Systemic Racism: a Rationalist Take (Part 7 of 9)

Do blacks have more police encounters than whites?

In 2015, the percentages of whites and blacks who experiences police initiated encounters with the police were equal, a little over 10% each. (Bureau of Justice Statistics.) These rates seem to debunk a popular alternative narrative appearing in the Boston Globe, among other liberal sources, that claims that blacks get killed more often by police than they should (see below) because they have more police encounters, including some prompted by racial profiling. There is also evidence from the Stanford Open Policing Project that police in various places stop black drivers at a higher rate than they do white or Hispanic drivers. The Prison Policy Initiative (a liberal organization) also asserts that African Americans share on average more contacts with police than whites and are stopped proportionately more often than whites, both on foot and in automobile traffic.

This disagreement is important for the following reason: Suppose that the probability of being killed by police depends completely and impartially on the probability of being stopped by police, say, in traffic. Think of a sort of deadly lottery: If you are stopped by police, there is a constant and equal probability of ending up dead. In this case, if police are more likely to stop black citizens than white citizens then, they are automatically more likely to kill black citizens, even for frivolous reasons or, in some other way, unfairly. Imagine further that black citizens are mostly stopped in traffic for having a missing taillight on their car and such, while white citizens are only stopped for such egregious conduct as going through a stoplight at 65 miles an hour. In that hypothetical situation, the killing of black drivers could easily be a result of police animus against them. Remember that this would be true although police would be just as likely to kill the whites as the blacks they stopped.

In the same situation, the killings of whites citizens by police would be more likely to be justified. In the hypothetical situation I describe, some black victims of police would be indirect victims of racism though black and whites would be stopped the ones as often as the others. In this scenario, the possibility of police hostility against African Americans could even remain in a situation where more whites than blacks are killed by police after a traffic stop. An extreme formulation of this perspective would go like this: Police kill whites they encounter as frequently as they kill blacks that they encounter, but all the blacks they kill are completely innocent while all the whites they kill are all guilty of some serious or violent legal trespass.

So, I ask, are there reasons other than racial animus, legitimate reasons, why police would stop black citizens more frequently than they do white citizens? This is a hard thing to figure out but it’s worth trying because the answer contains a potential explanation beyond the simple findings that police kill the whites they stop as much as the blacks they stop.

Is there any reason other than racial prejudice or animus why police would stop African Americans more than they do whites?

Ideally, a detailed study of police stops at every level of seriousness of suspected offense would answer this question. I think such a study does not exist. (I hope it will, soon.) So, I will use a trustworthy proxy for all forms of lawbreaking: murder convictions.

The assumption I make here is that there exists a continuity between such offenses as homicide, armed robbery, DUI, running stoplights, and driving with a broken taillight. Underlying this continuity would be a propensity to break the law. If no such continuity, no such propensity exist, my conclusions at the end of this section are correspondingly in doubt. I already know that my choice of indicator is imperfect in one respect: There is probably no continuity between crimes of passion and other transgressions. However, those are a small number of the total. This scheme also leaves aside whole categories of serious crimes that are almost certainly preponderantly white crimes, such as financial transgressions – because they seldom give rise to impromptu encounters with police.

I choose homicide as a substitute for all lacking measures of lesser categories of law breaking for several reasons. First, homicide is almost always an unambiguous act – as opposed to jumping stop signs, for example. Second homicides are more likely to be scrutinized than other forms of law breaking. Third, the race of homicide perpetrators is more likely to be known than the race of other crime perpetrators. Fourth, homicide is not as likely to be charged frivolously, without reason, as lesser offenses such as jumping stop signs may be.

Tech note: Below, I am dealing in broad orders of magnitude rather than is specific quantities. I mean that if I cut any figure I proffer by one third, the associated reasoning would remain intact.

African Americans regularly account at least for 40% and up of all homicides (“Race, Ethnicity and Sex ….” 2016 Crime in the United States, Table 3, already cited above.). The real situation is worse than this. It turns our that females in general commit ten times fewer homicides than males. So, it would be closer to the truth to state that something like the 6 to 8% of Americans, who are black and male, commit around 70% or more of the homicides in the United States. It matters to my reasoning that this is a long-standing situation. In 1976, a black male was 12 times more likely than a white male to commit homicide; in 2005 however, he was only 9 times more likely ( James Alan Fox, Northeastern University and Marianne W. Zawitz, BJS Statistician; BJS: Bureau of Justice Statistics Homicide Trends in the U.S. 2010.)

Why then would police stop African Americans preferentially? The seemingly ingrained current, as well as traditional answer is “racism.” One major objection to this view is that, as we have seen, the same supposedly racist police then and against expectations, are no more likely to kill the blacks they stop than the whites when given the opportunity.

The inference I make above may explain this paradox. Police, being well aware of a higher black propensity to break the law, stop African Americans more frequently that they stop whites. They would do so, perhaps, because stopping black offers a better yield than stopping whites. For the same reason (a speculation), they treat them more brutally, irrespective of degree of compliance (Fryer, Roland G. Jr: “What the Data Say About Police.” WSJ, 6/23/20, already cited). If racism guided their actions rather than a harsh but but basically rational realism, they would also kill the African Americans they stop more often than whites. The fact is that they do not.

Are black police officers as likely to kill black suspects as are white officers? If white officers preferentially killed blacks and black officers did not, or any attenuated version of this divergence would contribute to establish the thesis of systemic racism. So, what are the known facts on this? A 2018 Rutgers University study by Charles E. Menifield, Geiguen Shin, and Logan Strother, based on 2014- 2015 nation-wide data about all police killings answers a with a clear “Yes.”

