Before the Violence Strikes

Towards a Decentralized System of Harm Mitigation and Gun Control

Photo by ActionVance on Unsplash

Violence is a fact of life. And yet, we should continue to do what we can to reduce it. Questions around policies are debatable. But we’re not interested in policy debates here, as this is not a policy prescription. We’re laboring under the assumption that blanket policies, as such, are either generally ineffectual or come with unintended consequences. Voluntary, open source, decentralized systems are more promising. Why?

This concept, at least theoretically, is something we could start working on right now. So even if some policy were a magic bullet, it would take an eternity to pass. What follows then is a conversation starter about how we might create a system to reduce violence, particularly gun violence. It’s no panacea, but an idea worth exploring.

Controlling Guns or Reducing Violence

The very phrase “gun control” evokes linear thinking, when the problem is complicated and requires lateral thinking. We want to reduce violence, and violence reduction requires a different kind of approach.

Banning the sale of this type of gun or that is not likely to move the needle in reducing overall gun violence. We have to think of ways to keep guns out of the hands of potential perpetrators, and this means more focus on people. In other words, one of the fundamental assumptions here is that people who seek to do harm are impelled by cultivated malice, mental health problems, chemical dependency or extremist beliefs. We want to know if someone has the potential to harm innocents. Theories abound about what sows the seeds for different types of violence. We’ll leave that for study.

What we can say, though, is in many cases people — friends, family, neighbors, or colleagues — saw it coming.

In the recent Florida school shooting, there were more than a dozen tips to central authorities about Nikolas Cruz’s behavior prior to that awful day. Police are set up to respond to events after they happen. Thus, the centralized system of tips resulted in nothing helpful.

Is there an alternative system that, though not perfect, could be more helpful?

Again, we have to think about intersession at the level of the person, not just the level of the weapon.

What is it?

Imagine a simple app that would work from any platform. Call it “VIA,” or Violence Intervention App. VIA’s open source system would create a relationship among four basic classes of people: Early Warners, Gun Sellers, Persons of Concern, and Interveners*.

• Early Warners — Those who use the app anonymously to report persons of concern.
• Gun Sellers — Those who use the app system to check at the point of sale.
• Persons of Concern — Those whose patterns of behavior prompt concern in Early Warners.
• Interveners/Mental Health Professionals*— Professionals in the vicinity of Persons of Concern trained to intervene before a crisis point.

The idea here in brief is that Gun Sellers check to see if a customer has been flagged by Early Warners. If the system reports risk up to a certain threshold, the Store Owners are recommended to refuse sale of the weapon.

But as with any other business, Gun Sellers are in the business of making money. We might be able to count on the Gun Sellers to “do the right thing” without incentives, but we might not. So it will be important to align the incentives of everyone in the system.

Early Warners should find it simple and free for anyone to download the app and offer warnings about a person’s behavior.

Gun Sellers should be given a significant portion of the sale price of the weapon in cryptocurrency, or money to download.

Interveners (mental health professionals) could find ways to get to Persons of Concern before any time-bombs go off. And there needs to be incentives and processes in place for intervention.

*I admit this is the part about which I am most tentative. It might be that VIA functions without this class at first.

The theory, though, is that when all of these groups are getting what they need, overall violence will be reduced.

How it Works

Sadly, it might take another “availability cascade” to ensure that such an app got visibility necessary for mass adoption. But mass adoption will be necessary to cover the greatest number of people.

Early Warners sign on to VIA, provide identity, and agree to tell the truth, with reputational consequences for demonstrably false reports.

Early Warners answer a series of questions designed specifically to generate a risk profile. These would have to be determined by experts, but we can imagine questions such as:

• Can you list the full name, pictures, any social media profiles, or other unique identifiers for the Person of Concern?
• Has the Person of Concern made threats of violence towards you or people you know?
• Has the Person of Concern made generalized statements, whether in person or online, glorifying violence?
• Has the Person of Concern shown an unusual interest in, or fascination with, violence?
• Have you ever witnessed the Person of Concern harming animals in a context that most anyone would find strange or inappropriate?
• Has the Person of Concern engaged in any behavior that suggests he or she might be violent?
• Does the Person of Concern often become angry or belligerent when using drugs or alcohol?
• To your knowledge, does the Person of Concern associate with gangs (is in a gang)?
• Does the Person of Concern have a short fuse or quick temper?
• To your knowledge, has the Person of Concern experienced trauma or abuse in their lives?
• Does the Person of Concern behave in ways most people would consider bizarre, irrational, or obsessive?
• Has the Person of Concern demonstrated extremist views on politics, religion, or society?
• Does the Person of Concern have a history of negative interpersonal relationships?
• Does the Person of Concern have wild or unpredictable mood swings?
• Do you know other people you trust who might share your concerns?
• To what degree are you confident that they will corroborate your accounts?
• Would you be willing to invite them to use this app?

Once the questionnaire is complete, information about the Person of Concern is incorporated into a secure database. As others complete the questionnaire corroborating the accounts, the Person of Concern accumulates a risk profile.

