When I started college, I was assigned a roommate I didn't know, a guy I later came to refer to as the Jewish Nazi. He argued, among other things:

"If I am walking on the street at night and I see a black man coming, I move to the other side of the street."


His justification was essentially that the crime rate for Black males in the US is significantly higher than for the rest of the population, so it was rational to be afraid of them in such a situation and take appropriate action. Now, I found this stance pretty repugnant, but at the same time there seemed to be some logic to the argument. I had generally held that prejudice was just plain irrational. His argument seemed relatively rational, though I still felt that it was ethically wrong. My goal here, inspired by that interchange, is to investigate the nature of prejudice, whether it is rational, and whether it is ethical.

I want to start out with a relatively broad definition of prejudice that does not specify whether it is rational, so that we can explore the issue. Let's make the following definition of prejudice:

A judgment formed about a matter without all the relevant information, a preconception.


This follows largely the second definition given by Webster, with an attempt to stay neutral on the subject at the outset. Webster also suggests we may include, “leaning toward one side of a question from other considerations than those belonging to it,” but that is clearly irrational and will as a result be unethical if the decision is at all important, so we'll ignore that case. Now that we have agreed on what prejudice is (for these purposes), we can explore whether it's rational or ethical.

Is prejudice necessarily irrational?

If we make a judgment without all the necessary information, is that always irrational? If we reach a conclusion while ignoring some information that we know may prove it false, then we have essentially reached a conclusion which does not really follow from the premises. That argument is, by definition, invalid, so it's probably pretty fair to call that irrational. But in real life (as opposed to formal logic) we seldom, or possibly never, have all the facts that might be relevant. In practice, we generally collect enough information that we feel pretty sure, and we think we know there's a pretty small chance (but always non-zero) that there are important things we've ignored. What this means is that the difference between ordinary judgments and prejudice becomes much more gray, because the question is "how much evidence is enough"? The answer comes from the fact that our conclusions, to be logically valid, are never 100% certain; we can only say we're "pretty sure", and how sure we are depends on our evidence.

So, we might divide judgments into three kinds. The first are judgments that are fairly certain, based on quite a bit of evidence. We'll call those sure judgments. They're never 100% positive, but they should be getting there. An example would be that I live in the USA. It's conceivable that people could have pulled the wool over my eyes all these years, but it's not likely. The second are judgments that are made with not all that much evidence that are stated to be not all that reliable. We'll call those rational prejudices. An example of a rational prejudice would be that you don't see any clouds in the sky in the morning so you conclude it will not rain. Now we've probably all been burned before by such a judgment, but then we know it's just a guess from the beginning. The third class of judgments would be irrational prejudices, which are judgments where we don't have all the evidence, yet we assert that we're sure about them. An irrational prejudice would be if you went to New York City, didn't like it , and then refused to go to any part of New York State, because you were sure you'd hate it. Since NYC is in the state, that is a valid piece of data, but it's not nearly enough to base the decision on. So, we can conclude that forming a judgment without all the evidence isn't necessarily irrational, as long as we remember that that judgment is quite dubious.

Prejudice in real life

So we've decided prejudice as we've defined it isn't always irrational, at least not by definition. But it's important to then ask, “Ok, but how is it applied in every day life?” I think there are probably many instances of what we've called rational prejudice in our daily life. Especially if a decision is not that important, it's often easier to make a judgment based on whatever information we have at hand rather than to get all the information necessary to make a sure judgment. However, in most of the cases that people usually apply the term prejudice, I claim that it is irrational prejudice.

