For about a generation now the police in the U.S. have followed a general practice of escalating any confrontation, with the intention of escalating faster than the opposition can respond. This has saved lives, but I think we’ve reached the point where the negative consequences of this practice are becoming highly visible in a way that’s going to force changes.

The police still patrol in ones and twos the way they always did, but starting in maybe the 1970s they quit responding to confrontations that way. At the first sign of trouble, they’d call for backup. A confrontation with one person would be met by four or six officers. A confrontation with several people would be met by a dozen officers. A confrontation with an armed individual would be met by countless officers with rifles and shotguns.

In many cases this has no doubt saved lives. Certainly it has saved the lives of police officers, but it has probably saved the lives of suspects who were persuaded to surrender in the face of overwhelming force.

The problem with this tactic is that it’s obviously inappropriate in many circumstances. It’s basically a military tactic—hit with overwhelming force—and the general public views it that way and reacts with scorn (when used against people who are obviously harmless) together with either fear (when used against people like them) or shame (when they see it used in their name).

And that last, I think, is the change that’s going to make a difference: People feel shame when they see a rapid escalation to a disproportionate use of force done in their name, and they’re seeing it more in this age of smart phones.

So far when rapid escalation leads to uses of force that seem clearly unnecessary or disproportionate, and yet police officers face few or no consequences, people have been surprised. They shouldn’t be. The whole point is that these practices are codified in law and in the procedures the police departments follow. When the police rapidly escalate a confrontation they aren’t doing anything illegal, and they aren’t violating police department policies.

And those things—laws and procedures—are what need to change. Fortunately, in a democracy, when the public decides something ought to change there is a real chance of change forthcoming.

So, what do we do instead? We can go back to what policing ought to be—attempting to deescalate confrontations, reserving escalation for cases where public safety really requires it.

There are many instances where violence is a possible result—where somebody is angry or drunk or stoned or stupid or having a psychotic break or just a really bad day. In many instances violence could be avoided with deescalation.

For a very long time—as long as there has been government, right up until recent decades—escalation had to be slow, because the communication options and the manpower available didn’t allow for rapid escalation. There tended to be one representative of the government—a sheriff or tax collector—who kept order largely through moral suasion, together with having some call on overwhelming force (in the form of a posse or a platoon of redcoats or something). Unless a situation was such that calling in the marines would be appropriate, it pretty much had to be handled through deescalation rather than escalation.

Many of the lives saved by rapid escalation are police officers’ lives. Many of the people who will be saved through deescalation are not especially sympathetic—the petty criminal, the drunk, the mental patient, the burly man who is developmentally disabled. But others are, and we’re beginning to see the losses as part of a pattern, rather than as a series of unfortunate incidents.

When a guy has a stroke at the wheel of his car, manages to stop it, but is unable to respond to a police officer’s instructions (“Show me your hands! License and registration! Get out of the vehicle!”), what is the police officer’s appropriate response? Knowing the situation, obviously calling an ambulance. But under current polices there’s far too much chance that the guy will be dragged out of his car, roughed up, and dumped in the drunk tank until brain damage from the stroke is irreversible. Or, if he doesn’t have white skin, very possibly shot instead.

Now that we’re seeing people die at the hands of the police, and now that we’re hearing the testimony that these killings are legal and are in accordance with police department policy, we may finally see some changes in the laws and the policies.

Because that, rather than feelings of anger or shame, is what will make the difference.

A shift to deescalation will probably mean that people will die who might have lived. But people are dying now. People will die either way. A shift to deescalation may mean fewer people will die. It will almost certainly mean that fewer people will be shamefully killed in my name.

I think I know why it has been so difficult to pass any sort of significant gun control legislation in the United States.

Most democracies in the world have gun control laws, which serves as an existence proof that it is possible, and yet in the US gun control has been very much a matter of x steps forward, y steps back (with what you see as the values for x and y very much depending on your political position).

You can point to “historical reasons” or “cultural reasons” for our unique situation, and you’d certainly be right, but I don’t think that serves as much of a guide to anyone who wants change (in either direction). Instead, whether you advocate or oppose gun control, think for a minute what it means to have gun control—or any kind of government control—in a democracy.

