Many people, including myself, are highly skeptical of the modern systemic racism narrative: the idea that black people, and other minority groups are systemically oppressed in the United States, via our institutions and culture, which undervalue them, and that this oppression severely limits any individual black American’s chances of success in the year 2020. There is some truth to the idea, but the extent to which racism is currently gripping society has been wildly exaggerated. For instance, black economist Roland Fryer intended to demonstrate that there was a bias against black individuals in the lethal use of force by the police, even after confounding factors, such as the suspects’ behavior during interactions with police, were considered. His study found no such bias, much to his surprise, though it did find bias in non-lethal uses of force.
The media narratives about George Floyd’s murder at the hands of the police give the impression that America has not moved past the 1960s. But a critical examination of the issue exposes the fallacious nature of the claims that our society is in the grip of systemic racism.
An abundance of evidence has been presented to support the idea that racism is systemic or is proliferating in our society, but I am unable to draw that conclusion from the proof on offer. The evidence I have seen almost always suffers from at least one of three flaws.
The Disparity Fallacy
The first flaw is the idea that disparity equals bias. (I discuss one disparity study in detail here.)
Most phenomena are complex and multifaceted. Racism is just one potential explanatory factor among many. If an experiment or observational study does not account for the other confounding variables, then it is impossible to say with certainty what percentage of the observed outcome is explained by these variables. Controlling for confounding variables can neutralize the effects of these other influences on the results of the experiment and the practice is therefore of paramount importance if the true nature of a phenomenon is to be discovered.
Compare and contrast the statements below. The first is from the Roland Fryer study mentioned above. The second is from historian Ibram X. Kendi, director of The Antiracist Research and Policy Center, in response to a study published in the New York Times, purportedly analyzing the effects of racism on black life outcomes:
Fryer: A simple count of the number of police shootings that occur does little to explore whether racial differences in the frequency of officer-involved shootings are due to police malfeasance or differences in suspect behavior … Put simply, if one assumes police simply stop whomever they want for no particular reason, there seem to be large racial differences. If one assumes they are trying to prevent violent crimes, then evidence for bias is exceedingly small.
Kendi: As an anti-racist, when I see racial disparities, I see racism. But I know for many racist Americans, when they see racial disparities they see black inferiority.
There is a radical difference in mindset here. Fryer shows an open-mindedness and willingness to adjust his perspective based on data and reason; Kendi relies on confirmation bias and will castigate anyone who proposes alternative explanations as a racist. The latter mindset has poisoned public discourse and led to the maltreatment of many good men and women. It must be fervently opposed by those who value rationality over dogma.
Some of the worst studies and observations of disparity make zero mention of confounding variables. For instance, a Washington Post article discussing racial disparities in school discipline procedures states, “black students accounted for 15 percent of the student body in the 2015–2016 school year, but 31 percent of arrests.” But this statistic does not show that there is bias in how schools or the police handle disciplinary incidents among children. The article attempts to deny the importance of one variable by stating that there are no differences between the behaviors of black and white children at school, but without providing a reference; however, several studies indicate that there are statistical behavioral differences between black and white children. This disparity statistic does not factor in socioeconomic or geographic factors, among a plethora of other likely confounding variables. If racism or bias is claimed as the predominant causal factor, its effects must be isolated from those of other potential causal factors. (For more on differentiating causation from correlation, see this Steven Pinker lecture.)
The disparity in disciplinary incidents is of particular importance, due to Obama era guidelines punishing school districts that cannot sufficiently explain any racial disparities in their school systems. If the disparity in disciplinary incidents is due to differences in the children’s behavior—and not to racial bias on the part of those administering the punishments, any attempt to fix it by adjusting punishment rates would necessarily result in underpunishment of black students, or, worse, overpunishment of white students. In other words, if this policy is predicated on bad statistics, it will result in de jure racial discrimination. The policy intended to punish bad actors and create a more equitable system would do exactly the opposite: just as the zealous over-reaction to sexual assault statistics on college campuses has led to an erosion of basic due process rights and hundreds of discrimination lawsuits against universities, thanks to the Obama era Title IX guidelines. Imagine the injustice that would ensue if we applied the same type of reasoning to the gender disparity in our prison system, under the misguided notion that group disparities must necessarily stem from bias in the system. (See this video from Jonathan Haidt for elaboration on this point.)
Most people have probably heard that black men are more likely to be pulled over than white men. Racial bias on the part of police officers is one plausible explanation for this. However, another plausible explanation is that, on average, black and white drivers behave differently behind the wheel.
