The September/October issue of Foreign Affairs published the exchange of letters on the U.S. Nuclear Primacy article by Keir Lieber and Daryl Press. The exchange includes my letter, which is a very short version of the response that I posted here in March.

The exchange is very interesting and I encourage everyone to read it. I thought that Keir and Daryl handled the letters from Peter Flory (Assistant Secretary of Defense) and Keith Payne very well, arguing quite convincingly that even if the United States does not intentionally give its weapons counterforce capabilities (as Flory insisted it doesn't) or if it has never admitted in public that a preemptive first-strike has been considered a legitimate option (as Payne tried to demonstrate), it does not mean that that capability and that option do not exist. Of course, they do. In fact, in my view one of the weak points of the original article was exactly that it suggested that the drive toward genuine first-strike capability has been a deliberate U.S. policy. It's most certainly not, but this doesn't make it less real or less dangerous. Had Lieber and Press emphasized this point in their original article, it would have been a much stronger piece.

Now to their response to my letter. Well, I think people may disagree about how successful the Bulava missile program is or whether the R-39 Bark program "ended in failure". After all, the very record that I keep here shows that there have been some delays indeed. I just believe that the story is more complex and to say that the delays with Bulava (or problems with other projects) suggest deterioration of the Russian forces is to get only part of the picture and the least interesting one at that.

Then, we are in disagreement about the gap in Russian early-warning radar coverage. Keir and Daryl insist that the gap is there. They even posted a file with pictures of some missile trajectories that are supposed to support their conclusion. But I cannot replicate these results - virtually all the missiles that I "launch" from the same locations get detected by the radars in Pechora or Mishelevka. There might be a very narrow corridor if one would launch a missile from the Arctic coast of Chukotka toward the ICBM base in Irkutsk, but that's about it. I guess we just use different data. For example, I noticed that on their map the Pechora radar is moved to the North from its actual location and its fan is a bit narrower that I believe it actually is - less than 90 degrees vs. 110 degrees (see Table 1).

I would be glad to compare the notes, but I don't think it matters all that much - even if the gap between Pechora and Mishelevka is as wide as Keir and Daryl suggest (which I doubt), there are other radars - for example, a missile launched from the Pacific toward the Dombarovskiy ICBM base would be detected by the radar in Pushkino about 12 minutes before impact. Not very much, but it is as good as it ever going to get for Russia. It's not very different from detecting a launch from a submarine located off the coast of Norway, for example (see Figure 1 in my accidental launch article, for example). And it's always been that way, so the gap in the East, whether it exists or not, does not create any new danger.

Over all, I think it was a very interesting and ultimately useful discussion, which I'm sure is not over yet - International Security will be publishing responses to the full version of the artcile that appeared there.