In a democracy, you’re going to end up with laws that are (to a first approximation) supported by a majority of the people. In fact, in the sort of representative democracy we have in the US, it’s very hard to get any substantive change in the law unless it’s supported by substantially more than a bare majority, because it’s so easy for a determined minority to delay or block changes.

What that says to me is that, in a democracy, changing the law has to begin with changing people’s opinions.

Trying to do things the other way around—by pushing for legislation in advance of majority support—leads to exactly what we’ve seen these past few decades:

  1. Laws that are ineffectual, because they are loaded up with compromises needed to cobble together a majority in the legislature.
  2. A stiffening of opposition to the legislation, because opponents feel their viewpoint has been ignored.

If you want to get something like this done, your best bet is to follow the model provided by Mothers Against Drunk Driving. They wanted stiffer penalties and lower blood-alcohol limits, but their efforts in the legislator were initially quite ineffective. Where they were first influential was in changing public opinion.

In the early 1960s, drunk driving was something to be laughed at. Many a comedy bit was created around taking some guy so drunk he couldn’t stand up, pouring him into his car, and sending him off weaving down the road. They were funny. Watch those same sketches now and they’re appalling.

Once public opinion was changed, changing the laws was easy—because we live in a democracy.

There’s a second reason public opinion needs to change first: The police are  unable to enforce laws that aren’t broadly supported, except through police-state tactics.

If you have broad support—not just a majority, but a general consensus across society—then it’s easy to enforce laws. Few people break them. When they are broken, witness come forward to report the crimes. When people are prosecuted, juries convict and judges impose sentences as prescribed by law.

How would you enforce a gun law that was opposed by half the population?

It would be easy enough to enforce a law against open carry of a firearm, but enforcing a law against concealed carry would require the police to stop and frisk people on the street. (Of course, this happens already in certain neighborhoods. People who live in those neighborhoods are properly outraged, as are those who believe in freedom. But let it start happening in neighborhoods full of middle-class people—neighborhoods full of voters—and the laws would get changed back very quickly indeed. At least, I hope they would.)

It would be even harder to enforce a law against owning a firearm in your home. Even in the sorts of neighborhoods where people are routinely stopped and frisked, the police are not yet so bold as to enter and search people’s homes. But without that step, illegal guns would linger for decades. Indeed, for generations. And with that step, I suspect we’d see the law changed back very quickly: Many of the same people who support gun control would still object to the police-state tactics that would be necessary to impose a ban on guns.

And, lest I be accused of arguing against a straw man, on the grounds that “no one” is arguing for a ban (just reasonable regulation), I’d like to point out that the enforcement problem is the same.

I suppose the intermediate step that gun control advocates anticipate would be laws that regulate gun ownership, but with enforcement happening only when guns are discovered incidentally: If your house is searched for some other reason, then your illegal guns will be found and their presence will be used to pile on additional penalties. I most particularly object to that scenario, just as I object to all scenarios where ordinary people are required to keep their papers in order or face harsh penalties.

My main point here, though, is that seeking changes in the laws should always be a second step. First, seek a consensus in society that things should be different. Do that and it’s easy to change the law and easy to enforce it.

I’m generally against zero-tolerance policies. I’ve read too many stories about kids expelled because of an asthma inhaler or a pocket knife forgotten in a jacket pocket (or in the trunk of a car) and accidentally brought to school. Those sorts of harmless, technical violations of the rules are exactly the sort thing that should be tolerated.

But there’s one zero-tolerance policy that I’d really like to see. Prompted by the gruesome story Occupy Oakland: second Iraq war veteran injured after police clashes, about a man beaten so badly by police that his spleen was lacerated, who was then denied medical care for 18 hours, I think we need a zero-tolerance policy for failure to provide medical care to prisoners.

Every person involved in taking or holding a prisoner—police, guards, staff, managers—should be absolutely responsible for doing everything necessary to ensure that needed care is provided.

If needed care is not provided, everyone who heard the prisoner request care, saw the prisoner in distress, or got a report that the prisoner needs or has requested care, should be fired.

There should be no exceptions.