Gun Sellers all have VIA downloaded onto their mobile and point-of-sale (POS) devices, and receive notifications if:

• A Person of Concern is using an app that makes him trackable, and is in the vicinity.
• A Person of Concern in the database attempts to purchase a weapon.
• A Person of Concern in the database attempts to purchase (or successfully purchases) a weapon within a certain radius.

Gun Sellers are encouraged and incentivized to turn away the Person of Concern, and are given language that helps them successfully deal with any issues or conflicts that might arise in turning someone way.

Predicting Violence: Bayesian 360 Profile

When multiple parties report similar things about a Person of Concern it is possible to form what we might call a Bayesian 360 profile. The “Bayesian” part is to form a statistical likelihood of some future violence based on past instances or patterns of behavior. The “360” part is borrowed from organizations who conduct performance reviews based on reports by those who work closely with the employee to get a 360 degree view of the person’s performance. Put these two parts together and you have the basis for a profiling system that should yield a profiling system that allows communities to intervene before violence strikes.

Aligning Incentives

It’s important to align the incentives between the App Creators, Gun Sellers and Early Warners so that the ecosystem works. What does a well-functioning ecosystem look like? Can such a system be tokenized?

1. Early warners are incentivized to track truth and not to use the system for personal enmities.
2. Gun Sellers are incentivized to pay attention to warnings and to refuse custom to Persons of Concern.
3. Persons of Concern will encounter barriers to acquiring weapons.
4. Persons of Concern might get the help they need before they reach a crisis point.

This system is not likely to be fool proof or perfect. But it could reduce violence along a number of dimensions — including those that don’t make sensational media appearances, such as domestic violence, revenge killings, and even suicide.

The important thing to remember is this: If this is even viable, it cannot and won’t be a perfect system. It it a mechanism for buying time and finding other ways.

Michael Murphy Art Installation

Potential Problems and Unintended Consequences

There could no doubt be perverse consequences if such a system were to become widely adopted and operational. We might as well set out some potential objections.

1. Black Markets — Such a system could push Persons of Concern onto black markets. (Or expand the market for 3-D printed guns.)
2. Panopticon — Authorities could use the information to monitor and track citizens in a way that could violate due process, or introduce privacy or civil liberties concerns.
3. False Claims — Without the correct level of checks and accountability, people might use the platform to initiate false claims, which could damage the reputation of innocent or undeserving parties.
4. Adoption — The system might not grow to be widely adopted so won’t work very well.
5. Burden of Proof — The system violates the spirit of the Innocence Presumption, that is “innocent until proven guilty,” which is a cornerstone of the U.S. Republic.
6. Illiberal Countries — It’s not clear that this would work in every country. Indeed, if there is no presumption of innocence or vestige of protection from unreasonable searches, so this technology could be ripe for abuse.
7. Intervention — It might be that we are overestimating the standards, processes, and abilities of trained professionals to intervene with different Persons of Concern.
8. Subjective Takes — One person’s melancholic goth kid is another person’s ticking time bomb. It will be important to tell the difference.
9. Funding — the app doesn’t generate enough revenue to sustain the system and incentivize all the parties.
10. Right to Be Forgotten — Europeans have laws that allow people to erase damaging claims or records after a prescribed period. Do Persons of Concern deserve this right after a time? Does one’s appearance in such a database permanently mar one’s reputation, or ability to find employment?

I won’t attempt to refute the above here. Instead, we understand that all are concerns anyone should consider — and seek to mitigate — before building such a system.

The Marketing Power of Mass Shootings

To repeat, it would take a critical mass of users to ensure that this system worked well. “Availability cascades” — sensational, viral news events — mean that people can become hysterical en masse. At least these phenomena have the ability to spread information fast. In the sad event that another mass shooting transpires, well-placed PR/mentions could be enough to get this system into the hands of enough conscientious people to make a difference.

The Enemy of the Good?

I must admit I have had some concerns about publishing this concept at all. It’s seems sort of obvious on the one hand given the current state of technology and tokens. On the other it might just be too unwieldy a system to develop and incentivize — that is, too many points of failure to be viable.

Source: Minority Report

Then the early-warning issues that call to mind the pre-cogs of Minority Report, and of course there is the “Black Mirror” weirdness of keeping tabs on each other. In some respects one would rather live in a world with more violence than a world without privacy or civil rights. But maybe there are smart people out there who can balance all of these concerns and develop a system that reduces violence and protects privacy and civil rights.

Dans ses écrits, un sage Italien
Dit que le mieux est l’ennemi du bien.

It could be that this concept, as a billion others, finds its way into the graveyard of unborn concepts. And that’s okay. Conception starts with a conversation. And innovating ways to reduce violence is a conversation worth having.

Update: Social Evolution contributor Tomasz Kaye suggests that VIA could be “combined with an opt-in insurance system. You get coverage if you submit to registering in the system.” Could work in tandem with Harrington and Davies proposal for a national gun insurance program.

Max Borders is executive director of Social Evolution. Support them here.