Law & the Courts

Shouldn’t Police at Home Exhibit at Least as Much Discipline as Soldiers at War?

NYPD officers on 5th Avenue in New York City in 2016. (Lucas Jackson/Reuters)
A response to Jack Dunphy.

On November 22, 2007, I flew into Forward Operating Base Caldwell in eastern Diyala Province, Iraq. I was the squadron judge advocate for the Second Squadron, Third Armored Cavalry Regiment. One of my primary responsibilities was advising the command in making shoot/don’t shoot decisions according to the rules of engagement and the laws of war. In the course of that year, I helped make countless life-or-death decisions — decisions that not only placed the lives of my friends and brothers at risk but also determined whether we killed terrorists or killed civilians.

My commander was a wise man, and he knew that a JAG officer whose sole experience consisted of watching video feeds or hearing radio transmissions would render advice based not in the reality of war but in the abstractions on the screen. So I went outside the wire on patrols — not nearly as often as the daily missions of the cavalry scouts and armor officers I served with, but often enough to understand the reality of the danger. I walked the streets of local towns and villages. I experienced tense situations where you didn’t know whether to shoot or hold fire. I learned what it was like to see a car approaching and not know whether the occupant was a terrorist about to blow himself up or a civilian oblivious to the presence of American troops.

I learned what it was like to see a car approaching and not know whether the occupant was a terrorist about to blow himself up or a civilian oblivious to the presence of American troops.

It was a tough deployment, one that ultimately earned my squadron the Valorous Unit Award. In the months we were in Diyala, our troopers faced constant attacks. IEDs claimed lives. Men died to ambushes. Indirect fire was a frequent threat to our combat outposts. Our troopers fought pitched battles in the streets, called in air strikes, fired thousands of artillery rounds, and killed, wounded, and captured dozens of terrorists. By the end of the deployment, they’d reclaimed thousands of square kilometers from al-Qaeda and left it a broken, spent force.

Do you know how many innocent civilians we killed in that entire deployment, which spanned hundreds of engagements with the enemy? Exactly two. One to small-arms fire and one to a wayward artillery shell.


Over the past three years, as the issue of police shootings has come to periodically dominate American discourse, I’ve noticed a disturbing pattern. While many controversial police shootings are lawful and justifiable, many others would be surprising to see in a war zone, much less in the streets of America’s cities. Some of the names come easily to mind — Philando Castile, Daniel Shaver, Walter Scott, and (most recently) Stephon Clark.

Why is this case? One part of the answer is easy. Cops are human, and since they’re human some will be incompetent, some will panic, and some will be racist. But such shootings happen often enough in a nation that is still enjoying a respite from the horrific crime waves of the late 1980s and early 1990s — a nation where policing is far down on the list of the most dangerous jobs — that “cops are human” simply isn’t a sufficient answer.

Another answer came from Jack Dunphy, the “nom de cyber” of a retired LAPD officer who often contributes pieces to NR. Dunphy offers a valuable cop’s-eye view of police controversies, and his recent response to my piece criticizing police actions in the Clark shooting is worth reading in its entirely. One passage in particular stands out. Responding to my citation of statistics demonstrating that police killings in Sacramento were extraordinarily rare, and this “background risk” should be considered in ambiguous encounters, Dunphy says this:

Surprisingly, David also asks that officers, before deciding on a course of action in response to a fleeing suspect, perform an instant, on-the-spot risk analysis based on historical data.  “What is the background level of risk here?” he asks. “According to the City of Sacramento, it’s been almost 20 years since a cop was shot and killed in the line of duty.”

Yes, the last Sacramento Police Department officer shot to death was William C. Bean, Jr, who died in 1999. And the last before him was Doyle A. Popovich, who died in 1974. But the fact that a police officer is murdered in the city only once every 20 or 25 years does not lessen the risk of any individual encounter.

The surprise here is mutual. I’m surprised Dunphy would make this argument. Of course officers should have situational and tactical awareness of individual risk. Of course the infrequency of police killings (as one factor among many) is relevant to “any individual encounter,” especially when the encounter is with a suspected trespasser and vandal not known to be a violent felon.

But rather than emphasizing odds, probabilities, and patterns, training sometimes fills cops’ minds with ideas like, “The worst can always happen” or, “Any encounter can go bad.” These statements are true, but incomplete. They’re not the same thing as saying, “Every encounter is equally likely to go bad.” Good officers, like good soldiers, know that each encounter takes place against the background of a much larger context, with multiple factors influencing the outcome.

First, it’s important to understand that the mission must come before personal safety. When you sign up to wear the uniform, you’re tacitly acknowledging as much. This doesn’t mean you’re required to be reckless with your own life, of course: Prudence and self-protection still matter. But they come behind the purpose of the police force itself. If you have any doubt about this fact, ask the Broward County Sheriff’s office. The armed deputy at Marjory Stoneman Douglas High School certainly succeeded in protecting himself during Nikolas Cruz’s massacre. But he failed to do his job, placing his own safety above the safety of the innocent kids he was sworn to protect, and he rightly had to face the consequences afterward.

Second, it’s important to fully understand the mission. When your job is to preserve the safety, security, and — crucially — liberty of a community, each individual encounter is conducted against the backdrop of those broader, over-arching goals. So, a call to pursue a suspected vandal and trespasser (like in the Clark case) presents a multi-faceted challenge: Apprehend the suspect, protect his civil liberties, understand the community you’re policing, and protect the liberties and security of those others who live there, as well. Every confrontation is potentially dangerous, sure, but every confrontation is also complicated by the multifaceted balancing act we ask of our cops. One may argue that we ask too much of our cops, but I don’t think so; younger soldiers perform the same balancing act in more dangerous circumstances for less pay every day.