Let us return to the Jewish Nazi's example. He says that he will avoid a black man on the streets at night because a black man is statistically more likely to commit a violent crime against him than the rest of the population*. He claims this is rational. Now, if one were put in this situation and had absolutely no other information to go on other than a person's race, this might be a rational choice; however, in such situations we actually have a wealth of information at hand. This information includes most importantly body language and appearance. As with other animals, the body language of a human often reveals his intents and is by far the most relevant piece of information. People actually use this sort of judgment all the time. Also, for better or for worse, a person's clothes can tell you a lot about their social standing and attitudes. The point is that compared to those observable reflections of behavior and intent, race is an extremely poor indicator of someone's intentions (speaking statistically), and it is, indeed, irrational to pick out that one characteristic as important to making a judgment about the person. Now, if you see a person walking down the street with a shotgun and she doesn't appear to be a police officer, then it is completely reasonable to assume she is a danger to you, because it is not normal to walk around brandishing a firearm and would be a good indicator that someone is about to commit a crime.

You may object that all the ways of judging a person's intent that we've mentioned are extremely superficial and probably not all that reliable, and that is true. However, they are appropriate to the “man you meet on the street” situation, because it's not as though you can have an extended conversation with each person you pass on the street to understand his world view and intent, especially if you think that person may pose a danger to you. Generally, though, when people argue about prejudice, the argument usually isn't centered on whether irrational prejudice is bad (which most people would probably agree on) but whether a certain judgment is rational or irrational prejudice (though it wouldn't be put in those terms). That is an argument of fact that has to be decided for specific cases, but it should be noted that, in most cases that will come up, the things that are usually labeled prejudice are irrational prejudice. This “man you meet on the street” scenario is artificial, because it is one of the situations where it is most justified to make judgments with very little information. In most other situations, like school or the workplace, the BEST way to make judgments about a person is simply to get to know her. Talking to him and observing his actions is by far the best way to make judgments about him. In those cases using any superficial characteristics like race or appearance, and usually even things like religion, is completely irrational. In those cases, we may simply say the proof is in the pudding; actions speak louder than nationality, race, creed, etc. Racism is probably the form of prejudice that is most clearly irrational in most everyday situations, but other prejudices such as those based on sex, clothing, speech, nationality, or religion are often irrational as well. Generally, if we stop to think about it, they are used to judge about things that one could just ask about or observe in interactions with others.

Why do people hate and fear prejudice?

Here I will speak from my vantage as a citizen of the USA at the beginning of the 21st century on how people feel about prejudice and why. Prejudice is generally considered a pretty nasty thing, and saying someone is prejudiced is generally considered to be a fairly harsh insult. History instructs us on some of the reasons that people fear prejudice. Slavery (e.g. of blacks in the USA) and genocide (e.g. Jews in Nazi Germany or native americans in the USA) are generally considered to be amongst the darkest chapters in human history. What makes irrational prejudice frightening is that someone who is acting irrationally may well take actions a rational person would not (as in the case of lynching, witch burnings, or the Nazi gas chambers), so that they are indeed quite dangerous, and if I am discriminated against because of irrational prejudice then there is a good chance that rational appeals will not influence those who are discriminating against me. This, I think, is why forms of prejudice like racism are the most frightening, because in that case an individual is being judged on the basis of something he cannot change. If people decide it is wrong to wear black clothes, I my disagree and I may defy them, but clearly I have the option of avoiding trouble by conforming. If someone decides that it is wrong to have black skin, a black person has no recourse. So, even from an abstract analysis, irrational prejudice is clearly a very dangerous thing for a society, and history certainly confirms that conclusion.

Is prejudice always a bad thing?

Now that we have outlined that prejudice may sometimes be rational and sometimes irrational and that there are very good reasons to fear irrational prejudice, we may ask whether it is ever right to act on prejudice. Certainly, the answer will depend on each person's code of ethics. Some people may believe in a code of ethics that says prejudice is inherently wrong, no matter the circumstances. If we just talk about wanting to achieve our own goals, whatever they may be, then we can already say that irrational prejudice is wrong, simply because it is irrational. If we make irrational judgments, we will not be able to accurately determine how to achieve our goals, which is bad. That's pretty obvious, but then we could ask whether it is right to act on rational prejudice. For most people this will depend on what the judgment is about and the possible effects. If you are making judgments about the weather or what route to take to work, then I think almost all of us would find rational prejudice (what you might think of as a “snap judgment”) perfectly acceptable. The questions arise when it comes to judgments regarding people.