In response to allegations of racial profiling, the state of New Jersey conducted a study to determine whether black motorists were unduly targeted. While only 13.7% of the New Jersey population is black, the study found that 25% of those recorded speeding were black, but only 23% of those pulled over for speeding were black. In other words, the black population was slightly under-represented among those stopped by police. As is often the case in these disparity analyses, as more confounding variables are accounted for, the unexplained gap between the groups in question begins to close. Simply assuming that any two or more groups must be equally distributed with regard to whatever phenomenon you are studying, and, by extension, assuming that any disparity between those groups must therefore be explained by prejudice or power dynamics is demonstrably false.
Many studies do adequately take into account the confounding variables that influence the results of data collection. One such study, published in Nature online, found that, on a national level, black individuals were more likely to be stopped by the police before than after dark, even while controlling for time of year, day and location, among other factors. The data, summarized in the table below, show that the percentage of stopped drivers who were black dropped from around 26% before dark, to approximately 21% after dark, indicating that racial bias does influence decisions to stop motorists. This difference is sizable and worth addressing, but, looked at in context—particularly in combination with the Roland Fryer study—it does not justify the levels of anger and hostility we see on the streets today.
Extrapolation from Single Isolated Instances
The second flaw common among those pushing the systemic racism narrative is the impugning of motive in single, isolated incidents, which are treated as microcosms of a broader societal trend. We see this with the George Floyd case, as with many other cases that have made headlines in the past. In 2018, two black men were arrested inside a Starbucks, after an employee called the police on them. The men arrived at the Starbucks early for a business meeting and, upon arrival, asked to use the restroom. They were denied access, in accordance with Starbucks policy at the time, since they had not ordered anything. The gentlemen then sat down at a table, the manager came over and asked if they wanted to order anything, and they replied that they were waiting to meet someone. Upon hearing this, the manager called the police and had them removed from the establishment.
The manager was completely out of line in escalating the situation by phoning the authorities so quickly, without first attempting to understand the situation. But the motives of the police should not be in question here. As soon as the manager asked the gentleman to leave and they refused, the police officers were acting in their rightful and lawful capacity in arresting them, especially as the officers first gave the men ample opportunity to leave of their own accord, without incident.
It is reasonable to question the motives of the employee who called the police. This employee may have been a racist, or implicitly biased, and more inclined to call the authorities over a minor infraction when black people were involved. On the other hand, maybe that particular Starbucks has an issue with loiterers and the manager had grown impatient over time. Or maybe the employee was an ardent stickler for the rules and intolerant of anyone who questioned his or her authority.
Some might claim that the infraction involved was so minor and the response so excessive that no reasonable person could have taken such an action; therefore, racism must have been the motivation. I agree with the first part of the logic: no reasonable person would have called the police in that scenario—but who said that this person was reasonable? Most people can probably think of a time when they responded unreasonably to a situation—even if not to the same extent as this individual. To claim that racism or implicit bias is the only plausible explanation here is absurd. The vocal people on the social justice left should give such individuals as the manager the benefit of the doubt when the facts are uncertain. However, this would undermine their aims, which is why so few activists engage with the points made here and repeated many times elsewhere.
If we use instances like this as a barometer of racism, no amount of real social progress will ever satisfy us. The odds that any one person will call the police on another for indefensible reasons are incredibly small. But run the probability analysis in a country of 350 million people, experiencing billions of social interactions every day, and the likelihood goes from incredibly small to almost guaranteed. This is simply the law of large numbers. You can find plenty of such interactions occurring between two white people, yet the news of such incidents barely makes it past the block on which they transpire. Swap out one of those white people with a black person, and the incident becomes worthy of international attention.
If we could somehow erase every trace of racism and bias from the hearts and minds of every single individual in this country, incidents like this would still occur.
Similar arguments are made each time a white police officer kills a black individual, like George Floyd. Sometimes, the use of lethal force is justified, in which case there is no reason to believe that racism played a contributory role, and, at other times, the use of lethal force is unjustified—as it was in the George Floyd incident—in which case racism may or may not have played a role. To illustrate this point, let’s look at two prominent cases from the past: the police shootings of Michael Brown and Tamir Rice.
Michael Brown was shot and killed by officer Darren Wilson, after Wilson stopped Brown, following a convenience store robbery. During the confrontation, Brown fought with Wilson for control of Wilson’s gun, causing an accidental discharge. This fact was confirmed by multiple eye witness accounts and forensic evidence. Later in the confrontation, Brown charged at Wilson, and Wilson fatally shot Brown in self-defense.
