Joshua PollackThe Logic of a Silo Conversion Ban

By now, New START aficionados (if that’s the right word) will know all the talking points about missile defenses. Treaty critics in search of unacceptable restrictions on interceptors have homed in on Art. V, Paragraph 3, which prohibits either the conversion of ICBM silos or submarine missile tubes into interceptor launchers, or conversion the other way around. Administration officials and military leaders have responded that there were never plans to conduct such conversions, and it’s more cost-effective and technically feasible just to dig new interceptor silos if necessary. Fair enough. But why is the conversion ban in the Treaty text? What’s the point of having it there?

Back in April, Jeffrey pointed out one good reason: putting new interceptors into dual-purpose silos probably would have brought the interceptors into the scope of the New START inspection regime. And earlier this week, Ivan Oelrich similarly observed that, absent the two-way conversion ban, enforcing the Treaty’s limits on ICBM or SLBM launchers (i.e., silos or tubes) would have become more complicated. It probably would require more inspections to verify that interceptor silos were not, in fact, treaty-limited ICBM silos.

The apparent point, then, is to keep interceptors out of the Treaty’s numerical limits and inspections regime. That seems like a straightforward and perfectly adequate explanation for the conversion ban. So the additional explanation from STRATCOM commander Gen. Kevin Chilton, given at the June 16 SFRC hearing, is that much more interesting.

The World’s Biggest Game of Russian Roulette

First off, Chilton endorsed the view — voiced by Sen. Dick Lugar — that dual-purpose silos would have to count against the ICBM or SLBM launcher limit.  He considered this disadvantageous; why trade SLBMs for interceptors if you can have both instead? He also made the familiar point about cost and technical disincentives for conversion. But then the STRATCOM commander offered an altogether different explanation about why the two-way conversion ban is an all-around good idea:

From an ICBM field perspective… there would be some issues that would be raised if you were to launch a missile defense asset from an ICBM field, with regard to the opposite side seeing a missile come off and wondering, well, was that a missile defense — a defensive missile — or is that an offensive missile?

In another context, this issue has been called “the ambiguity problem.” If the Russians don’t know whether the missiles flying at them are defensive interceptors or nuclear-tipped, they might respond with a nuclear attack. As discussed in this space last year,  GMD interceptor missiles launched from North America against inbound North Korean ICBMs would fly out in the direction of Russia. In some scenarios, the intercept attempts would actually occur over Russian skies.

A Picture is Worth 1,000 Missiles

In case you missed the earlier discussion, here are the illustrations again. Observe carefully. The two red tracks are North Korean ICBMs shown bracketing potential North American targets. (Hawaii is a fundamentally different scenario.) The blue tracks are GMD interceptors flying out from Ft. Greely, Alaska against the attacking missiles. The black fans are early warning radars, both America’s and Russia’s.

Now let’s consider the Russian view of this scenario. Note the yellow zone, which shows where the interceptors may overfly and re-enter the atmosphere, and where it overlaps with Russian territory. With two interceptors flying out against each attacking missile, absent one or more major technical failures, one to two interceptors per missile would fall to Earth inside this area.

Even in the best case, the opportunity for an unfortunate misunderstanding gives one pause. Don’t try this at home.

Now just imagine that the interceptors are flying out of the ICBM field at Minot AFB instead of Ft. Greely. Add to that the possibility that Russia’s early-warning network may not detect the North Korean launch in the first place, and this scenario could spoil your entire day.

Perhaps the concern is merely hypothetical, since we’re not going to use GMD, barring some horrible miscalculation. As Kim Jong Il once remarked, “It is ridiculous to claim that North Korea will be able to beat the U.S. by developing intercontinental ballistic missiles and blasting them off to the U.S.” Nevertheless, in the run-up to both North Korean ICBM tests, back in 2006 and 2009, there was a clamor to shoot that missile down! There will probably be additional ICBM tests and further calls for interception. Some future President might yield to that sentiment. If so, I sure hope that she remembers to tell the Russians in advance.

In short, the two-way conversion ban is there for good reasons. Missile-defense advocates, certainly, should be only too glad to embrace a provision that keeps defenses out of the Treaty limits, keeps inspectors away from interceptors, and reduces the chance that an intercept scenario will escalate into WWIII — a margin of safety that might make a future leader more willing to use GMD, in theory. The Russians may have raised the idea, as they’ve apparently expressed concerns about silo conversion potential before. Or the Americans may have raised it, for all of the reasons given above. Either way, both sides have sound reasons to want this provision in the Treaty.


  1. anon (History)

    You’ve answered a really interesting question — “Why would the United States NOT want to put missile defense interceptors in ICBM silos?” And the answer is because launching them could inadvertantly lead to World War III.

