Using Science to Take a Look at Police Violence

With the current climate in our country, regardless of what side of the debate that you are on, it is very clear that something incredibly awful is happening. Almost every day we are hearing about another attack either on or by officers. Innocent lives are being lost in what is turning into a war that no one seems to want. The whole thing is dripping with racism, hate, fear, and very little comprehension of how both sides seem to be making the whole thing worse.

But I’m not here to write about the racism or hate. Not because they aren’t important (in fact they are the foundation for much larger areas of causality here), but because Racism isn’t something that can so easily be boiled down into numbers. Racism is not binary. Actions are. So bear with me as we take a look at dismantling two of the biggest arguments in favor of excessive force that are not directly attached with racism.

1. “An Officer Has a Right to Defend Him/Herself”

Now this is a big one. And on the surface, it is very hard to argue without sounding like I do not think an officer’s life matters. It does. I know it does. So just hear me out. To wean us into this topic, let’s first start with a little talk on Signal Detection Theory [1].

In Signal Detection Theory, there are two main variables: a “signal” and a “response.” To best understand the difference, let’s take a look at airport security. Imagine you are walking through a metal detector (The old kind, not the new fancy scanners that remind the TSA agents that I have, in fact, not been training for my marathon). The detector can be set on a range of sensitivity, meaning it could pick up everything as small as a foil gum wrapper if it is too high, or miss a block of lead if it is too low. So how does the airport decide where to draw the line? Well they come up with what is called a criterion. They try to find an acceptable middle ground in which most of what they are trying to catch is caught, and most of what they don’t care about is let through. However, this still will make mistakes on both ends because the detector cannot be specific enough. This means that every single time someone walks through a detector, one of four scenarios plays out:

A) Nothing in pockets (no signal); Nothing detected (no response) = Correct Rejection
B) Nothing in pockets (no signal); Threat detected (yes response) = False Alarm
C) Something in pocket (yes signal); Nothing detected (no response) = Miss
D) Something in pocket (yes signal); Threat detected (yes response) = Hit

We want ‘A’ and ‘D’, but need to limit ‘B’ and ‘C’. In any case in which such a criterion must be made, there has to be a decision about what is worse. Are we more worried about someone having to remove the gum from their pocket and try again (False Alarm), or a gun accidentally making it onto a plane (Miss)? In the case of airport security, we have a very low criterion because having 10,000 people empty their pockets again is worth it to prevent even a single gun on a plane. Easy.

Now let’s convert this to the topic at hand. Police interactions. For this, we need to consider every single time an officer interacts with a citizen. Granted, this number seems impossibly high, but it’s okay, since the VAST majority of them fall into Correct Rejections, and are therefore not useful to determining if a criterion is being met, remember, that is the job of the misses and false alarms.

First, we need to set a criterion. And, I am not going to lie, this is going to be hard, and may seem insensitive, but for the sake of getting to the bottom of this question, this is a must. To determine the criterion, we need to answer the question, “What is worth more, an innocent life, or the life of an officer?” Anyone want to tackle this one for me? No? Okay, well I’ll give my take on it. I say the innocent life, by a hair. Not because I think an officer is less important, but that it is part of the job to risk their life (and most do day in and day out) for the innocent lives around them. If you don’t like my answer, try to answer this impossible series of questions:

Situation 1: Two men stand in front of a police officer, and the officer knows BOTH plan on shooting him if he doesn’t act immediately. Should the officer shoot?

Situation 2: Two men stand in front of a police officer, and the officer knows NEITHER plan on shooting him at all. Ever. Should the officer shoot?

Situation 3: Two men stand in front of a police officer, and the officer knows ONE man plans on shooting him if he doesn’t act immediately, but he doesn’t know which one. Should the officer shoot them both or neither?

Most would agree the officer should fire in the first, not in the second, and may feel torn in the third. If you feel this way, good, you’re a complex human being with complex human emotions. But really look at that third option. Does an officer have the right to shoot an innocent man because he wants to live?

This gets even more complex when you increase the number. What if there were ten men, NINE of which were planning on shooting the officer. Should the officer shoot them all?

If you are looking at your screen right now refusing to even consider these options, you are being clouded by emotion, and not allowing yourself to actually make the straight answer, when all else are equal, whose life is more important. The officer or the innocent person?

