Drop the Knife: Policing the Unpoliceable
I spent last week watching hundreds of videos of police officers shooting people. What prompted me to do this was working on the downloadable pdf workbooks which accompany the release of each episode of The Essential Sam Harris. I was building out a section for the upcoming episode on violence which features some of Sam’s most controversial arguments on topics like torture, profiling, and gun ownership. It’s a doozy (likely out in a month or so). But I went down the rabbit hole on the complicated ethics of gun ownership and crime which naturally led me to the quandary of how to police a society like the United States which is flooded with easily accessible guns.
In the pdf, and in this essay, I recommend the best collection of bodycam footage of police discharging their weapons which I’ve yet found. All of the linked videos included here are from that YouTube channel. I have to issue a huge graphic content warning. These are videos of violent human encounters. Some are videos of people losing their lives or getting seriously injured. These are videos of people in disturbing states of minds. These are also incredibly sad videos. These videos are not for everyone. I’m not sure they are for anyone really. But I ended up watching a ton of them. This, of course, does not make me any kind of expert. But I am feeling, thinking, and pondering things that I really just have to get off my chest. You, lucky reader, are on the receiving end.
I am going to try to refrain from making sweeping statements about the data on police violence and what exactly it implies. I used to keep a live google sheet which pulled in data from a good database of fatal police encounters that the Washington Post kept. Their data is here. I used to check my data from time to time and carve it up by race, gender, age, state etc… I honestly don’t think it tells you that much. But here is one general sweeping statement that I can make and firmly stand behind. No other country on Earth produces datasets like these from their citizens’ encounters with police. Not. Even. Close.
That statement is not meant to castigate the police as a bloodthirsty corrupt group of thugs. My week with violent police videos certainly does not leave me feeling that way. And I don’t think you should feel that way either. The issues here are complex and even when you process the data and locate correlations with variables like whether the suspect has a gun, a fake gun, an ax, a knife, a stick, a finger shaped like a gun, or none of the above - is never enough to condemn, praise, absolve, or prosecute the cops. The race, age, or gender of the suspect nor the police officer tells you all that much either. Emotionally charged conclusions about racial disparities, class inequalities, or murderous police are usually the game plan of the outrage dealers of the media. Don’t fall for that stuff. (I have a bit more on that later.)
Looking at a line in a dataset doesn’t tell you how or why these shootings are happening. In fact, that linked dataset just includes the incidents where the subject happened to die. A gunshot (or several) has fairly unpredictable outcomes when it comes to fatality. A single bullet can hit the right/wrong spot and a barrage of them can happen to miss those same spots. Though, of course, receiving one bullet is better than several of them if your aim is trying to survive.
Every shooting is unique. Every situation is different. They should all be scrutinized. But that is the interesting part. We are getting (finally) some increased ability to scrutinize each one due to the use of body cams and generally ubiquitous security cameras. Here is some good recent data about the increased usage and functionality of body-worn cameras (47% of general-purpose law enforcement agencies had acquired body-worn cameras; for large police departments, that number is 80%).
So, let’s scrutinize some encounters. I want to share some videos of specific incidents which contribute to the current jumbled thoughts and feelings in my conclusion. I categorize the videos into different “thought groups.”
Again, graphic warning… for real. And here is my “stay away from the comment section” caveat that basically should be applied to anything on the internet.
Thought #1: “Boy, there seems to be a lot of suicide by cop.”
There are a ton of examples of the police being called into a situation where a person is wielding something like a knife, or gun, and threatening himself or others. These encounters can go in several ways.
These examples go on and on. Sometimes, there are attempts to deploy “less-lethal” methods to subdue the subject. Oftentimes, they don’t work. When they don’t work, they can make the situation much worse. The subject hears a loud bang and suddenly has a strange fishing wire prong on them. This is the point they can lash out and charge at the officer - and get hit with the lethal option.
Thought #2: “Can’t they use tasers or something?”