    But that doesn’t answer the question that Senator McCain, and others have asked. Even if its true that we’d never want to do this, WHY is the limit in the treaty? Why did we include a ban that it is meaningless, as far as affecting our plans or operations?

    The ban does make it possible to deny Russia access to interceptor silos in the monitoring and verification regime. And that includes the silos at Ft. Greeley (because it is a two-way ban, and, absent the ban, they’d presumably want to check out the interceptor silos to make sure they don’t hold ICBMs). But there’s a far more simple explanation for why we agreed to ban something we never wanted to do. Russia doesn’t like the fact that we converted 5 silos at Vandenberg. They either wanted those to count under New START (a bad precedent that could spread to Ft. Greeley silos), or they wanted us to convert them back. They harped on this a lot in the negotiations. So, we made a trade. They agreed to leave those five silos alone if we agreed we’d never convert anymore of them. We gave up something we didn’t want to do in the first place to shut them up about something we didn’t want to undo. We won on both counts. The benefits to us were very high (keeping the 5) and the costs were very low. For the Russians, the costs were high (they didn’t get access to any interceptor silos) and the benefits were very low.

    So, bottom line, the critics argue that this is a concession to the Russians and proved that we were poor negotiators. In reality, we refused to concede what they really wanted, and gave them something they didn’t ask for and we didn’t mind giving up. Sounds like the Russians proved to be pretty poor negotiators.

    • joshua (History)

      Thanks for sharing this fascinating account. On this basis, we might conclude that Jeff and Ivan were looking in the right direction, but had no way of guessing how events unfolded across the table.

      Still left unexplained is the exact nature of the Russian objection to the Vandenberg silo conversions.

  2. steeljawscribe (History)

    Advocates of dual use conversion will likely focus on the conversion of SLBM tubes vice land-based, as was the case in discussions for deploying KEI in converted Ohio-class SSBNs. While technically feasible (setting aside for the moment the issue of missiles primarily fueled with hypergolics and their deployment in subs), the challenges of the CONOPS associated with this form of interceptor employment (stationing, communications, target updates to interceptor, etc.)are typically overlooked by said advocates…
    w/r, SJS

  3. Allen Thomson (History)

    Remind me why we (the US) should object to some reasonable provision for letting the Russians inspect interceptors now and then.

    It’s certainly the case that GBI silos could hold a smallish ICBM, and maybe that a GBI modification, swapping burn-out velocity for payload, could be that smallish ICBM. What’s to be lost by letting Russian inspectors have access to a randomly picked interceptor in its silo a few times a year?

    And, BTW, DARPA is looking at a hypersonic payload that would be boosted by an SM-3 Block II of the sort the US contemplates having in Poland and Romania at the end of the decade. ( Questionable that it will ever become real, but overall it doesn’t seem unreasonable to accommodate concerns about such matters by agreeing to inspection.

    • joshua (History)

      An entirely fair question, Allen, and if you want to be very precise about it, the Treaty actually provides for two exhibitions of the five converted silos at Vandenberg AFB, so the Americans can show the Russians the features that distinguish converted silos from ICBM silos, and then give them a chance later on to ascertain (among other things) that the silos haven’t been changed back. (See the Seventh Agreed Statement in the Protocol.)

      If we can accept Anon’s explanation above, it would seem that the Russians find silos that change back and forth to be very bothersome. So what’s gained most of all from the conversion ban is a clean demarcation of the two launcher types — ICBM launchers controlled by the Treaty, GBI launchers not controlled by it. That ought to make both sides happy.

      For everyone’s reference, I will post Art. V Para. 3, the Seventh Agreed Statement, and the relevant excerpt of the June 16 SFRC hearings lower down in the comments.

    • A former inspector (History)

      If you want to see why letting the Russians take a look at a GBI every year in a converted silo, just check out slide 6 in this Power Point on the official Russian MOD website:

      The title of the slide is: The GBI ABM practically does not differ in size and shape from an ICBM.

      If you thought getting the Russians to agree to a lot of things at JCIC was tough, wait for the first time they saw a GBI in the hole.

    • Scott Monje (History)

      Former inspector,

      Thanks for pointing out that Power Point presentation. I notice that a couple of slides later it details the proposals that the Russians say Gates and Rice offered orally in October 2007 and then left out of the formal document presented in November. The Power Point presentation is attributed to Baluevskii as chief of the General Staff, so it must date from some time between November 2007 and June 2008, when he stepped down.

  4. George William Herbert (History)

    None of the likely threats a BMD would be defending CONUS against would benefit much from at-sea basing. The angles just aren’t there.

    Protecting Japan, Europe, or Hawaii are different stories, to some extent, but not what this system is designed to do.

  5. joshua (History)

    Here are the source materials for this post and the discussion in the comments so far.

