Is there an explosive increase in police violence? There is no question that American police use their weapons more than police in any other developed democracy. Unfortunately, we don’t have fully accurate information about the number or nature of homicides at the hands of police. Despite a 2006 law requiring the reporting of this information (reauthorized in 2014), many police departments do not comply. Researchers have had to rely on independent information such as local news stories to cobble together numbers. One effort by the Guardian and Washington Post documented 1,100 deaths in 2014, 991 in 2015, and 1,080 in 2016—fewer than in the 1960s and 1970s, but still far too many.²

African Americans are disproportionately victims of police shootings; black teens are up to twenty-one times more likely than white teens to be killed by police,³ though these rates are often proportional to the race of gun offenders and shooting victims more broadly.⁴ Racial profiling remains widespread, and many communities of color experience invasive and disrespectful policing. The recent cases of Ferguson and North Charleston are hardly outliers; blacks and Latinos are overwhelmingly the targets of low-level police interactions, from traffic tickets to searches to arrests for minor infractions, and frequently report being treated in a hostile and degrading manner despite having done nothing wrong.⁵ In New York City 80 to 90 percent of those targeted for such interactions are people of color.⁶

This form of policing is based on a mindset that people of color commit more crime and therefore must be subjected to harsher police tactics. Police argue that residents in high-crime communities often demand police action. What is left out is that these communities also ask for better schools, parks, libraries, and jobs, but these services are rarely provided. They lack the political power to obtain real services and support to make their communities safer and healthier. The reality is that middle-class and wealthy white communities would put a stop to the constant harassment and humiliation meted out by police in communities of color, no matter the crime rate.

Those who question the police or their authority are frequently subjected to verbal threats and physical attacks. In 2012, young Harlem resident Alvin Cruz, who had been repeatedly stopped and searched by police without justification, taped an encounter with police in which he questioned the reason for the stop. In response, the police officer cursed at him, twisted his arm behind his back, and said, “Dude, I’m gonna break your fuckin’ arm, then I’m gonna punch you in the fuckin’ face.”⁷

Even wealthy and more powerful people of color are not immune: in 2009, Harvard professor and PBS personality Henry Louis Gates Jr. was arrested by Cambridge police in his own home; he had lost his keys, and a neighbor had called the police to report a break-in. The incident prompted President Obama to state:

Excessive use of force, however, is just the tip of the iceberg of over-policing. There are currently more than 2 million Americans in prison or jail and another 4 million on probation or parole. Many have lost the right to vote; most will have severe difficulties in finding work upon release and will never recover from the lost earnings and work experience. Many have had their ties to their families irrevocably damaged and have been driven into more serious and violent criminality. Despite numerous well-documented cases of false arrests and convictions, the vast majority of these arrests and convictions have been conducted lawfully and according to proper procedure—but their effects on individuals and communities are incredibly destructive.

Such training ignores two important factors in Garner’s death. The first is the officers’ casual disregard for his well-being, ignoring his cries of “I can’t breathe,” and their seeming indifferent reaction to his near lifelessness while awaiting an ambulance. This is a problem of values and seems to go to the heart of the claim that, for too many police, black lives don’t matter. The second is “broken windows”-style policing, which targets low-level infractions for intensive, invasive, and aggressive enforcement. This theory was first laid out in 1982 by criminologists James Q. Wilson and George Kelling.¹² They presented existing behavioral research that showed that when a car is left unattended on a street it is usually left alone, but if just one window of the car is broken, the car is quickly vandalized. The lesson: failure to indicate care and maintenance will unleash people’s latent destructive tendencies. Therefore, if cities want to establish or maintain crime-free neighborhoods they must take action to ensure that residents feel the pressure to conform to civilized norms of public behavior. The best way to accomplish this is to use police to remind people in subtle and not-so-subtle ways that disorderly, unruly, and antisocial behavior are unacceptable. When this doesn’t happen, people’s baser instincts will take hold and predatory behavior will reign, in a return to a Hobbesian “war of all against all.”

