Mario, I certainly agree with you that an Iranian withdrawal of cooperation from the IAEA should be treated as the crossing of a huge red line, perhaps triggering military action. But it does strike me that calculations of timing are relevant, both politically and substantively. Why didn’t Clinton strike in 1994? The administration relied on the fact that we had time, and struck an agreement with North Korea in the hope that the regime would fall before they actually obtained nuclear weapons. Like it or not, even if Iran kicks out the inspectors, many will argue against action on the grounds that it will take years for Iran to perfect a bomb.

If memory serves, in the aftermath of 9/11, America’s options on North Korea were severely constrained by the belief that they probably had enough fuel to make one or two rudimentary bombs. True, we were also deeply concerned by NK’s ability to take out Seoul with artillery alone. But there has been a widespread belief since 9/11 that the North Koreans, even without having successfully tested a bomb, might nonetheless have one or two rudimentary devices. And this belief has constrained the calculations of others.

In that sense, it seems to me that time does matter, as does the possibility that Iran might be able to make even a bomb or two from what they produce, prior to the achievement of true “industrial” status. All Iran needs is to make the world think that it might have one or two rudimentary bombs, and they will be treated as a de facto nuclear power, just as North Korea is now. That knowledge will give Iran power over many nations. For other countries, it will jump-start rival nuclear programs. True, actual nuclear testing does matter, and does push calculations to a new level. But you don’t have to go all the way to make your rivals worry that you just might have a working device or two.

Of course, all of this in a sense validates your basic point. It’s another way of saying that, once Iran ramps up production and pulls the veil of secrecy, they have to be treated as a de facto nuclear power. But it’s important to spell out some of the reasons why you don’t have to wait for a full-scale nuclear production line and testing regime for there to be a serious problem. And again, that is true for substantive reasons, and also because there are plenty of folks out there who will resist your own red line, even though the Korean case pretty much proves you right.