Police shootings of unarmed black Americans raise the question of why they occurred. While some might quickly claim the explanation is racism and brutality, the most plausible account involves several factors.

The first, which is connected to racism, is implicit bias. Studies involving simulators show that officers are more likely to use force against a black suspect than a white suspect. This has been explained in terms of an implicit negative bias. These studies show that even black and Hispanic officers are more likely to use force against black suspects. The studies also show civilians are more likely than officers to use force and exhibit more racial bias.

One reason why an implicit bias can lead to the use of force is how it shapes actions and perception. When a person thinks they are at risk, they become vigilant and anticipate the possibility they will be attacked. As such, the person starts to see the world through a “threat filter.”  So, for example, a person reaching rapidly to grab his wallet can be seen as reaching for a weapon. Perceptual errors occur often. For example, people who are afraid of snakes often see every vine or stick as a snake. These perceptual errors also help explain shootings: someone might think they saw the suspect reaching for a weapon.

Since the main difference between officers and civilians is most likely the training, it is reasonable to conclude that the training is having some positive effect. However, the existence of racial disparity in the use of force shows there is still a problem. One point of concern is that the bias might be so embedded in American culture that training will not eliminate it. As such, eliminating bias in police requires eliminating it in society, which goes far beyond addressing problems with policing. But is something that should be done.

A second often mentioned factor is “warrior culture.” Visually, this is exemplified by police use of military equipment, such as armored personal carriers and combat infantry equipment. However, the warrior culture is not just a matter of equipment, but also one of attitude. While police training includes conflict resolution, there is a significant emphasis on combat skills, especially firearms. On the one hand, this makes sense. People who are going to be using weapons need to be properly trained. On the other hand, it is reasonable to be concerned that there is more focus on combat training than the peaceful resolution of conflicts.

Since I have seen absurd and useless “training” in conflict resolution, I understand there are concerns about such training. I also understand that conflict resolution is often presented as if it is just “holding hands and drinking chamomile tea together” and hence not appealing to people who are interested in “real” police work. However, it is a critical skill. After all, in a crisis people fall back on habit and training. Those who train primarily for combat will fall back on that skill set. Naturally, there is the worry that too much emphasis on conflict resolution could put officers in danger. However, this is a practical matter that can be addressed.

A critical part of conflict resolution training is also what Aristotle would see as moral education: developing the right sort of character to know when and how to act correctly. As Aristotle said, it is easy to be angry, but it is hard to be angry at the right time for the right reasons, towards the right people and to the right degree. As Aristotle also said, this is hard, and most people are bad at it. This does present a challenge even for a well-trained officer since conflict resolution involves (at least) two people and the person they are dealing with is probably horrible at it. One possible solution is training for citizens, in interacting with the police (and each other). Expecting the full burden of conflict resolution to fall upon the police is unfair and not a successful strategy.

The final factor I will consider is the principle of the primacy of officer survival. A primary goal of police training and practice is officer survival. It would, obviously, be absurd to claim that police should not be trained in survival or that police practices should not put an emphasis on the survival of officers.  However, there are legitimate concerns about the consequences of this approach.

Part of the problem, as some see it, links to the warrior mentality. The police, it is claimed, are trained to see their job as incredibly dangerous and policing as a combat mission. This, obviously enough, shapes the reaction of officers, which ties into perceptual bias. If an office believes she is going into a combat zone, she will perceive people and actions through this “combat zone filter.” So, people will be seen as more threatening, actions will be more likely to be interpreted as hostile and objects will be more likely to be perceived as weapons. As such, it makes sense that training officers for survival by getting them to see police work as a combat mission would cause more civilian causalities than would alternate approaches.

Naturally, it can be argued that officers do not, in general, have a “combat zone” attitude and that academics are presenting the emphasis on survival in the wrong way. It can also be argued that the “combat zone” attitude is real but is also correct since some people target police officers and almost any situation could become a battle for survival.  As such, it would be morally irresponsible to fail to train officers for survival, to not instill in them a proper sense of fear, and to not engage in practices that focus on officers making it home at the end of the shift. Even if this approach results in more civilian deaths, including the deaths of unarmed civilians.

This leads to a moral concern about the degree of risk a person is obligated to take to minimize the harm to another. This matter is not just connected to the issue of the use of force by police, but also the broader issue of self-defense.

I do assume there is a moral right to self-defense and that police officers do not lose this right when acting in their professional capacity. That is, a person has a right to harm another person when legitimately defending her life, liberty or property against an unwarranted attack. Even if such a right is accepted, there is still the question of the degree of force a person is justified in using and to what extent a person should limit her response to minimize harm to the attacker.

In terms of the degree of force, the easy and obvious answer is that the force should be proportional to the threat and sufficient to end the threat. For example, when I was a boy, I was subject to the usual boy versus boy violence. Since these attacks usually just involved fists and grappling, a proportional response was to hit back hard enough to make the other boy stop. Grabbing a rock, a bat or pulling a knife would be disproportional. As another example, if someone is shooting at a police officer, then she would certainly be right to use her firearm since that would be a proportional response.

One practical and moral concern about the proportional response is that the attacker might escalate. For example, if Bob swings on Mary and she lands a solid punch to his face, he might pull out a knife and stab her. If Mary had simply shot Bob, she would not have been stabbed because Bob would be wounded or dead. As such, some would argue that the response to an attack should be disproportional and aimed at ending the encounter through superior force. In terms of the moral justification, this would rest on the fact that the attacker is engaged in an unjust action and the person attacked has reason to think, as Locke argued, that the person might intend to kill her.

Another practical and moral concern is that if someone “plays fair” by responding in a proportional manner, she risks losing the encounter. For example, if Big Bertha swings on Small Sam and Sam sticks with his fists, Bertha might be able to beat him. Since dealing with an attacker is not a sporting event, the idea of “fair play” seems absurd—hence the person who is unjustly attacked has the moral right to respond in a disproportional manner. 

However, there is also the counter-concern that a disproportional response would be excessive in the sense of being unnecessary. For example, if Bertha swings at Sam and Sam shoots her four times with a twelve gauge, Sam is now safe. But if Sam could have used a Taser to stop Bertha, then the use of the shotgun would seem to have been wrong—after all, he did not need to kill Bertha in order to save himself. As such, it would seem reasonable to hold to the moral principle that the force should be sufficient for defense, but not excessive.

The obvious practical challenge is judging what would be sufficient and what would be excessive. Laws that address self-defense are usually vague: a person can use deadly force when facing a “reasonable perceived threat.” That is, the person must have a reasonable belief that there is a threat and there is usually no requirement that the threat must be real. To use the stock example, if a man points a realistic looking toy gun at an officer and says he is going to kill her, the officer will have a reasonable belief that there is a threat. Of course, there are problems with threat assessment. As noted above, implicit bias, warrior mentality and survival focus can cause a person to greatly overestimate a threat (or see one where it does not exist).

The challenge of judging sufficient force in response to a perceived threat is directly connected with the moral concern about the degree of risk a person is obligated to face to avoid (excessively) harming another person.  After all, a person could “best” ensure her safety by responding to every perceived threat with maximum lethal force. If she responds with less force or delays her response, then she is at ever increasing risk. If she accepts too little risk, she would be acting wrongly towards the person threatening her. If she accepts too much risk, she would be acting wrongly towards herself and anyone she is protecting.

A general and generic approach would be to model the obligation of risk on the proportional response approach. That is, the risk one is obligated to take is proportional to the situation at hand. This then leads to the problem of working out the details of the specific situation. Which is to state the obvious: the degree of risk rests heavily on the circumstances.

However, there are general factors that would impact the degree of risk on is obligated to take. One would be the relation between the people. For example, it seems reasonable to hold that people have greater obligations to accept risk to avoid harming people they love or care about. Another factor that seems relevant is the person’s profession. For example, soldiers are expected to take some risks to avoid killing civilians, even when doing so puts them in some danger. To use a specific example, soldiers on patrol could increase their chance of survival by killing any unidentified person (adult or child) that approaches them. However, being a soldier and not simply a rampaging killer requires that soldiers accept some risks to avoid murdering innocents.

In the case of police officers, it could be argued that their profession obligates them to take greater risks to avoid harming others. Since their professed duty is to serve and protect, it can be argued that the survival of those who they are supposed to protect should be given equal weight to the survival of the officer. That is, the focus should be on everyone going home. In terms of how this would be implemented, the usual practice would be training and changes to rules regarding use of force. Limiting officer use of force can be seen as generating greater risk for the officers, but the goal would be to reduce the harm done to civilians. Since the police are supposed to protect people, they are (it might be argued) under greater obligation to accept risk than civilians.

One obvious reply to this is that some officers already have this view and take considerable risks to avoid harming people, even if they would be justified in using force. These officers save many lives, although sometimes at the cost of their own. Another reply is that this sort of view would get officers killed because they would be too concerned about not harming suspects and not concerned enough about their own survival. That is a reasonable concern and there is a challenge of balancing the safety of the public and the safety of officers.