Here’s a thesis statement for New Urbanists to consider: it’s hard to enjoy being a flaneur if someone is trying to mow you down with a motor vehicle, or if the cops are shooting to stop the mowing. That is, a pedestrian taking in the sights ought not to have to worry about being struck by a car or truck, or perhaps being hit by a stray bullet.

Such thoughts are summoned to mind by this Washington Post headline: “Police in D.C., New York revise shooting policies in response to vehicle ramming attacks.” As the Post‘s May 1 story details, in the wake of the spate of ramming attacks around the world, police forces are reversing their decades-old policies against shooting at dangerous moving vehicles.

Reluctantly, one might conclude that the authorities have no choice. It’s one thing to tell the police that they shouldn’t shoot at, say, getaway cars, because the collateral risks are too high; it’s another to demand they hold their fire at the moment of murderous vehicular mayhem.

One could further say that the April 23 attack in Toronto, which left 10 dead and 15 injured, is particularly scary because the alleged killer was not a jihadi. Instead, he was apparently an “incel,” which is one of those new categories of pathology that the Internet seems good at discovering, and in a perverse way at glamorizing and multiplying.

In other words, ramming has transcended the stereotype about killers as ISIS wannabes. It’s now an equal-opportunity thing, a wicked kind of performance art, a sick-yet-easy option for anyone with a sufficiently malevolent turn of mind.

So what to do? How to respond to this threat? For the moment, let’s confine ourselves to the immediate issue of safeguarding streets and walkways. As for the underlying problem of human nature, let’s wish the healers and therapists the best, even as we remember the realist wisdom of Immanuel Kant: “Out of the crooked timber of humanity, no straight thing was ever made.”

Still, as we live our lives in a bent condition, we can at least try to make straight the path to greater urban safety. Here are four possible categories of solutions.

First, applying on-the-spot restraint. This includes, of course, shooting at the driver. Without a doubt, firing at moving targets is at best a least-worst strategy, given the difficulty of aiming and the great risk of hitting innocent bystanders.

So we might ask: are there better mechanisms of immediate restraint? Could we have pop-up barriers or force fields? Traps or nets of some kind? Lasers or some sort of pulse device that turns off the vehicle? In speculating on such technological fixes, we should have confidence that there will be plenty of clever ideas for gadgets—and some might even work. Of course, we should also be mindful that any safety device that kicks into gear will likely be controlled by some sort of artificial intelligence, and, as this author has argued, that opens up new vistas of the problematic.

Second, banning cars in pedestrian-dense areas. Many cities have been moving in this direction: in the past, the rationale has been aesthetic or environmental, but nowadays safety is another valid reason. Indeed, even without ramming, motor vehicles bring trouble, killing some 6,000 pedestrians each year. So in some places, banning vehicles would also solve the ramming problem.

On the other hand, in a country of 325 million people, there are a lot of pedestrian-dense areas, and so prohibiting vehicles for the sake of safety is hardly a scalable solution. Moreover, as President Trump pointed out while speaking to the NRA convention in Dallas on Friday, “van control” would raise all the same objections as “gun control.” And, of course, cars have a lot more support than guns.

Third, accelerating the push toward driverless cars, also known as autonomous vehicles (AV). That is, if we can’t ban cars, maybe we can ban car drivers. The AV idea, actively pushed by Silicon Valley, seemingly offers the promise of eliminating wheeled human murderousness, because the computer would be doing the driving—or perhaps, in the name of safety, the emergency overriding.

Yet on the matter of safety, a string of accidents, including fatal accidents, has shaken public confidence in AV. Actually, come to think of it, maybe the public has never wanted AV. Maybe all along it’s been a top-down lobbying campaign pushed by Silicon Valley’s money and prestige.

In the wake of an AV fatality in March, Uber announced a pause in AV testing, which seemed like a fair acknowledgement of public concern. And then, just last Friday, we learned that another tech company, Waymo, owned by Google, had still been AV-ing—and was involved in another accident, this one mercifully non-fatal.

Obviously, the tech lords have gotten way ahead of themselves on AV—and more to the point, way ahead of public opinion. Yes, techsters have much invested in their AV projects, but the American people have much invested in their own lives. And as long as it’s votes—as opposed to campaign donations—that are counted on Election Day, politicians will have to heed the wariness of voters more than the eagerness of donors. So in light of these murky politics, an AV solution to ramming is nothing to count on.

Fourth, building barriers. In the last three and a half decades—ever since the 1983 truck-bombing of the Marine barracks in Beirut, Lebanon—a pattern has emerged: if a building is judged to be truly important, it is to be defended by an immobile phalanx of bollards or other kinds of obstacles, including, sigh, the distinctly unsightly jersey barrier.

Typically, these protected buildings are public structures, but some high-profile private buildings, too, have the benefit of barriers. For example, the immediate area around Trump Tower in Manhattan is a car-proof fortress, and other name-worthy private buildings—such as the nearby Bank of America Tower—have the same fixed counter-terror protection.

As this author wrote for TAC last year, bollards can be seen as “passive defense.” Passive defense is just what it sounds like: it’s just there, always on guard. The alternative is, of course, “active defense.” And while active defense might seem better because, well, it’s active, it will certainly be more expensive and might sometimes even be less effective.

After all, with active defense, there will always be technical complexity—and thus the greater risk of Murphy’s Law. By contrast, passive defense, in its low-tech simplicity, is always “on,” whether or not it’s plugged in or wired up, and that makes it about as snafu-proof as anything humans can create. (This author has sung the praises of passive defense earlier, here and here.)

To be sure, the issue of putting barriers around buildings is controversial. Back in 2006, for example, New York City began restricting the placement of barriers around buildings. After all, some of them were definitely eyesores, and probably all of them had at least some inhibiting effect on traffic and people flow.

Yet the active march toward passive defense is likely to continue—for the simple reason that, in its stark simplicity, it does the job. Moreover, with some thought, barriers could be made prettier; they could be disguised as benches, or tables, or statuary. They could even be trees—perhaps, if need be, reinforced trees.

In the meantime, another new phenomenon in the built environment, protected bicycle lanes, points the way to another kind of passive solution. That is, bike lanes with curbs that separate bicyclists from motorists are an obviously good idea for the protection of the two-wheeled. And it’s easy to see how curbed bike lanes could double as passive defense for pedestrians and buildings from the four-wheeled threat.

Of course, a lively debate about such matters is not just inevitable; it’s useful. The dialectic is the fastest route to sustainable solutions. In that vein, it would be interesting to hear more from major landlords, including federal ones like the Department of Housing and Urban Development and the General Services Administration. And it would be equally constructive to hear from artists, urbanists, activists, and the citizenry. For the sake of aesthetics as well as safety, a national conversation is needed.

Not everyone, to be sure, will like the idea of passive barriers, because there’s always someone who objects. As Voltaire said, it’s easier to conquer the universe than to get a single village to agree—and America is a mighty big village.

Still, until we get to the world where humans are made from non-crooked timber, it’s likely that a critical mass of people will agree on the need for barriers as a relatively cheap—and extremely important—urban safety measure.

James P. Pinkerton is an author and contributing editor at TAC. He served as a White House policy aide to both Presidents Ronald Reagan and George H.W. Bush.