Sunday, July 10, 2016

Moral Outrage is Appropriate, but It Doesn't Solve the Problem

Moral outrage is a perfectly human response to many situations. The problem with moral outrage is that it’s not really a solution. Declarations such as “This has to stop!” and “This is totally unacceptable!” need to be accompanied by substantive policy recommendations. How do you reduce the number of police shootings of innocent civilians?

You can’t just say “Stop making mistakes!” Any encounter between police and civilians has a non-zero chance of becoming one of these fatal escalations. So adding up the millions and millions of encounters, most of which are peaceful, will inevitably lead to a few horrific mistakes. A substantive discussion would include some policy changes that fix the problem. Notice that even a slightly more specific framing of the problem still isn’t helpful: “These shootings happen because too many police officers are racist, and their fear of black people causes them to be trigger-happy.” This may or may not be a correct diagnosis, but it raises the question of what policy tweaks can we implement to make people less racist? It’s very hard to change culture, and it’s very hard to change the nature of people. We don’t have a policy lever that turns bad people into good people. We need to tweak our institutions so that we get the best possible result out of the crooked timber we’re working with.

We could implement a policy that bluntly leads to fewer interactions between police and civilians. You *could* require stricter criteria for making a traffic stop so that there are fewer of them, but this would mean that it’s harder to make traffic stops for legitimate reasons and catching fewer real criminals. There’s a false positive/false negative trade-off. Do you want to harass fewer innocents and let more criminals off the hook, or do you want to catch more real criminals and harass more innocent people? There’s an inverse relationship between the two kinds of errors (taking everything else as a given: fixed resources, a given level of law enforcement experience, our institutions as they currently exist, etc.). I picture an ideal world where everybody understands this principle and sees the same inverse curve (with false positives on the X-axis and false negatives on the Y-axis) and the substantive argument is about *where* on that curve we want to be.

Or perhaps we can reduce both kinds of errors by throwing more resources at the problem. Some kind of sensitivity training or tactical training for police officers might work, but it needs to be established that such a training program *actually* works or the officers will resent and ridicule it. There could even be a backlash, where police officers resent the treatment so much it has the opposite of the intended effect. Not to mention it would be a waste of money and a distraction of our scarce law enforcement resources. Any high-sounding solution needs to be tested for effectiveness.

Maybe you can do better screening for problematic police officers, but this will raise the cost of hiring law enforcement personnel by limiting the pool of applicants. It may be a cost worth paying, but we should face it with our eyes wide open. Also, this once again presents the false-positive/false-negative trade-off. If you raise the bar, you will exclude more problem officers, but you will also exclude more perfectly good candidates who happen to score low on your imperfect selection criteria. This should be evaluated in cost-benefit terms *and* in terms of fairness to the individuals.

(A minor, slightly technical aside here. There is no test that tells you with certainty “This is a problem officer” and “This is a good officer.” But you could come up with a test that says, “Given this person’s background and test score, this person has a 5% (or 10% or 15%) chance of becoming a problem officer.” You can work on making the test as accurate a discriminator as possible, but you will never separate your population cleanly into 100% good and 100% bad police officers. Wherever you set the bar, you are always running the risk of turning away good officers and accepting bad ones. Both kinds of errors will always happen. The choice you face is the relative proportions of these kinds of errors.)

There are other questions. What level of government should respond to the problem? Is there even a “lever” to pull here? As in, a policy change that affects the problem? Should a higher level of government step in if a local government is intransigent in implementing a solution? Does such an intervention raise concerns about democracy and voter sovereignty? “Stop being so racist” might be good advice, but is there a policy lever to address it? (As in, “The more money and resources we throw into this machine, the less racism we have in law enforcement.”)

It reminds me of when my kids are fighting over a toy, and my lazy response is to just say, “Stop fighting, guys.” And they both still have their hands on the toy. This isn’t a solution; I have to get in there and adjudicate the dispute. I have to decide which one of them gets the toy, which means I have to break the other one’s grip. Possibly, I also need to set a policy, such as “These toys are yours, these toys are his. Don’t play with your brother’s toys without permission.” Or “The toys are collectively owned and anyone can play with any toy, so long as your brother isn’t playing with it at the moment.” Unfocused moral outrage accomplishes about as much as me telling my kids to “stop fighting.” When I hear things like, “Let’s all just get along” and “This should never happen” or “Can’t people just stop being shitty to each other?”, I think the speakers are really dodging the important questions. For some reason it’s considered cynical to talk about politics when something tragic occurs, but I think it’s necessary to have a serious discussion about which policies lead to fewer tragedies. (I am in no way backing down from my previous position, that we shouldn't react to individual tragedies.)  Policy setting *is* politics. I understand why many people want to be non-combatants in this fight over policy-setting. It does get ugly. It ruins friendships. But it’s a fight that has to happen. Unless we have that discussion (no, that heated argument) about which policy fix to implement, we won’t solve the problem. Denouncing bad behavior while bowing out of the policy fight is like telling your brats to “just stop fighting” while leaving both of them still grasping the toy. You’ve expressed your wish that a problem would go away, but haven’t really done anything about it.

With all that said, it’s possible that unfocused moral outrage actually *does* accomplish something. Police forces are surely scared to death that the wrath of society will come down upon them if they make the wrong mistake, and this surely has some effect on their level of caution. And some policy fixes have been implemented already. There are more body cameras, and maybe my perception is wrong but I’ve seen more instances of officers being punished or fired for shameful behavior on social media. Body cameras and personnel vetting are real policy fixes, and I could think of a few others. Legalizing drugs would eliminate the need for many police-citizen interactions, as I’ve said before. “Stop and frisk” policies, where the police randomly search (mostly young minority) pedestrians, could be eliminated. “Community policing” initiatives, where the police try to promote a good relationship with the public they serve, would give police officers a lot of non-violent tools for dealing with problem individuals or potential suspects. Angry protests, even if they don’t specifically ask for any of these fixes, might accomplish some of them anyway. Community leaders (like the mayor, city council, or police leadership) facing the choice, “Do *something* or face more angry traffic-snarling, commerce-choking mobs” might get their collective shit together and implement some of these solutions, which is great. “Moral outrage” is an important mode of the human experience and it can initiate real change. But it’s more constructive sometimes to shut it off and think about how to fix the problem. 

No comments:

Post a Comment