The last clip above contains an ineffective use of a less-lethal option (a bean-bag round) before the barrage of lethal bullets are discharged. Here are some taser deployments:
Here’s an example of the ineffective use of a taser:
And another:
And another:
And here is a particularly disturbing one where the suspect was carrying a shovel.
And another awful example of the unsuccessful use which results in the aggressive response and lethal option:
Here is an example of the effective use of a taser:
Taser effectiveness might only be in the 40% range.
There is a large amount of “suicide by cop”. This phenomenon seems to be understudied in psychology. But here is one meta study which suggests it is between 10-36% of police shootings. That wide range is likely due to the vague criteria one would have to use to define an incident as suicide when it masquerades as a crime. Is someone who wanders down the streets of Hollywood with a lighter shaped like a gun in an unstable state of mind actually motivated by self-destruction and a subconscious type of knowledge that this is their life-exiting flourish? Or do we need the suspect to announce something as definitive as “Kill me! Kill me!” like this one which also unfortunately included a malfunctioning taser? Or this one with a suspect asking for the police to “end my existence”.
Think about how those 3 incidents would look in a datasheet. In none of them did the suspect hold a real firearm. The races, ages, and genders of the suspects and officers are all over the place. That kind of data would tell us almost nothing of interest.
How would one ever know how to characterize the “suicidal” variable of an incident like this one. Or like a few of the earlier incidents which include almost no conversation and dialogue between the police and the suspect (that last one is particularly disturbing and might result in disciplinary measures for the officer).
Some psychological speculation on “suicide by cop”: fear of pulling the trigger yourself must be real. It’s just hard to kill yourself. People fear that they will chicken out in the end. So, charging at a cop with a knife in a country like the United States where the police are instructed to shoot someone who is doing that is a decent bet. There is no hassle when it comes to getting a firearm to commit suicide. And I think even more importantly, there is likely a desire to make a very dramatic statement about the state of your mind when it comes to a vague sense of legacy about your life story. Shooting yourself and leaving a note carries with it a bit of shame and defeat, whereas charging at a cop after trying to rob a store and going out “like that” carries with it a response from family or friends of something like “Man, I guess he just snapped.” Or even better perhaps, “He must have gone crazy.” There is some kind of deep existential cry to “Notice my pain! Look how maddening the world is that I would do such a thing!” in those moments which is secretly relatable to many people.
The most obvious “suicide by cop” incident I happened to come across was this one. I have a feeling the 10%-36% range is a fairly big undercount if “suicidal” is defined a bit more broadly.
One note about the vast majority of the shootings of these types - the police officers often follow up the moment of pulling the trigger with something like “God, dammit!” The cops don’t usually “want” to shoot these people. You can hear the frustration when they begin to realize they are about to be caught between a rock and a hard place. “The rock” being the protocol which justifies and seems to demand the use of the firearm if the suspect makes an aggressive move towards them while brandishing a weapon, or is even pretending to. “The hard place” being the heaviness and trauma of taking someone else’s life. Simply ignoring a command to “show your hands” is enough to demand the protocol that the cop assumes you are holding a gun unless he can actually see your hands. See this example for a clear case of the mandated assumption that “unseen hands” actually contain “a loaded gun”. Even when there is some reason to believe that the person is in clear mental distress (of a romantic variation it seems) and is likely suicidal.
But even if the suspect doesn’t have a weapon, the police officer does. As is often noted, there is always a gun on the scene at these calls… at least in the USA. If a suspect knocks you out, or successfully grabs your weapon, you are dead. Here is an example of that factor. And another. And another.
Thought #3: “Sometimes the shot has to happen and you want it to be a good one”
There are plenty of situations where the police arrive in the midst of an incredibly dangerous and tense situation and the lives of non-suspects are in direct threat. These are clear hostage situations. These are the moments when the shooting itself is probably most easily justified - though the desire for a less-lethal option is still strong. Sometimes the skill and shooting techniques of the officers here is crucial and “impressive”. These can also be some of the saddest situations possible. But they show up on a spreadsheet absent the absolutely necessary context.
The flip side of the most easily morally (and legally) justified use of deadly force in these hostage situations is that police can always claim a general danger to the public when the suspect has a weapon… even if he or she is not actively holding a knife to anyone’s throat.
Thought #4: “Things can go from calm to deadly VERY quickly”
This thought is the one that every cop has in their mind at all times. They approach a car on a routine traffic stop and suddenly a gun is in their face and the trigger is pulled. This must be the reoccurring nightmare for every officer and family member of one. And it is a constant threat. How often does it happen? I have no idea. But the possibility of it happening is ever-present. In a few cases, it seems that the police failed to check for weapons or other dangers and that better training could have avoided the situation. But sometimes there is absolutely no warning for these deadly shifts and no way to avert them.
Thought #5: “Cops are terrible therapists”
So many of the incidents I watched were just crying out for a police officer to actually help someone through a mental health crisis. It seems obvious to state that having a gun pointed at your chest by your therapist is probably not very effective. The entire institution of policing seems unqualified to be the thing trying to talk people through these kinds of things. But, of course, the people can be threats to themselves and their family members in these distressed mental states. The justification for a police officer with a gun on the scene is not a hard thing to argue for. But it also seems absolutely nuts to have them do most of the talking. And the psychological techniques used by the well-intentioned officers in these clips are just… terrible.
Thought #6: “Sometimes these are VERY questionable”
I watched a lot of shootings that were borderline illegal… and some that absolutely were (and the courts agreed). There really do seem to be a lot of situations where the first 2 shots from the police officer’s gun were “justified” and understandable… but the next four or five or fifteen that come from the gun aren’t okay. When the first bullet is legally justified because the suspect did not drop a weapon when the cop asked them to, it seems that the next few bullets are also justified by the same disobedience if they happen quickly enough. And, of course, we have the luxury of watching these incidents in slow motion and slow them down with frame by frame analysis. These shootings usually happen very quickly. But these are very frustrating to watch. Some of them suggest an eagerness to use the gun. Or a reveling in the “seal being broken” by the legal justification of the first bullet and an effort to cram as many bullets into that legal window as possible.
Thought #7: “Some cops are really bad murderous people”
Bodycams are revealing some incidents that we probably would have had no chance of seeing in the previous decades. That last clip seems to be a homeless man. The shooting itself is borderline. I’m personally not okay with it. There may be some disciplinary consequences for the officer. It’s impossible to tell from the video itself if the cavalier shooting by the officer is influenced by the apparent low socioeconomic class of the suspect. But I do shutter to think of just how many incidents like this used to happen in the complete darkness with only a written report to tell the story… authored by the police themselves.
This one is the absolute worst video I watched. It was a downright execution.
Thought #8: “Accountability?”
The police officer in that last horrendous clip is in legal trouble. His was indicted for manslaughter until a judge recently sent that charge back to a new grand jury to hear new evidence. But, if I had to bet, I’d guess that he will be going to jail for several years. Perhaps a decade. Keep an eye on it. In fact, if you were brave enough to watch all the videos included in this essay, you may be encouraged to find out that many of the “iffy” ones are actually facing real charges - not all of them, but some of them. We all know about the high profile cases like George Floyd’s death which have resulted in significant charges and murder convictions. But even the clips in this post, which you probably didn’t hear much about, are being investigated.
And how come you probably didn’t hear about some of them? How come that last shooting didn’t lead to national protests and calls to defund the police? I’ve written about the racial complications in the data of police shootings before and why it matters and doesn’t. I recommend taking a look at that article… but here is another thought:
Thought #9: “The media is awful… and loves this carnage.”