We may all differ on our level of a criterion. Me? I think that the minimum idea of criterion is that innocent life = that of an officer. That is, at most, an officer should only be able to endanger the public to the level in which he/she is in danger him/herself.

So how does this look when we place it in SDT? Let’s consider this in an encounter between an officer and someone else. The signal will be whether or not the “suspect” intends to kill the officer (more specifically, the signal is whatever the officer perceives as life threatening); whereas the response will be whether or not the officer uses lethal force. So we have:

A) No threat (no signal); No threat detected (no response) = Correct Rejection
B) No threat (no signal); Threat detected (yes response) = False Alarm
C) Actual threat (yes signal); No threat detected (no response) = Miss
D) Actual threat (yes signal); Threat detected (yes response) = Hit

Or more specifically:

A) Correct Rejections: 99.9999999% of all police encounters
B) False Alarm: Innocent people SHOT by officers
C) Miss: Officers murdered by any method
D) Hit: Officer successfully neutralized a threat

(Note: Because this post is about officers using their weapons in a moment of danger, only officers using their firearms are being counted for false alarms, but ALL murders of officers are counted for a miss. This also considers an event in which an officer perceived the threat, but not in time, as a miss.)

Now remember, there MUST be a criterion, and it has to be balanced; if it were too low, it would mean all hits and false alarms (anytime an officer saw a quick movement, he/she would shoot), but if it were too high, it would be all correct rejections and misses (the officer would never shoot, even if a gun was pointed at him/her).

We can’t really use 2016 data for this exercise, because it is still constantly shifting, and it is still barely halfway done, but we do have accurate data from 2015.  Unfortunately, the difference between a False Alarm and a Hit is sometimes hard to determine (i.e. was the person really a threat or not?). For this part of the exercise, I am going to go ahead and do something that we all know is not accurate, but let’s just say, that 100% of the time that someone had a deadly weapon, they would have killed the police officer (again, this is not even remotely true, and as my 2nd amendment loving friends don’t shut up about, it isn’t illegal to carry around a deadly weapon). Here are the results:

A) Correct Rejections: 99.9999999% of all police encounters                    (Some huge number)
B) False Alarm: Unarmed/Toy gun people shot by officers                         (127) [2]
C) Miss: Officers murdered by any method                                                      (41) [3]
D) Hit: Officer shot someone with a deadly item (inc. unknown)      (863) [2]

Look at ‘B’ and ‘C’. This means with the current level of criterion, Officers are 3x more likely to pull the trigger too quickly, than to miss the threat. And this is if you decide, across the board, to say that all deaths in which the person had a weapon were in fact valid threats. With this criterion, we are saying that it is okay for an officer to kill up to 3 innocent people if his/her life is on the line. And this just isn’t acceptable.

Now, remember that word I said I wasn’t going to bring up? Racism? Ok, well I lied a bit. But I promise it was for your own good, and this is important. What if we look at these numbers when it comes to white vs. minority encounters. The base rate is 3.097 (innocent lives taken per officer deaths). Now, it can be hard to tell who exactly was on the end of killing an officer, since some murderers are never found, but the FBI did a comprehensive study from 2004-2013 showing that roughly 52% of officer deaths during that time were caused by white attackers [4]. Applying that figure to 2015 (and still going with the borderline absurd assumption that all those with a weapon would have killed an officer) for encounters with white people we have a rate of  2.571, which is still unacceptable, but it grows for minorities to 3.65!

Let that sink in. Not only with the current criterion is an officer’s life worth more than 3 average innocent Americans. But the life of a person of color in America is about 50% more expendable than a white person? These numbers show the rates in which minorities are killed compared to when they are “actually” a threat. If race were not a factor, these numbers would not be so different. Maybe Racism CAN be put down into numbers.

Okay. Back to non-directly racism related points.

2. “If They Were Innocent, Why Did They [insert crime here]”

One of the most empty defenses in response to the Michael Brown case 2 years ago was that he had recently stolen items from a store. I heard this same despicable line of thought in the recent tragedy of Philando Castile when I heard someone say “Yeah, but did you hear they had weed in the car?” No. Just stop. None of these events or offenses bear any relevance. The ONLY time in which committing a crime could potentially be justified with a lethal response is in the case of a dangerous felony. Hell, most states already institute a Felony Murder Rule, which means if you are engaged in a felonious act, all lives lost in the proceeding events can be placed under your fault, including their own.