If black officers kill black suspects as readily as do white officers, it’s unlikely that white officers’ killings of African Americans are generally expressions of any racism while the equal propensity of black officers to do the same is not. It’s more reasonable to suppose that the equal probability of black and white officers to kill blacks is the expression of, or is associated with something other than racism.

While I wish that we had a bigger study covering more years than the study cited above, it is the only systematic treatment of the data we have. For the moment, I would rather go with it rather than with a dozen well documented, dramatic, lamentable episodes spread over five or six years. A principled new study covering more years showing that white officers are more likely than black to kill African Americans could undermine my provisional conclusion below any time. Incidentally, and here again, I am surprised that the many supporters of the thesis of systemic racism in academia have not yet produced such as study. Or, have they?

I am well aware of the adage that “absence of evidence is not evidence of absence” but it’s only partially correct. If you said that a presumptive wolf got into your house every other night and stole one of your children, and if you did not proceed to study wolves, I would quickly suspect that something was amiss. I would believe in a short time that your narrative is all askew. Or, I would even judge that your lack of due diligence conceals something important

Disposing of a slightly risky inference If the inference I make above about a kind of continuity of lawlessness from homicide to other, milder forms of law breaking is not credible, it can be replaced with the following more basic formulation: Police officers are likely to be aware of the high black homicide rate. As a consequence, they consider black people in general more dangerous than white people. This knowledge is the basis for their stopping black citizens more frequently than white and also for their rougher treatment of black citizens than of white.

In this scenario, police officers are acting in a rational way although it may be objectionable. Again, African American police officers kill blacks as often as white police officers do.

Perhaps these practices amount to systemic racism. I think the case has to be made explicitly and clearly. It should affirm that police officers should not act differently with those they consider dangerous than with those they do not. The explanation should also include an evocation of how police should act differently based on the information available to them.


Here is a detour, an obviously necessary detour. The analysis above does not seem to me to support the concept of systemic racism but it leaves plenty of room for charges of racial unfairness. The legions of African Americans who think of rolling through a stop sign as significant lawbreaking would be, according to the same analysis, possibly, just possibly, at greater risk of being killed by police than their white fellow-citizens just because of some African physical features. The unfairness resides mostly in the fact that such features could not be erased or masked would they wish to do so. (I don’t suggest they would or should.) Yet, differential treatments based on such or similar ascribed characteristics are common in other phases of real life and normally seen only as regrettable but unavoidable facts.

There is a large category of Americans who are systematically required to pay 20% higher premiums for life insurance than the majority of the population, the category of reference. This category is “Men.” It’s unfair because, in general, as a rule, men cannot stop being men. Insurance companies routinely advance the justification that, at every age, men are more likely to die than women thus creating a higher risk of disbursement for the company. Men are also known generally, on the average, to engage in more risky behavior than women. Yet, the premium surcharge imposed on all men is obviously unfair to some male human beings like me who never touch alcohol or any other drug, don’t smoke, exercise two hours of every day, and eat only tofu and kale. It’s even unfair to the probably many less saintly men who do not lead riskier lives than do women in general. Do you see the parallel?

If the police tendency to stop blacks more than whites based on general numbers and an ascribed characteristic constitutes systemic racism, isn’t it true that the absolute preferential treatment insurance companies afford women is “systemic sexism”?

I must add this: African Americans have allowed themselves to be treated as members of a “community” for at least fifty years or more (perhaps, since the Civil Rights Movement.) This makes it difficult to advance claims based on individual traits, like this: “I don’t smoke and I don’t even roll through stop signs.”

Putting all the numbers together, this higher risk for African Americans to die at the hands of the police is compatible with the idea that they are stopped more frequently than are whites although, once stopped – as we have seen – they suffer no greater risk of being killed. Though these figures indicate that African Americans are more likely to die from police action than others, they don’t demonstrate systemic racism in law enforcement. They may be compatible with that concept through some other path I have not discovered. Perhaps, more research is needed. I have trouble believing this. I think that if it were possible, it would have been done by one of the several organizations dedicated to the welfare of black Americans, or by any one team of American liberal academics.

Confusing Yesterday and Today

One must ask why much of the general public, helped by the largely left-leaning media seems to accept a narrative starkly negated by available figures. It seems to me that the explanation resides in a massive confusion in the mind of that fraction of the general public that is intellectually honest about racism.

The confusion concerns the passive collective inheritance of slavery. Those whose ancestors came here in chains and against their will (instead of being highly self-selected like all other immigrants. See my “Why Immigrants are Superior.”), those descendants of slaves who received a systematically inferior education or none at all, those whose grandparents were limited in their occupational choices by legal segregation, such members of American society will do less well economically and socially than those whose antecedents suffered no such limitations on their talent and character. Throw in thousands of lynchings and the occasional deadly race riot and you have a societal design for the failure of some.

If you could conduct an experiment replicating those conditions with people selected at random, marked with a blue tattoo on their left hand, and made to breed among themselves you would certainly observe in them below average rewards of life on any conceivable indicator. This would happen in the absence of any current (current) mistreatment of their descendants. The now vague factor of “racism” would not have to be investigated. The historical explanations above would suffice.

[Editor’s note: Part 6 can be found here, or you can read the whole thing here.]

Please keep it civil

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s