Part of what prompted me to write this essay was coming across a recent Jon Oliver clip on the media and police shootings. That whole clip is here.:
There is plenty of interesting analysis in the clip about local media outlets accepting the police narrative of events without a critical and skeptical approach… but the last bit of the clip was the part that really irked me (Timestamped here). That’s a show called The Rising. That’s a guy named Robby Soave being dunked on by the public defender and then by Jon Oliver. Maybe if you watched it, you feel the same way as Jon. She really does have a point. And, of course, the ethnicity of the two people at the desk and the suspect involved in the shooting they are talking about feel relevant. But the part that irked me is that I had just come off this fresh week of slogging through these videos. And I knew the shooting Jon was talking about. And I knew there was a video. In fact, I knew there were many bodycams rolling on that incident.
Jon’s lead in to the clip was this: “…Jayland Walker who was killed by police after fleeing a traffic stop. Watch her push back when the host seems a little too eager to accept the police line.”
And then in the clip the public defender says “…too often the media reports on what are police stories as though it is the ultimate truth…”
What is everyone doing? There. Is. Video. Here it is:
I’m not sure what you think of the video. For me, it falls in the, “Okay fine, it’s legally justified… but that sure does seems like a lot of extra bullets” category. But yes, he had a gun. Yes, he fired it at police after leading them on a car chase. And yes, the police officers seemed eager to unload their weapons when they were given the legal permission.
I understand that these videos are very graphic. But that’s the state of things in America right now… very graphic. Did Jon Oliver’s writing team watch the video of shooting? Did Robby Soave? Did the public defender? How come they all seemed to act as if it didn’t exist?
I remember the initial push for bodycams and calls for transparency from police use of force as soon as small cheap wearable cameras were invented. But now that the videos are available, I’m not so sure anyone is watching. Maybe they don’t like that the shootings are never simple. The videos aren’t clearly about any one thing in particular: not race, not class, not abuse of power, not corrupt police, not poorly trained police, not well trained police. They are about all of those things in various degrees depending on the specifics of the incident. But the media does not do nuance. The media is not interested in why police shoot and kill people. They love this stuff.
Final thought #10: “Ugh.”
I really haven’t gone too much into specific police reforms and how they are playing out. Clearly, there is a messy effort underway into establishing mental health crisis responders. That sounds like a good place to start. But I want to give some air to the thought that this entire situation is next to impossible.
Watching the 40 (yes 40) videos which are linked in this essay leaves me with the feeling that police officers need an immense amount of training to start to do a little better here. The policies and legal boundaries which constitute a “justified” shooting seem difficult to adjust. But let’s say that we did. Let’s say that we made it abundantly clear that a single bullet fired after the suspect had fallen to the ground would be considered unjustified… even if that bullet was fired 0.1 seconds after the first justified one. I think I am okay with endorsing a strict shift like that. I would guess that about 50% of the videos included here would immediately include some unjustified shootings. And the second, or third, or fourth bullet which was fired after the suspect was on the ground would result in the prosecution of the police officer. I’d be fascinated to see how the incidents would shift and just how reluctant the officers would be to fire a second bullet. And, of course, what that might do to shrink the aforementioned bonkers American dataset of police shootings resulting in death.
But just imagine the training that it would take to get an officer to have the mental calmness to adhere to such a policy? The lightning quick thought process would have to go like this: the suspect has a weapon and starts to charge at you… you fire one bullet… assess if the suspect is going down… if yes, then no more bullets, if “no” then fire another… repeat. That is a very tall task when you are under stress. Part of me wants to argue something like, “Well, hell that’s the job! Being a police officer is an incredibly hard thing and we demand incredibly well trained officers!” I’m okay with that. Pay them more. Train them well. But… is it even possible?