If someone is robbing a bank, or engaging in a high speed car chase, then yes, their actions may lead to their death. But we have laws (and penalties for breaking those laws) in place against smaller infractions, and NEVER should potential of death from a confused officer be one of those. Stealing a pack of cigarillos, or driving with marijuana in your car, does not make your life worth less. And it doesn’t justify anyone taking your life. Period.

Oh, and Don’t Forget the Racism

“What the hell! You promised!” I know. I know. But it really is impossible to separate this issue from racism. Not only because it is found in every stinking inch of the whole damned thing, but also because there is an even more insidious type of meta-racism being ignored in this mess. The racism of disparaging those who are suffering from racism.

All Lives Matter. Blue Lives Matter. Whatever you put in your catchy little hashtag or meme to try and draw attention to the fact that everyone matters. You don’t realize how awful what you are doing is.

Black Lives Matter is not about saying Black lives are better or more important than others. It is about drawing attention to an epidemic of American (read: white) culture that dismantles and devalues Black culture, opinion, and, most importantly, lives.

Saying “All Lives Matter” is like if you were in a swimming pool and you were holding little Timmy underwater. Then when he comes up and gasps “I need air!”, you laugh and say “Ohh, Timmy, we all need air!”, and then you shove him right back underwater.

That is what you are doing. Mainstream (again, read: White) culture is perpetuating a system of oppression and racism, which is Mathematically (see above for just one of many examples) devaluing Black lives. And then you have the audacity to mock their struggle by equating it with your own?

And stop the complaining about your precious roads being blocked. I see people post “They’re not going to win anyone over by doing that!” Really? Because you were going to call them up next week and ask them what you can do to help, but not now that I-94 is backed up? Not only is racism alive and well in America, but half of the country pretends that it doesn’t even exist. And you want those who are being oppressed, or those who are standing with them, to stay out of your way so that you can keep on believing that. Your roads are blocked because they are trying to force action. They are done caring if you agree with them, because you obviously already made up your mind that the final page in America’s racist history was written when we elected a Black man as president, and they are done asking for your permission to be treated as equal, because somewhere in your twisted logic you think that they already have equality, and they are done waiting for the next generation to hopefully come to their senses, because we are the next generation, and as a whole we are just as clueless as the last.

So the next time you complain about BLM, or dismiss the unregulated and legally accepted violence towards innocent lives, stop and ask what you’re really mad at. Try and listen with an open mind to what others are saying. Consider for just a moment, that they might have something worth being upset about.

Now, the events in Dallas, Baton Rouge, and, as of yesterday, Kansas City, are all disgusting acts of retaliation. Honest officers put their lives on the line every day, and all they ask is that we do not go out of our way to put their lives in danger. 

And although we do need to hold the system in which they live accountable for its actions, it shouldn’t come as a surprise that physically attacking the police is not the way to do that. 

Even from a preservational standpoint, this puts the police even more on alert, raises their sensitivity to threats, and will only lead to MORE False Alarms, and therefore more innocent lives.

Violence begets violence. MLK saw that half a century ago. He may not have understood Signal Detection Theory, but he knew that what is needed is compassion for each other, and that more can be done by bringing awareness and understanding than fear.

Now I know a few of you are rolling your eyes right now, thinking “Who does this white, straight, cis-gender, male think he is, quoting MLK and talking about equality?”

I have no personal skin in the game, and even I can look at this situation and know it is sickening. Shouldn’t that tell you something?

I’m bringing this to light so that maybe some of you out there can take a step back and take it at face value. I am not trying to sell you on anything. Just maybe open you up a little to what is really going on.

I don’t intend on putting guilt or shame on those who say ALM, or who do not understand why the protests are happening, but I am begging that you stop diminishing what is going on here.

Police violence is a symptom of the society we live in. And if we don’t change how we view others, and fight to break down our own irrational fears about those different than us, it will just get worse. And to be honest, I am not sure how much more it can take before it reaches a level worse than we all would have believed to be possible.

I just hope it it’s still not too late.

1: Signal Detection Theory:
2: 2015 Officer Kills:
3: 2015 Officer Deaths:
4: Black or White killers: