It's Verification, Stupid!

The old START is about to expire, and new START negotiations are entering their final stage.  While negotiations are unlikely to be finished by the self-imposed December 5th deadline when the old treaty expires, they will likely be completed by the end of the month (or at worst early next year).  In the interim, there will likely be some sort of bridging agreement (though some, such as Pavel Podvig, are less optimistic).

So in the final days and hours of negotiations, what’s the biggest hold-up?  Is it the number of operationally deployed warheads? The number of strategic missile launchers? Restrictions on missile defense?  How to restrict conventional capabilities? It appears that the answer is none of the above.  Instead, the biggest obstacle remaining is verification.

Last month, Linton Brooks (correctly) identified verification as the most difficult part of negotiations based on his experience negotiating START in the early 90s.  And now, the AFP reports:

[T]alks have reportedly hit a snag over the monitor missions to Russia.
Moscow wants to jettison any controls of its missile production under a new treaty, while Washington says monitoring is needed to ensure Russia complies with limits on the number of its nuclear-capable missiles.

Fortunately, both sides appear willing to find a compromise that replaces the inspections at the Votkinsk factory with some other form of verification.  According to Nicholas Kralev in the Washington Times:

The monitoring facility's present mandate ends with START's expiration Saturday, and the Obama administration has decided not to seek another agreement allowing Americans to remain, administration and congressional officials said.
"U.S. and Russian officials signed on Oct. 20 a series of documents, which establish the procedures to be followed for the completion of U.S. monitoring activities at the Russian ICBM production facility at Votkinsk," a State Department official said.

Kralev continues:

Obama administration officials said they take Moscow's missile program seriously, and they are negotiating other verification measures to replace the permanent U.S. presence at Votkinsk. They declined to be more specific.
Mr. Kimball said that other ways to find out how many missiles Russia produces include regular inspections, data exchanges and intelligence gathering, such as tracking missile test flights.
"How significant [the loss of Votkinsk] is depends on what other monitoring mechanisms will be worked out," he said.

It's too early to tell whether the new treaty will include an adequate verification replacement, but it's not too early to figure out the key stumbling block of future arms control agreements:  It's verification, stupid!

In negotiating a new START treaty, the U.S. and Russia aren't breaking a lot of new ground.  The original treaty included limits on operationally deployed warheads and nuclear missile launchers.  While the new treay pushes the numbers down a little further, it will do so using a lot of the same verification techniques that we've been using for almost 20 years:  inspections of nuclear launchers, data exchanges, tracking missiles, etc.  In fact, one of the main arguments for negotiating a new treaty instead of extending the old one is the need to simplify verification procedures.

So if negotiations to simplify verification are causing problems, imagine how difficult negotiations over new verification methods will be.  Linton Brooks raised this concern on a recent Hudson Institute panel called "The Future of U.S.- Russia Arms Control Efforts."

Brooks pointed out that arms control advocates, including many in the current administration, see a new START as a down payment on the treaty they really want.  In the so-called treaty after next, the administration will try to negotiate a bilateral treaty that brings total warheads on each side to around 1000.

On the surface, it seems like there could be agreement.  The U.S. wants limits on Russian non-strategic nuclear weapons, while Russia wants limits on U.S. warheads to limit potential uploading.

The problem is that we currently have no treaty that limits the total number of nuclear warheads in the nuclear stockpile.  Instead,  START only counts operationally deployed warheads.  And, according to Brooks, neither the U.S. nor Russia has any idea how to verify limits on total warheads and, in fact, don't even have good conceptual models.  The technology isn't there, neither side is prepared for the intrusive inspections that are required, and there's no way to verify the numbers that each side starts with.

The practical problems with current negotiations and theoretical problems with future negotiations make it clear that verification will be the key stumbling block for future arms control agreements.  This has led Brooks (and many others) to conclude that if the U.S. is serious about arms control, it should focus its financial resources and intellectual energy on generating new and creative ways to solve the verification problem.