How about the infuriating part of the videos where the suspect is down and obviously incapacitated after taking a few bullets, but the officers still can’t see their hands? It’s maddening to watch the officers, guns still drawn, screaming for the mortally wounded suspect to “show their hands” because they want to come and give aid. Can that part of the policy be adjusted? Is putting handcuffs on a suspect who is bleeding out and dying really necessary? It certainly looks terrible. And it certainly sounds terrible when a talking head on a crummy internet news show describes it to a smug reporter. I did not come across a single video which showed a suspect who had been shot several times and who was laying on the ground still stab or shoot an officer when aid was attempted to be delivered. If anyone has a video like that, feel free to pass it along. But I don’t doubt that it is possible for that to happen. Plenty of suspects in these videos still did cling to a knife or weapon with a vague sense of trying to use it on a hostage, even after taking a few bullets. That realm of “possible for the danger to the officer to still exist” seems to be all that matters here to dictate the policies of engaging officers.
How do you police a country that’s flooded with guns? Any type of policy reform that I’ve wondered about certainly increases the risks to officer safety. The videos where things turn deadly in an instant is what is driving all of this madness… and they are very real. I think in the 40 linked videos there are 2 police officers who end up dying from their injuries. It’s awful. This is a warzone. This is madness.
My dark week of thinking about violence, and police violence in particular, has left with a few concluding thoughts.
The less-lethal options need to improve a lot. There are some “lasso” type options in development such as BolaWrap. Here it is in action. I get the feeling like it has a pretty specific use case at a fairly close range. There are also robot dogs and actual dogs. But from what I’ve seen generally, the less-lethal stuff is usually pointed at the suspect at the same time as the lethal stuff. Officers sometimes deploy them at nearly the same time. There seems to be very little trust in them from police officers, and for good reason. The failure of the contact prongs of a taser to land just right and take down a suspect can create a frightened enraged armed person charging at an officer in the very next moment… with predictable lethal response from the officers.
We do need to expect more from officers. But a country like America is a difficult, perhaps impossible, situation to police ethically. Many officers are out of shape and likely not very confident that they could run away or after a fit armed suspect. I’d be interested to see studies correlating the fitness of the officer to the likelihood of discharging their weapon.
But my final overwhelming thought is that the spreadsheets and data are never going to tell us what we need to know or even help us figure out how to think about these issues… other than that the datasets are abnormally huge in America. But we already knew that. What we needed to get access to was bodycam footage. We seem to finally be getting a steady stream of them. And there are things to learn within each of them which will always remain hidden in a spreadsheet. What watching them adds up to tell me is that this job might just be impossible to do with an “acceptable” overall level of bullets flying and deaths at the hands of police officers.
This does not mean we should just give up on improving the situation. There are some videos where the bullet is “acceptable”, if not always a kind of deep tragedy. The existentialist philosopher and novelist Albert Camus may have captured this notion best when he wrote “violence is both unavoidable and unjustifiable.”
Okay, fine. But we simply have to do better with the conversation on this topic. I’ll start with this obvious one which always seems to get buried under the layers of race and historical inequities. There are just too many goddamn guns. Before any police officer would even entertain any of the legal adjustments that I am wondering about, they’d have to have more confidence that there is almost certainly not a gun hiding in the bedroom drawer when they get called into a domestic dispute. In America, you can never be sure. And worse than that, the possibility is high enough that the logic of assuming it to be true is reasonable enough. Is it too late to solve the problem of too many easily accessible guns? Probably.
Let’s hope for rapid innovations in less-lethal methods and some improvements in officer training and mental crisis intervention. In the meantime, I won’t hold my breath for the media to improve. If these videos prove anything, they show that when the “right” demographics align, the media will be happy to scroll past the all the “wrong” ones and steer the conversation towards the most unhelpful possible airing and light the country on fire. And even when the video does exist, they will be happy to pretend it doesn’t.
Maybe you would like to pretend they don’t exist as well. And I know these videos are extremely challenging and I hated looking at them. But I do think it was a worthwhile thing for me to do. It’s the only way to shrink that absolutely awful giant spreadsheet… which just keeps growing.