The emergence of this theory in 1982 is tied to a larger arc of urban neoconservative thinking going back to the 1960s. Wilson’s former mentor and collaborator, Edward Banfield, a close associate of neoliberal economist Milton Friedman at the University of Chicago, parented many of the ideas that came to make up the new conservative consensus on cities. In his seminal 1970 work The Unheavenly City, Banfield argues that the poor are trapped in a culture of poverty that makes them largely immune to government assistance:

Wilson, following Banfield, believed strongly that there were profound limits on what government could do to help the poor. Financial investment in them would be squandered; new services would go unused or be destroyed; they would continue in their slothful and destructive ways. Since the root of the problem was either an essentially moral and cultural failure or a lack of external controls to regulate inherently destructive human urges, the solution had to take the form of punitive social control mechanisms to restore order and neighborhood stability.¹⁵

Wilson’s views were informed by a borderline racism that emerged as a mix of biological and cultural explanations for the “inferiority” of poor blacks. Wilson co-authored the book Crime and Human Nature with Richard Herrnstein, which argued that there were important biological determinants of criminality.¹⁶ While race was not one of the core determinants, language about IQ and body type opened the door to a kind of sociobiology that led Herrnstein to coauthor the openly racist The Bell Curve with Charles Murray, who was also a close associate of Wilson.¹⁷

What was needed to stem this tide of declining civility, they argued, was to empower the police to not just fight crime but to become agents of moral authority on the streets. The new role for the police was to intervene in the quotidian disorders of urban life that contributed to the sense that “anything goes.” The broken-windows theory magically reverses the well-understood causal relationship between crime and poverty, arguing that poverty and social disorganization are the result, not the cause, of crime and that the disorderly behavior of the growing “underclass” threatens to destroy the very fabric of cities.

Broken-windows policing is at root a deeply conservative attempt to shift the burden of responsibility for declining living conditions onto the poor themselves and to argue that the solution to all social ills is increasingly aggressive, invasive, and restrictive forms of policing that involve more arrests, more harassment, and ultimately more violence. As inequality continues to increase, so will homelessness and public disorder, and as long as people continue to embrace the use of police to manage disorder, we will see a continual increase in the scope of police power and authority at the expense of human and civil rights.

The order to arrest Eric Garner came from the very top echelons of the department, in response to complaints from local merchants about illegal cigarette sales. Treating this as a crime requiring the deployment of a special plainclothes unit, two sergeants, and uniformed backup seems excessive and pointless. Garner had experienced over a dozen previous police contacts in similar circumstances, including stints in jail; this had done nothing to change his behavior or improve his or the community’s circumstances. No amount of procedural training will solve this fundamental flaw in public policy.

Many advocates also call for cultural sensitivity trainings designed to reduce racial and ethnic bias. A lot of this training is based on the idea that most people have at least some unexamined stereotypes and biases that they are not consciously aware of but that influence their behavior. Controlled experiments consistently show that people are quicker and more likely to shoot at a black target than a white one in simulations. Trainings such as “Fair and Impartial Policing” use role-playing and simulations to help officers see and consciously adjust for these biases.¹⁸ Diversity and multicultural training is not a new idea, nor is it terribly effective. Most officers have already been through some form of diversity training and tend to describe it as politically motived, feel-good programming divorced from the realities of street policing. Researchers have found no impact on problems like racial disparities in traffic stops or marijuana arrests; both implicit and explicit bias remain, even after targeted and intensive training. This is not necessarily because officers remain committed to their racial biases, though this can be true,¹⁹ but because institutional pressures remain intact.

American police receive a great deal of training. Almost all officers attend an organized police academy and many have prior college and or military experience. There is also ongoing training; large departments have their own large training staff, while smaller departments rely on state and regional training centers. Many states have unified Police Officer Standards and Training (POST) agencies that set minimum standards, develop training plans, and advise on best practices. While police training standards are still more decentralized in the United States than in many countries that have national police forces and academies, the new POST system has gone a long way in raising standards and creating greater uniformity of procedures.

However, even after training officers often have inadequate knowledge of the laws they are tasked to enforce. Police regularly disperse young people from street corners without a legal basis, conduct searches without probable cause, and in some cases take enforcement action based on inaccurate knowledge of the law. In Victoria, Texas, an officer assaulted an elderly man he had pulled over for not having a registration sticker on his license plate. The man tried to explain that the vehicle had a dealers’ plate, which in Texas is exempt from the sticker requirement. When the officer refused to listen, the man attempted to summon his boss at the car dealership where the confrontation was occurring. Rather than working to resolve the mistake, the officer attempted to arrest the man and in the process injured him with a Taser so badly that he was hospitalized.²⁰ In the subsequent inquiry, the officer insisted that the man’s passive resistance was a threat that had to be neutralized. Since the incident was recorded on the dashboard camera of the police cruiser, the officer was fired.

The training police receive at the academy is often quite different from what they learn from training officers and pe...