this is not merely an issue of semantics. The faulty view of prejudice is a matter of substance, and not some benign triviality, relevant only to English-language pedants. It actually has some very bad consequences. In particular, this faulty understanding of rational judgment leads people to disparage experience and empirical judgment as some kind of impediment to reason. It causes them to reject stereotypes without any examination of their foundation and, most importantly, whether they are true!
Philosophically, this attitude stems from the so-called "rationalism" associated with philosophers like Spinoza and Leibniz — the view that all knowledge can be derived intellectually through deduction, without any appeal to experience. More specifically, it stems from the false view that groups of people must be equal in all empirical characteristics and behaviors (a fortiori if these have any normative implications) regardless of what the actual facts show us.
It is only through empirical judgment that we form beliefs as to what characteristics commonly go together, and what do not, such that we can make rational predictions about what to expect from people we know little about. For example, the Chris Rock joke I mentioned at the start of this essay is rife with stereotypes and the so-called "prejudices" that we have just discussed, but it is actually evidence of proper empirical observation and inductive reasoning. The humor in the joke derives from the contrary nature of the facts arrayed and their collective improbability compared to historical patterns — it is funny because most of the best rappers are black, most professional golf players are white, most of the tallest basketball players are not Chinese, and certain German (and earlier Prussian) governments' have had a less-than-stellar record for peaceful foreign affairs. It is precisely because stereotypes are true in large numbers of individual cases that the situation highlighted in the joke (though cherry-picked) is both surprising and humorous.
Of course, if one were to drastically underestimate the probability that, despite the common trends and useful stereotypes, there could be a highly popular white rapper, a champion golfer who is black, or a very tall Asian basketball player, then this would indeed be reason to reexamine one's inductive reasoning. Similarly, if one were to refuse to recognize the present situation as evidence that goes against the stereotype (and therefore weakens its predictive ability) then this would also be a reason to reexamine one's inductive reasoning. But to even understand the joke requires one to know that there are group trends in behavior and characteristics that are predictive of individual cases. This is what lets us know that the current circumstance referred to in the joke is uncommon and therefore funny.
Don't be prejudiced about prejudice! Make sure you look at the facts, and consider the basis on which judgments are formed. Instead of following the outlook of those who disparage experience, we are much better to listen to Voltaire, who understood that "prejudice is opinion without judgment."
 The modern word "prejudice" is derived from the Latin "praejudicium" which is the conjunction of "prae" (before) and "judicium" (judgment) (see Harper, D. (2001) Online Etymology Dictionary. The etymology of the term makes clear that it has always referred to the making of a judgment before the consideration of some fact.
 In the foregoing examples, I have included an element of coarseness in the empirical assertions that could rightly be a basis for legitimate criticism, depending on the context. I do this because it is often the case that group judgments like this are expressed in a vague or unrefined way, without proper stipulation of the exact meaning of the assertion or any relevant caveats. Nevertheless, it is usually clear from the context that such statements, even when presented coarsely, are empirical assertions of correlations in group behavior or characteristics. Thus, if someone says that "Irishmen are a bunch of drunks" they actually do not intend to assert a literal one-to-one relationship; they are saying that the incidence of drunkenness is substantially higher among the Irish than among other groups. Similarly, if someone says that "women can't drive" they would not intend this literally (since clearly women can drive); rather, they are saying that the average driving proficiency is substantially lower amongst women than men. Hence, I take such statements to be coarse statements of correlation.
 Empirical support for the proposition that (on average) native Americans are more likely to gamble than others can be found in several sources. Anecdotal evidence can be found from the presence of many casinos run by native Americans (though this also has other historical and political causes). Empirical studies of gambling also lend some support to this proposition (see e.g., Welte, J.W., Barnes, G.M., Tidwell, M.O. and Hoffman, J.H. (2008) The presence of problem gambling among US adolescents and young adults: results from a national survey. Journal of Gambling Studies 24, pp. 119-133.)
 Recent research shows higher rates of alcohol abstinence, but also higher rates of binge drinking in Ireland compared to other European countries (see Ramstedt, M. and Hope, A. (2004) The Irish drinking culture. http://www.drugsandalcohol.ie/5841/1/2396-2528.pdf). Empirical support for high levels of alcohol induced problems amongst Irish people can also be found in earlier sources, though there is some conflicting evidence regarding different measures (see e.g., Lynn, R. and Hampson, S. (1970) Alcoholism and alcohol consumption in Ireland. Journal of the Irish Medical Association 63(3), pp. 39-42; Walsh, D. (1970) Alcoholism in the Republic of Ireland. British Journal of Psychiatry 115, pp. 1021-1025; Walsh, B.M. and Walsh, D. (1973) Validity of indices of alcoholism: a comment from Irish experience. British Journal of Preventative and Social Medicine 27, pp. 18-26).
 The most compelling empirical evidence for this assertion is the racial composition of basketball teams and leagues in countries such as the United States, which have a very high proportion of black men compared to their proportion in the general population. (Empirical support for this assertion can also be found in a novel piece of research on the likes and dislikes of different self-identified racial groups on an internet dating site (see OKTrends (2010) "The REAL 'Stuff White People Like'").
 For a more precise account, we can turn to statistical inference and decision theory. This tells us that a decision-maker should take account of all presently known information, and should seek out further information to improve the decision in cases where the expected value of the additional information exceeds to costs (e.g. inconvenience) of acquiring the additional information. The expected value of the additional information is the difference between the expected benefit of the decision without the additional information, and the expected benefit of the decision with the additional information. Hence, the value of additional information will depend on the usefulness of this information in improving the decision in question, and the importance of the decision in conferring a benefit (or avoiding a loss) to the decision-maker, or others.
 Emphasis added.