ben vandgrift · on living the dream

We have a policing problem in the United States.

Written from Charlotte, NC following the shooting of Keith Scott and the aftermath1

I don't know what happened. I know there are conflicting stories, and maybe a video that the police have but won't release because 'take our word for it', and that sounds a little shady to me. A protester was shot in the head last night; people at the scene say an officer did it, the official word from the police is that they didn't, because 'take our word for it'. At the end of the day, I don't know what happened.

There are people who say it isn't my problem because I'm a few blocks away and a reasonably affluent white dude, and these problems are never really going to be my problems. Thing is, I'm also a human being. So yes, they are my problems.

Here's what I know: if you're a black man between the ages of 15 and 29, you're three times as likely to get killed by an officer as if you're a white man2. Since September 14, at least 23 people have been killed by the police in the United States. Two were mentally disabled, five were black, and the race of seven of the dead were unspecified. The two unarmed dead were black, as was the 13 year old 'armed' with the BB gun.

I also know that testing shows that the opinion of black Americans consistently demonstrate an implicit bias against black Americans. That is, a black officer is just as likely to be suspicious of a black suspect and react accordingly.

Given that Keith Scott was black and he was killed by the CMPD, this is a little about race. It's just as much, if not more about the quality and restraint that law enforcement nationwide isn't showing. Solving that will solve the racial disparity on the streets, and hopefully the sentiment will trickle upward into the courts, where there are other obvious bias problems.

Our police are undertrained. They have too little oversight. The wrong people are entering law enforcement. Consider:

The average amount of time that police are trained for their job before being handed a gun is 19 weeks, 16 weeks in North Carolina. The average for hairdressers is 26 weeks, 37.5 weeks and required performances for a cosmetologist's license in NC. A funeral service worker usually require two years of education and a year of internship. An basic certification as an EMT requires between 26 weeks and two years, depending on state and level of certification.

Near as I can figure, in North Carolina: the people who make you beautiful, save your life, and prepare your death all require more training than the person who is authorized to take your life at the least provocation. Does 'least provocation' seem inflammatory? Terrence Crutcher, an unarmed black man, was killed by police with his hands up, for the crime of being in the middle of the road3. Tyre King, 13, was shot to death while running away from police because he was carrying a BB gun4. These were days ago, and are not abnormal behavior for police.

There are around 800,000 officers with arrest power employed either full- or part-time in the United States. Last year, those officers killed 1,1465 people, 230 of them unarmed. Nineteen of them were under 18. One of those was unarmed, six years old, and a passenger in his father's car.

Prior to 2015, the number of killings by police each year were tracked (badly)6 by the Bureau of Justice Statistics, and averaging 928 per year during all years examined (2003-09)7.

1,146 people last year may not sound like a lot in the context of a nation of 319 million, but consider: in the 15 years since the 2001 invasion of Afghanistan by the United States, 6,749 Americans have died in Afghanistan8 and subsequent war in Iraq9. According to the BJS estimates (which have been found low by about a quarter) at least 13,920 people were killed by our police force during the same time period. In two overseas wars spanning 15 years, fewer Americans have died on foreign battlefields at the hands of our national enemies than in our streets at the hands of our law enforcement, by a factor of two.

Let's slice the data a different way, using 2014's numbers and population, the most recent available on WISQARS10. In 2014, 15,809 homicides were committed in the U.S. (pop. 318,857,056). That gives an incidence rate per 100,000 population of 4.96. Including all injuries leading to death (199,756), the incidence rate is 62.65—quite the jump, when you include accidents, suicide, fire, automobile, industrial machinery, etc. In 2014, using the low estimate provided by the BJS (12,064 over 2002-2014), the incidence rate for law enforcement was 116. If we used the corrected estimates, 116 becomes 155. If we use 2015's numbers from The Guardian5, we get 143.25.

One of the things I love about numbers is their purity. 116 is clearly greater than 62.65. As a citizen of the United States, you are more likely to die from injury at the hands of the police than from any other fatal injury by a factor of two. We can correctly identify law enforcement as the most dangerous organization in the United States.

What about terrorism? Including the World Trade Center incident, there have been 3,064 domestic victims of terrorism from 2001-201411. Excluding 2001 and the WTC as a clear statistical outlier12 and committed by foreign nationals, that number drops to 61 victims of terrorism between Jan 1, 2002 and Dec 31, 2014, compared to 12,064 law enforcement deaths over the same time period. We have no idea how many terrorists are living in the United States, so we can't establish an incidence rate. However, in order for 'terrorists' of any stripe to be more dangerous than law enforcement, in order to achieve an incidence rate per 100k population equivalent to that of law enforcement, there must be fewer than 4,048.45 terrorists living in the United States. In a nation of more than 318,000,000 people, a population smaller than the attendance of an average Nuggets13 game is insignificant compared to 800,000 well-armed and poorly-trained authoritarians deployed to every street in the country.

We have a policing problem in the United States. When discussing this recently, I received the following question:

"Isn't it misleading to talk about all police related fatalities without regard to the percentage that were justified?"

It is absolutely not misleading to talk about numbers absent motivation. It is in fact the only way you can talk about them. Numbers don't feel. Risk doesn't suppose. I don't know why law enforcement officials kill so readily. Only that they do. This is primarily a statistical summary, not an inference of motive.

Besides, where could I find impartial data on justification? There is an a priori question here: is the killing of an American citizen by a member of an organization sworn to serve and protect ever justified? Now we're out of statistics and into ethics. You first have to establish whether or not any justification is even possible, then you can talk about what it means to be justified. Until you can quantify justification in a measurable way, it has no place in a discussion of numbers.

Complicating matters: it is not possible to judge the innocence or guilt of all of victims of legal intervention under U.S. law, because at least some of them didn't see the inside of a courtroom, just a morgue. Given that, how can we establish justifiability on legal grounds? We could estimate, based on those who died due to injury while in police custody after having been convicted and sentenced, but then we introduce a sampling error due to the estimation. Is there another metric I can use? Does that metric normalize for observed conviction and sentencing bias? At what adjusted sentencing level is it justifiable for the officer on the scene to have killed the victim? 5 years? 25 years? My personal ethic says that it's never justifiable for an officer to kill anyone except to protect the life of an innocent bystander, but then, what does 'innocent' mean? I don't consider officers in the line of duty to be innocent bystanders. I'm not as convinced it's possible to build an effective legal system around these ideas. Given that our existing system is ineffective, where's the harm in trying?

At the bottom of all of this are hard, speculative ethical questions. They have nothing to do with statistics, risk, incidence, or definitions. I want to stick to numbers and absolve myself of opinion, but increasingly that seems like a coward's excuse. So I'll say it again, loudly, then suggest solutions.

We have a policing problem in the United States. The officers are undertrained, overarmed, militarized, and given far too much leeway in the performance of their duties. Given the serious nature of the problem, and law enforcement's unwillingness to address it with any seriousness at the local level, regulation has to come from the federal level. Consider the following, which would effectively solve the problem if implemented in total:

  1. Federally mandated minimum training and certification. At least as much training as the medical staff that have to clean up their messes. Two years of training, an unarmed apprenticeship, regular retraining and recertification by an independent body. If you're insufficiently trained, you can't carry anything more dangerous than a badge while on duty. Related: officers not on duty are not empowered to act as officers under any circumstances.

  2. Physical fitness requirements, assessed quarterly. Every officer should be able to effectively chase a suspect fleeing on foot without resorting to their weaponry.

  3. Removal of all military weaponry, vehicles, attire, and defenses from all law enforcement personnel save for a very limited supply of protective gear for bomb squads and rapid response teams, scaled by city population.

  4. Establish an independent court (Blue Court) whose only responsiblity is the investigation of police crime and misconduct, overseen by the Department of Justice. This removes the tendency of prosecutors to indulge bias against officers. Officer trials will be conducted solely by this court. Officers awaiting trail will remain in custody without bail.

  5. Body cameras on every officer, federally mandated. Choosing to pursue a career in law enforcement implies consent. Like other implied consent laws, if your camera is off, the officer's guilt is assumed, and the burden falls on them to establish their innocence. If it's off during a shooting, the charge is murder. If it's a beating, aggravated assault. Two cameras to be checked out and tested before the shift and returned after, with one on standby in the case of a equipment malfunction.

  6. Every officer at the scene without a camera during a shooting or other abuse of power shares the sentence.

  7. Every officer at the scene with a camera on who did not intervene is an accessory to the crime.

  8. Every discharge of a firearm, taser, pepper spray or any other violent or suppressive action that results in an injury carries with it suspension without pay until a DoJ investigation is concluded clearing the officers and accessories of any wrongdoing.

It's clear that the police are not interested in holding themselves to a higher standard, either personally or as an organization. Courts trying officers are jurisdictional, and prosecutors are inclined to protect the officers with whom they work and socialize. So the citizenry, and therefore the federal government must. Our justice system loves zero tolerance policies, implied consent, and mandatory minimums. It loves surveillance. Let's apply those same policies to the most dangerous civilians in the country.

Will this result in fewer people becoming officers of the law? Maybe. There's no way of knowing without studies that I'm not equipped to do right now, but my suspicion is that it will result in fewer of the wrong people becoming officers of the law. It will result in more of the wrong people being remanded into custody themselves. These measures create jobs within the Department of Justice as well, given the need for increased staff to monitor police activities.

How long do these emergency regulations need to remain in place? Until the culture changes. Until those who abuse their authority, or are willing to kill at the slightest provocation are behind bars. Until a) we can demonstrate that the problem is a few bad apples, or b) we have recycled the system entirely, and it's safe for all citizens to go about their business without fearing for their lives every day.

These problems are my problems, because I'm human. They're your problems too. As individuals, we're not very powerful when compared to the law enforcement complex, but we do have voices.

53 years ago, while sitting in a Birmingham, AL jail, the Reverend Martin Luther King, Jr. said this:

"I have almost reached the regrettable conclusion that the Negro's great stumbling block in the stride toward freedom is not the White Citizens Councillor or the Ku Klux Klanner but the white moderate who is more devoted to order than to justice; who prefers a negative peace which is the absence of tension to a positive peace which is the presence of justice;"

Sentiment-wise, not much has changed in the past half century. Do something—no one can do this by themselves, but contact the White House. Contact Congress. Make suggestions. Speak. Vote. Engage, nationally and locally. Or accept the fact that you're okay with these deaths, that offering aid and comfort is too much of an inconvenience.


  2. "If you were between 15 and 29 years of age between 2010 and 2014, the average incidence of injury death due to legal intervention if you were a black male was (291 / 26,093,455) 0.000011597 (0.001152%), and if you were a white male it was (478 / 127,417,292) 0.000003751 (0.0003751%). Simple division does the rest: (0.0003751/0.001152) => 3.07; your risk of dying due to police intervention was 3.07 times higher for black men as white men between the ages of 15 and 29, and between 2010 and 2014. It's still a REALLY small risk—say 1 in 266,000–but 3x higher than a white man's risk."



  5. 2









written: Sep 22 2016