Third, the prudent rules of engagement should vary by the nature of the encounter. As I wrote in my initial piece about the Clark shooting, situational awareness demands different kinds of risk tolerance. Pursuit of an armed robber is different from pursuit of a vandal, and both are dramatically different from rolling up on an actual firefight, like the incident that claimed the life of a Sacramento sheriff’s deputy in 2017. While each situation can potentially turn deadly, it’s a simple fact that some kinds of encounters are more fraught with peril than others, and greater inherent peril demands greater latitude for police use of force.

Articulating reasons for your fear is not the same thing as articulating “reasonable fear.”

Fourth, fear must be subject to reason. Public defenses of police shootings tend to revolve around questions of fear. Officers consistently escape conviction, prosecution, and sometimes even discipline altogether because they are able to effectively articulate why they were afraid for their lives the moment they fired the fateful shot. The legal standard to escape conviction, however, is that they must prove not just that they were afraid but also that their fear was “reasonable.” Articulating reasons for your fear is not the same thing as articulating “reasonable fear.”

Let’s take, for example, the statement of Jeronimo Janez, the officer who shot and killed concealed-carry-permit holder Philando Castile at a traffic stop:

I don’t remember how many rounds I let off. Um, I remember seeing the last two rounds go off and I remember seeing one of those rounds hit him in the arm. Uh, his glasses flew off. I’m not sure if it was from gunfire or from him, uh, whipping his head back or anything like that. Uh, but, uh as that was happening, as he was pulling at, out his hand, I thought I was gonna die and I thought if he’s, if has the, the guts and the audacity to smoke marijuana in front of a five-year-old girl and risk her lungs and risk her life by giving her secondhand smoke and the front-seat passenger doing the same thing then what, what care does he give about me? And, I let off the rounds and then after the rounds were off, the little girls [sic] was screaming. I held the suspect at gunpoint. His arms came up into view. And they were up by his chest. I can’t remember what I said. But I acknowledged this little girl first. ’Cause I wanted her to be safe and I told Officer Kauser or I told her one of the two to go run out of the car and run to Office Kauser or Officer Kauser to get her. And then I turned my attention to the front-seat passenger. I didn’t point my firearm at her. I still had it on the suspect. ’Cause he was still moving.

Got that? One of the reasons Janez was afraid was that he thought Castile had disregarded the life of a five-year-old by “giving her secondhand smoke.” So naturally, he then decided to fire his gun into a car containing that very same five-year-old. In reality, Janez panicked, and while that is understandable, it is not a justification for shooting Castile.


To return to Iraq, here’s a concrete example of how awareness of the overall mission, personal risk, and personal courage can combine to demand restraint even when deadly force is legally authorized.

It was a terrible night in late winter, 2008. An IED had hit one of our Humvees, injuring two soldiers and killing one. While the medevac chopper was inbound, our guys spotted a small group of what looked like men lying in the prone position in a ditch. From that vantage point it appeared they could engage the helicopter before it could land. No weapons were immediately visible.

So the question became: Shoot or don’t shoot?

The law of armed conflict said that we could shoot. Men lying prone in a position of tactical advantage when troops are in range can legally be responded to with deadly force. But if the shooting would be permitted, would it be proper?

Fire and you neutralize the possible threat, making sure that the medevac helicopter can complete its mission. But if you hit civilians, though the medevac will still complete its mission, you have endangered the squadron’s larger objectives  of securing the village and recruiting allies to join friendly militias.

Immediately engaging risked the larger mission too much. Simply ignoring the threat risked the medevac too much. So we chose a middle course. We asked our men to emerge from cover and move to investigate and perhaps detain the suspicious men. With the burning Humvee behind them, our cavalry troops cautiously probed forward and discovered not men but a group of unarmed boys. They’d heard the “boom” of the IED, ran out to investigate, and then took cover when they saw American troops swarming around.

They had no weapons. They had no cameras (al-Qaeda sometimes used kids with cameras to do damage assessments after IED strikes). They were just kids, and had we killed them where they hid, the repercussions could have been immense.

I bring up examples like this not to claim that police officers usually or routinely don’t show this level of situational awareness and restraint. They do. Through training, courage, and instinct most officers understand each and every truth I’ve outlined above. It’s one reason why, even though cops routinely engage in tense confrontations, we don’t have more problematic shootings.

But when the police do fail, and they shoot a man who was, in actuality, no threat at all, it is no answer to criticism to stampede to the minimum legal standards of conduct, spin out all the things that “could” have gone wrong if the facts were different, and finally circle your wagons around “officer safety.”

A person can be concerned about officer safety and realize the truth that officer safety isn’t the mission.

A person can be concerned about officer safety and realize the truth that officer safety isn’t the mission. A person can believe blue lives matter and understand that accepting sometimes extraordinary risk is part of the job. A person can support the police and still demand a very high level of tactical and strategic awareness even from the youngest officers. To put them on the street is to declare to the public that they are up to the job.

The legal bar for successful prosecution of an officer is appropriately high. We should not send a cop to jail if he makes the snap judgment to fire his weapon while reasonably fearing death or serious bodily injury. But I’m concerned that juries are too willing to excuse fear in the absence of reason, and I’m concerned that our bar for training, competence, and courage is often too low.

Men in uniform inspire respect not because of the uniform itself but because of what the uniform is supposed to represent. It’s supposed to represent not just a commitment to selfless sacrifice but also a commitment to excellence. Countless cops exhibit those very characteristics. Too many others do not. In the face of this reality, the least we can ask is that cops show as much discipline as soldiers at war.


The Latest