Airline security is a problem. Anyone who has to cope with the system is beginning to think the cure is worse than the disease.

That’s because it’s perimeter security. All of the efforts are put to keeping bad actors out of the system. You have the outside, where anything might happen; you have the inside, which is (in theory) sanitary and safe; and you have the perimeter, which is designed to be permeable to good guys and impermeable to bad ones.

Perimeter security is lousy security. That’s true no matter how good the people manning the perimeter are, and no matter how good the gear used to maintain the perimeter is. It doesn’t help that some of the definitions of “bad guys” are silly, but that’s an item. No matter how hard you work, you cannot make the perimeter barrier infallible.

Part of that is because the bad guys want to get in, and the perimeter is a handy target. Persistent probing will find the weak spot(s), and let the predators get in and feast on the goodies inside. The better the perimeter, the more “sanitary” the interior is, the more attractive piercing the barrier becomes and the more effort the predators will consider worthwhile.

The other problem is that perimeter security is necessarily a hassle for the people who seek to enter legitimately, and the tighter and more effective the perimeter is the worse that problem gets. This inevitably leads to legitimate entrants trying to defeat the perimeter in the interest of ease of entry. The “good guys” get in league with the “bad guys” looking for holes in the perimeter, and add to the forces trying to pierce it.

Perimeter security is necessary, but not sufficient. It only works if you also provide interior security. There will always be breaks in the perimeter, and you need a backup, something that will deal with the intruders. If the perimeter is of any use at all the breaks will be relatively small and transient, but that doesn’t make interior security easier, it makes it harder. It is very difficult for anyone to be alert all the time when nothing is happening, and a perimeter break that encounters lax interior security because of boredom is likely to be successful.

The question is, how to provide interior security for air travel. The only possible providers of interior security are the passengers; the flight crew are busy with their tasks and can’t take time to pay attention, and there aren’t enough of them anyway. Hiring people specifically to be interior security — “air marshals” — won’t work. An airline is a very low margin business, and the air marshal not only doesn’t pay for his seat, he expects to be paid for his time. The airline company can’t afford very many such people on the planes, which means that the likelihood of a responder being available when needed is low.

Generally arming the passengers is an attractive idea to libertarians. One wag suggested that the gate agent should ask passengers if they’re armed, and if not, offer them the choice of revolver or automatic. It wouldn’t work in practice. There really aren’t all that many people around who are actually competent in use of violence and weapons.

So: In order to provide effective interior security, we need to increase the number of passengers who are prepared to respond properly, but we can’t afford to hire security specialists to perform that role.

Solution: Reserve Deputy Air Marshals.

The model is the Concealed Carry Firearms License. In every State where they’re available, getting one involves taking a course or otherwise establishing that the applicant knows the law about using the weapon and has some idea how to decide on appropriate uses. It is a very successful concept. The typical CCFL holder is armed to the teeth, usually with a handgun that’s moderately to very powerful, but CCFL holders commit a vanishingly small fraction, nearly none, of firearms-related crimes.

Air Marshals take a course that acquaints them with the unique characteristics of operating on airplanes, the relevant laws, and correct procedures for evaluating situations and deciding on responses. Open that course to the general public — and not free, either; make applicants pay a fee that at least covers the expenses. That wouldn’t be cheap.

Why would anybody pay? Well, a Deputy Marshall would be trustworthy, having passed a reasonably stringent background check and been cleared by the instructors at the school as having the correct temperament. There would be no need to stop such people at the perimeter checkpoint; they need only show identification, get it verified (a computer could do that very quickly, like authorizing a credit card) and sail on past. No magnetometer, no taking shoes off, no search of carry-on luggage.

Road Warriors would go for it like a pack of rats. Not having to spend hours in line and subject onesself to humiliating searches and penetrating questions from the kind of brain-dead “security” people who arrested Michael Yon would be worth a lot to the kind of person who travels often. As a bonus, the typical Road Warrior is a person of vigorous constitution and disposition, and generally above average intelligence — exactly the sort of person who would make an ideal security guard if they could be recruited, but they can’t be recruited because of the pay and working conditions.

The result would be that a bad actor who penetrated the perimeter would face further examination in the interior, making it more difficult for them to act, and there would almost always be attentive people who realized that it was their job to react when necessary on the plane. It would make the job of a hijacker nearly impossible, and even mad bombers would face difficulties — I refuse to believe that anyone can walk around with a bomb in his underwear and not behave at least a little suspiciously. Deputy Air Marshals would also be trained to observe, so that if an incident did occur and damage was done, there would be some likelihood of catching any accomplices or confederates.

This isn’t randomly arming passengers, which I agree would be likely to produce bad effects. This is selectively arming precisely those passengers who are most likely to be able to do the job well and have demonstrated commitment by volunteering to do it and pay for the privilege. Your typical CCFL holder is notably better at both use of firearms and detecting whether or not their use is appropriate than the median policeman, and this selects the cream of the CCFL crop, so there is no reason to suppose they would be inferior to the directly-hired marshals.

Hey, maybe we could cut the wait at the perimeter to half an hour or so for the rest of the passengers… well, a fellow can dream.