If there were ever a justified use of lethal force against an unarmed man, this was clearly it. Despite overwhelming evidence showing that Brown intended to inflict serious physical injury on Wilson, the explosive charge of systemic racism was made, resulting in the perpetuation of the hands up, don’t shoot version of events (a lie). Given that Brown actually fired Wilson’s gun during the confrontation, there is no reason to believe that Wilson’s actions were motivated by racism and every reason to believe they were motivated by legitimate self-defense.
The shooting of Tamir Rice was very different. Rice was an innocent twelve-year-old boy, who was outside playing with a toy gun, when a bystander called 911. The bystander told the dispatcher that a potential juvenile was carrying a gun—although he believed it to be a toy gun—and pointing it at passersby. The dispatcher failed to communicate the apparent age of the individual or the nature of the gun to the officer. When the officer arrived, he saw what he claims to have thought was a fully-grown male, brandishing a real gun. The officer immediately exited his vehicle and shot Rice.
There is no denying that this is a tragic story, but there is more than reasonable doubt as to whether the shooting was motivated by racism. Even the New York Times described the shooting as “reasonable,” given its context. There are too many videos of otherwise ordinary traffic stops turning into lethal encounters in the blink of an eye: the public is not qualified to play Monday morning quarterback when it is not our lives that are on the line.
On the other hand, Rice was small in stature, indicating that he was a juvenile. The gun he was holding had an orange tip, indicating that it was not a real weapon. There is a reasonable argument to be made that the officer should have given Rice the opportunity to comply with his commands—even though I do not agree with that argument. But, in instances like the death of Tamir Rice, we should focus on the facts and use reason to determine whether improper force was used. We should not impugn guilt, based on the officer’s race or occupation.
The contrasting cases of Tamir Rice and Michael Brown illustrate that the justification for using lethal force is a variable independent of motive. I believe that the use of force in the Brown case was wholly justified because I do not think there is another reasonable interpretation of the facts. For the sake of argument, let’s assume the opposite is true of the Tamir Rice incident and that the use of force in the Rice case was wholly unjustified and there is no other reasonable interpretation of the facts. In both cases, the accusation of racism is still unfounded.
The Washington Post discovered that, from 2014–2016, 50 out of a total of 86 incidents of police use of lethal force involving the misidentification of a toy gun were committed against whites. If the vast majority of such shootings involved white victims, how can the instances involving black people be attributed to racism, rather than to the same factors that are assumed to be in play when both individuals are white? Why do the mainstream narratives draw such wildly different conclusions in the shootings of Rice and Brown, compared to the conclusions they reach after a shooting like that of Daniel Shaver?
Daniel Shaver, a white man, was shot after a bystander called the police, claiming they saw a rifle barrel sticking out of the window of Shaver’s apartment building, which was later determined to be the barrel of a pellet gun. The incident, like the Rice case, is tragic. The highly disturbing video shows the police officers giving contradictory instructions to Shaver to crawl on the floor towards the officers. For six minutes, Shaver did his best to comply with the officers, sobbing hysterically as the police, almost eagerly, repeatedly threatened to kill him if he failed to adhere to their commands. This resulted in the world’s most dangerous game of Simon Says—and, like all those who play that game, Shaver lost.
Given the cross-racial nature of these three incidents, and the overtly tragic nature of the Shaver and Rice cases, no rational, neutral observer could say with any degree of certainty that any of these instances were influenced by race. As John McWhorter notes, these incidents are directly comparable, but the public simply does not hear about the incidents involving white victims. As the video shows, the Daniel Shaver killing was at least as gut-wrenching as that of George Floyd.
Tragedies happen. Sometimes no one is at fault. Sometimes someone definitely is. Brutal, senseless murders also happen. It is unreasonable to believe that, every time such an incident occurs across racial lines, race was a contributing factor, given no other indication of motive. By focusing on the race of the individuals involved, when there is no evidence to suggest that race was a factor, the actual point of the debate is shelved and progress on the real issue—police brutality—is sidelined.
I am not against observing patterns in the data, but the fallacies I have outlined, which often crop up in these discussions, must be addressed, otherwise the observed patterns are just a case of apophenia. The failure to take confounding variables into account and the attribution of motive and intent, when neither are clear, lead to confirmation bias.
Some might argue that implicit bias tests can give us the data on racist motivations that we need. However, the implicit bias test is methodologically flawed. And, more importantly, averages taken from a set of data cannot be applied to individual data points, even if the data has been obtained in a methodologically sound manner. In other words, even if white people or police, on average, have implicit biases against the black population, this does not mean that one can use that fact to draw a conclusion about an individual person or event. Drawing conclusions about individuals from perceived, or real, patterns among the group to which they belong is the foundation of bigotry.
