I ran across this (hat tip: Marginal Revolution) today and it reminded me of a conversation I had with the Head of IT at work.
He's not an economist, but works in a building filled with economists and financial people, and is always asking questions to see if there is something we, collectively, don't know.
He asks the questions because he rarely finds a topic that we don't know something about.
But this is a glaring example that even a nobel prize winner in economics doesn't know everything.
Permissive Action Links or PAL don't work the way that Schelling thinks they do. They are not some magic failsafe device that allows turning off all US nuclear weapons via radio, which is what he clearly states here:
U.S. weapons, for example, have "permissive action links"— a radio signal code that arms weapons but that will also automatically disarm them it if launched at an unauthorized target.
Sorry, that's not how it works. That'd be much too dangerous: imagine if an enemy were to gain access to that code, allowing them to disarm US nuclear weapons at will. Bad, bad idea. Really bad idea. There's no way you can't be sure that the code hasn't been compromised, and Basic Cryptography 101 tells you that if you can't be certain that your code isn't compromised, then that code is worthless.
PALs take the form of an interlock that doesn't allow the weapon to be used unless the proper code is input: make one mistake, and you'll need to take it back to the manufacturer for it to be replaced. You can read more here.
Schelling's basic idea is correct: if a rogue nation is making nukes, we should hope that they make them so that if a mob takes over one, they can't detonate without some sort of safeguards. And these kinds of safeguards are what make a stolen nuke from the US or the former SovUnion more or less useless as a nuclear weapon: you have to rebuild them in order to use them, and that is, mildly, a non-trivial task.
But once the weapon is armed, i.e. once the weapon's PAL has been correctly activated, you cannot disarm it. You can't disarm them with a radio code signal: you can tell bombers carrying them to return to base, but that doesn't disarm the weapons. If a missile is fired, then it will, ceteris paribus, be detonated; you can't disarm missiles with a magic code. The same is true for cruise missiles, sea-based missiles and any weapon where a human is no longer in the loop.
Which is why the US continues to have bombers and why bombers would be, in push comes to shove, probably the weapon of choice if you want to maintain positive control over the usage of nuclear weapons until very briefly before detonation.
Which makes Iran's development of missiles that much more disturbing: they apparently aren't concerned about maintaining control, but want to use them.
Or that they really don't understand the world that they appear to want to enter.
Which makes making sure that they don't get them that much more important.
And if anything, we should be sure that India and Pakistan use PALs.
And there is a reason for countries like Iran not to have nuclear weapons: they are irresponsible. For them, nukes are the key to being able to attack Israel with impunity: the Israeli deterrence, even though officially non-existent, does work. Remove that and all bets are off.
And given Iranian rhetoric, don't think they wouldn't be used. For them, it's the whole point of developing them: to destroy Israel first and to prevent the US from retaliating second.
And with nuclear weapons comes great responsibility. That's why none have been used since WW2. That's why the cold war never turned hot. Once you have nuclear weapons, you become a rational player, if and only if you are responsible enough to understand that they can't be used. But give them to a madman, or a religious fanatic desperate for the tool for destruction of a real or imagined enemy, i.e. someone who does not play according to the rules of the game, and all bets are off.