Let us return now to the Jewish Nazi's “man you meet on the street” scenario from the beginning of this write up. We have already said that in actual fact, it is generally irrational to judge whether a person is a danger to you based on his race, because it doesn't tell you much about him compared to the multiplicity of other indicators about a person that are evident even from just superficial observation. However, for the sake of argument let us talk about the hypothetical situation that we are in this situation and cannot tell anything about the person other than their race and proximity. This is clearly unrealistic, but we want to ask about even the most extreme situation to completely understand prejudice. In this contrived situation one might argue that it's rational to cross to the other side of the street, because it's just a choice between more or less chance of being assaulted or mugged. That should be a simple choice.

We might be forced to agree, if we don't consider the effects of our actions on others. But consider now what it says to a person if others behave that way toward them, especially if there is a larger societal pattern of prejudice toward that person that we are playing into. I think most of us would feel it will weigh negatively on that person. We must view the action in the context of the larger, systematic prejudice against that individual's race. In that case, I think it can be argued that on the whole acting on prejudice (even if rational) has bad effects for society and, in the end, for each individual. Prejudice deprives people of opportunity and in some cases can even cause people to in the end fulfill the negative expectations placed upon them. So, given the usually small probability of actually being assaulted and the marginal increase in safety given by crossing the street, I assert that it does more harm to the other individual, by playing into a larger scheme of discrimination, than it does good for us by insulating us from danger. We could also object that maybe it's more reasonable to argue that we avoid danger best by avoiding any people we meet on the street. The analysis is much more clear if we change the scenario to say that there are two equally nondescript men, one black and one white, walking on either side of the street, and we must choose which one to walk past.

In the preceding, I am saying nothing more, really, than that I would certainly not want to be the recipient of that prejudice, and the harm done by it outweighs any possible good. This judgement may vary for different peoples' ethical systems and based upon the situation, but I think that in many cases even rational prejudice is not ethically justified when directed against other people. We should close by recalling that this is a highly hypothetical example and that in practically any realistic example we've found that such prejudice is irrational and bad under almost any ethical system.

What do people really mean when they say “prejudice”?

The purpose here was to examine the idea of prejudice, not the word itself. In my experience, most people use “prejudice” to refer exclusively to what we've termed “irrational prejudice”, and specifically when the judgment being made is negative. I simply invented the other terms to facilitate this discussion.

Who is prejudiced?

I think the short answer is everyone. All of us harbor irrational prejudices of one sort or another, even very good people do. Having a prejudice isn't a good thing, but it doesn't make you a bad person. It's just that you must continually challenge your ideas and seek uncover your prejudices and deal with them. I think we all have prejudice because it is a natural extension of human inference. We are constantly unconsciously trying to make connections between things, and sometimes those connections turn out to be false. When those false conclusions involve people, they can be particularly problematic in the form of prejudice.


I am very eager to get comments on this WU, since it's unlike others I have done. Also, I've done my best, but help with typos and misspellings is always appreciated.

* Note: For the purposes of this writeup, we assume that the crime rate for black males is significantly higher than for the general population. I think this is, in fact, true in the USA, but if you disagree then assume it for the sake of argument. It should also be noted that correlation does not imply causation, so one should be very careful in drawing conclusions from that statistic.

My goal here is to defuse prejudice by approaching it in an open and rational fashion that, hopefully, will speak both to those who agree and those who disagree. It is generally a subject which is semi-taboo, in that it's unacceptable for anyone to suggest prejudice might be ok. I feel that we must first suggest that idea in order to refute it, and by making an issue taboo we keep people from truly understanding why prejudice is bad.