Falsely Equating Past Injustices with Current Injustices
The third major flaw in these arguments about systemic racism is the conflation of past injustices with current injustices. No one denies that the United States went from a system of racist slavery to racist Jim Crow laws. Even after the passage of the Civil Rights Act in 1964, moral atrocities ranging from the Tuskegee experiments to redlining continued. However, it is now 2020, and we have made considerable moral progress. The atrocities of the past have ended and we are right to be skeptical of claims that America continues to be a deeply racist country.
Slavery and Jim Crow will always be part of our past and the effects of these policies continue to reverberate. Most of the disparities of outcome we see in the real world would probably still exist if we were able to extirpate all trace of racism from the hearts and minds of every American individual. If this is true, then present day racism is not a major contributing factor to the disparate outcomes we see today.
For example, take a black family who were the victims of the redlining practices of the 1930s or 40s. Due to that family’s inability to get a housing loan in the past, coupled with the steady compound inflation of home values that accounts for much of the average white person’s net worth, the descendants of that black family are likely to be significantly poorer than their white counterparts. This is a real tragedy. But the elimination of redlining ended that particular form of oppression. The effects of redlining on this hypothetical family are not due to racism in the present day, but to the effects of past racism on the present. A reasonable claim can be made that we ought to address this and other injustices inflicted upon any individual by our government, but it is far less reasonable to claim this constitutes ongoing discrimination. Moreover, such a claim of ongoing discrimination muddies the conversation when talked about in the same breath as other claims of discrimination, such as those involving police brutality. The claim that America is systemically racist is a claim about the present that cannot be supported by pointing to the past, but only by pointing to the present.
Rejecting Identity Politics
Even if there were no racial prejudice in the hearts of men and women, there would still be economic disparities, largely segregated neighborhoods, busybodies would still call the police on people for no good reason and cross-racial murders would still occur. It would take an active effort to change these things—not merely a woke epiphany on the part of white people. Of course, active change should be pursued, but as part of a liberal doctrine, which treats all individuals as equals, regardless of race—in stark opposition to the tribalism that dominates the current conversation. Invest in black communities should be invest in struggling communities and black lives matter should be a tautology.
Instead, the calls for reform are undergirded by a very specific ideology. As Ibram X. Kendi puts it, “The origins of racism cannot be separated from the origins of capitalism. The origins of capitalism cannot be separated from the origins of racism … In order to truly be antiracist, you also have to be anti-capitalist.”
The call to eradicate racism should not require an individual to accept or reject a specific economic system. Massive disparities exist within races, as well as across races. If the disparities themselves are unjust, then to focus heavily or solely on inter-racial disparities indicates that those involved in the anti-racist movement are not focused on the well-being of all, but only the well-being of some. There is simply no excuse for making the moral justification for one’s generosity dependent on race. The claim that black people are suffering, while true in a general sense, does not adequately capture the landscape of suffering in America. The white families on food stamps in the rural Midwest are no less deserving of charity or subsidies than similarly situated black families, just because they share a skin tone with Jeff Bezos or Bull Connor.
The changes we advocate to create a more just society should have a strong philosophical basis that does not rely on race to determine how time and resources are allocated. Help those who suffer from exposure to lead paint in their homes, for example, but do not justify the expenditure of resources on the grounds that some arbitrarily defined disproportionate number of those suffering are black.
Advocate for a change in the metrics by which merit is measured, to more adequately capture the aptitude of those who do not perform as well on traditional metrics, but have overcome other challenges than their more privileged peers. Do not use race as a proxy for those challenges. The SAT or LSAT, GPA, extracurriculars and strong essay-writing skills are incomplete measures of merit for those pursuing tertiary education. All other things being equal, an individual who scores a 1600 on her SAT but comes from a privileged, two-parent, middle-class household in a low crime area has demonstrated less merit or general competence than someone who scores 1600 but comes from an underprivileged, single-parent, poverty-stricken household in a high crime area. Race-neutral arguments like this, however, are often not much more than a footnote to conversations about disparate racial outcomes, when they should be our focus. Real, positive change could come about if we could just reframe the conversation from the SATs are racist—which can be easily rebutted as math and English questions cannot be racist—to the SATs underevaluate millions, which gives us the opportunity to make concrete proposals to change that. These misguided conversations hinder social progress or even cause us to regress. And that is truly a shame for all Americans, regardless of race.