    Art. V Para. 3 of the New START Treaty:

    Each Party shall not convert and shall not use ICBM launchers and SLBM launchers for placement of missile defense interceptors therein. Each Party further shall not convert and shall not use launchers of missile defense interceptors for placement of ICBMs and SLBMs therein. This provision shall not apply to ICBM launchers that were converted prior to signature of this Treaty for placement of missile defense interceptors therein.

    The Seventh Agreed Statement of the Protocol:

    Converted Silo Launchers of ICBMs at Vandenberg Air Force Base

    The Parties agree that, with respect to the five converted silo launchers of ICBMs located at Vandenberg Air Force Base, California, United States of America, that were converted prior to signature of the Treaty, the following provisions shall apply.

    1. In order to provide assurances that all five converted silo launchers of ICBMs remain incapable of launching ICBMs, no later than three years after entry into force of the Treaty, the United States of America shall conduct a one-time exhibition of the five converted silo launchers of ICBMs. The purpose of such an exhibition shall be to demonstrate that these launchers are no longer capable of launching ICBMs as well as to determine the features that distinguish a converted silo launcher of ICBMs from a silo launcher of ICBMs that has not been converted. On the basis of the results of such an exhibition, the distinguishing features shall be recorded in the inspection activity report. Such an exhibition shall be conducted, as applicable, in accordance with the procedures for an exhibition specified in the Annex on Inspection Activities to this Protocol, as provided for in Section VIII of Part Five of this Protocol.

    2. After the initial exhibition of all five converted silo launchers of ICBMs, the United States of America shall conduct one additional exhibition of the five converted silo launchers of ICBMs to demonstrate that the previously recorded distinguishing features remain unchanged, that the launchers have not been reconverted, and that the launchers remain incapable of launching ICBMs. Such an exhibition shall be conducted at an agreed time, but no later than 30 days after the receipt of a request thereof from the Russian Federation.

    From the Q&A of the June 16, 2010 SFRC hearings:

    SEN. LUGAR: I’d like to raise my final question. Some suggest we should place interceptor technology into existing silos, and perhaps even on existing strategic submarines, which New START would not permit. General Chilton and General O’Reilly, in the absence of New START, we could take Minuteman ICBMs and Trident SLBMs out of their launchers and replace them with interceptor missiles. If we did, wouldn’t we reduce our missiles, launchers and deployed nuclear warheads further below even New START limits on strategic offensive arms? And what are the dangers of placing interceptor missiles in ICBM fields or co-mingling them with our strategic offensive forces generally? General Chilton and then General O’Reilly.

    GEN. CHILTON: Sir, I would not support going down in those — in either of those directions for a couple of reasons. One, the missile — (inaudible) — that we have are valuable in the sense that they provide the strategic deterrent. And I think the value of the nuclear deterrent far — per missile — far outweighs the value of a single missile defense interceptor. So I would not want to trade Trident D5 and how powerful it is in its ability to deter for a single missile defense interceptor.

    From an ICBM field perspective — and General O’Reilly, I know — can address the cost and technical piece of this — but there would be some issues that would be raised if you were to launch a missile defense asset from an ICBM field, with regard to the opposite side seeing a missile come off and wondering, well, was that a missile defense — a defensive missile — or is that an offensive missile? So just in my opinion, I don’t see the — that either of those two options would be particularly beneficial, as you laid them out.

    SEN. LUGAR: General O’Reilly.

    GEN. O’REILLY: Sir, from a technical basis, and being responsible for the development of our missile defenses, I would say that either one of those approaches of replacing ICBMs with ground-based interceptors or adapting the submarine launched ballistic missiles to be an interceptor, would be — would actually be a setback — a major setback — to the development of our missile defenses. One, because of the extensive amount of funding required in resources to redesign both the fire control system, the communications system, but especially the interceptors. They’re of completely different size and completely different functionality — different fuels — so they are incompatible — our interceptors are — with submarines.

    And also, the submarine launched ballistic missiles have a — have a launch environment which is significantly different than what our interceptors have today. And the front end — the most critical part of our interceptors — would have to be completely redesigned in order to withstand the shocks in the other launch environments. So, in both cases, there would have to be an extensive redesign of our systems. And some of the basic fundamental engineering that we’ve been doing over the past decade would have to be redone in order to adapt them, for either one of those applications.

    SEN. LUGAR: Thank you.

    MR. MILLER: Senator Lugar, could I just add very briefly that — a couple of quick points. First, is that the possibility of deploying an interceptor in a submarine or the possibility of alternative deployments of interceptors on land, were both looked at extensively prior to us agreeing to this position in the New START Treaty, including studies going back several years, in terms of operational effectiveness and cost. And the conclusions that were reached then informed the negotiations and our willingness to accept this provision. And second is that the senior leadership of DOD was well appraised of these and comfortable with these — with Article 5, Section 3.

    SEN. LUGAR: Thank you, Sir.

Pin It on Pinterest