Teaching criminology courses, and especially police courses, the topic of "racial profiling" comes up a lot. It often surprises my students to hear me, Johnny Long Hair Hippie Professor, say that racial profiling basically doesn't exist. Sure, there are unfortunately plenty of examples of law enforcement who clearly set out to harass people of color. But fortunately, those folks are very much in the minority (pun most definitely not intended).
The real problem is that police are human beings. And like all human beings, they are subject to the social forces that all people are. So in trying to understand why there are huge racial disparities in our criminal justice system (despite there not being huge disparities in who is committing crimes) does not come from a bunch of racist police, prosecutors, judges, etc. setting about to harass and imprison people of color. Rather, it comes from the fact that we live in a racist society, and becoming a criminal justice official doesn't magically make on immune to social forces.
Take this recent study that gave law firm partners identical copies of a memo that contained 22 objective errors in spelling, grammar, and fact. The only difference in the study was that some of the partners were told the memo was written by a white person and some were told the memo was written by a Black person.
Well, as anyone even remotely familiar with American society could have guessed, those who thought they were reviewing the writing of a Black person were far more likely to find more mistakes than those who thought they were reviewing the writing of a white person. On a subjective scale of how well the partners thought the memo was written, the "Black" memo received an average rating (on a 5-point scale) of 3.2 while the "white" memo averaged 4.1 But more telling, amongst the objective errors (misspelled words and claims that were factually wrong), the reviewers found on average 2.9 mistakes in the "white" memos and 5.8 mistakes in the "Black" memos.
Again, these memos were the exact same thing. The only variable changed was whether the reviewer thought the memo was written by a white or Black person. Then even in things that are objectively either right or wrong, like spelling, the reviewers were much more likely to notice the mistakes in the "Black" memo. This is in line with many previous audit-style studies that have found things such as with identical resumes submitted for a job application, those with white-sounding names are significantly more likely to receive a call back/job offer than those with black-sounding names.
The point of such studies is to demonstrate that even in completely objective areas, the (perceived) race of the person being evaluated still makes a significant difference. These studies do a good job destroying the "well, if Black people didn't want [x bad thing happening], they shouldn't do [y thing]." But as the writing sample studies shows, Black people would have to be roughly twice as good with spelling as their white counterparts just to be viewed as equal.
So what does this have to do with racial profiling? Because we're all criminals. This is a point I make to my students constantly -- there are so many laws on the books at just the federal level alone that no one actually knows how many laws exist. The most conservative estimates put it at roughly 10,000 laws at the federal level alone. If you actually dig into all the laws on the books, you'll find that you're breaking multiple laws per day without probably even knowing it. For example, in nearly every state it is illegal to have anything hanging from the rear-view mirror of your car, or to have any stickers on any window of your car that were not issues by the state. And yet I know dozens of people who do exactly that.
This becomes a problem for equal enforcement -- if basically everyone is breaking the law at all times, how do you chose who to stop and who to let go? After all, police can neither practically nor politically detain every citizen every time they break the law, or we would all be in jail right now.
So instead, like all human beings, police use context clues to help them distinguish between who to investigate and who to ignore. When you combine that with a society that has an incredibly strong connection between dark skin and the suspicion of criminality, it's not surprising that so many officers would, again not intentionally, end up making people of color their main focus. And since we're all pretty much constantly breaking the law, they will more often than not find people of color doing something illegal, thus justifying their suspicions. Of course, were they to pay the same amount of attention to white citizens they would find the same amount of crime, but that counter-factual rarely enters into most people's thinking.
I really like studies like the writing study up top, because they demonstrate so forcefully how even in rather inconsequential areas of life race is an incredibly strong factor in how people are perceived. Of course in writing samples it may not be the biggest deal in the world, but when such disparities in perception move into the realm of criminal justice, hopefully I don't have to explain why they become much more significant. But more than anything else, they completely destroy the regressive, simplistic, and ultimately racist argument of "Well, if Black people don't want to be arrested so much, they should quite committing so much crime." Because as these studies (and dozens others like them) prove, the real formula there would be "Well, if Black people don't want to be arrested so much, they should quit being Black in public."