On Monday I was asked on live TV, “The film A House of Dynamite shows missile defense just doesn’t work, but the Pentagon disagrees. Who is right?” The short answer is the movie. The full answer is, like most technical subjects, more nuanced.
When I was asked last week by Bloomberg reporter Anthony Capaccio for my thoughts about the missile defense scenario in the new film A House of Dynamite, I didn’t know about the Pentagon memo he would reference in his Bloomberg story. The Pentagon argues that the events of the film are unrealistic and that the testing record for the $63 billion Ground-based Midcourse Defense (GMD) system “tells a vastly different story” than what’s depicted. But it’s not unexpected that the Pentagon would feel it had to defend the GMD system, which has strained to provide any credible defensive capability despite decades of work. (And actually, the United States has spent more than $400 billion in 2021 dollars on missile defense over the last 70 years.)
While I would love to have a one-word answer to a provocative question, as a scientist who has followed missile defense for a long time, context and details matter. I’d like to provide them here. I can say the bottom line is that the GMD system has not been shown to work under realistic conditions. As a recent American Physical Society report that I co-authored says,
Due to its fragility to countermeasures, and the inability to expand it readily or cost-effectively, the current midcourse intercept system cannot be expected to provide a robust or reliable capability against more than the simplest attacks by a small number of relatively unsophisticated missiles within the 15-year time horizon of this report.
As it turns out, the missile attack in the film is basically the simplest and most unsophisticated attack you could construct. In the film (spoiler alert!), operators launch two GMD interceptors against an incoming ICBM (intercontinental ballistic missile), which is presumably nuclear-armed, and which is heading for the US Midwest. Both interceptors fail to stop the missile. The film’s secretary of defense is angry and surprised that something that cost tens of billions had a coin-toss probability of success. Is this realistic? Yes, it’s an entirely reasonable plot point for the GMD system to miss twice. But the wider context is important, too. This comes in two parts.
The missile attack in the film is unrealistically simple
First, in the film, the threat is a single incoming missile that seems to be unaccompanied by any of the well-known countermeasures an adversary would use to increase its odds of success. These strategies include: sending multiple nuclear warheads at once; hiding the warheads among lookalike decoys that require the system to target all of them or risk letting the warhead through; anti-simulation, where instead of having decoys look like the warhead, all objects are disguised in different configurations of size, reflectivity, dynamic motion, etc.; radar jammers; direct attacks on key sensors (though the film hints that space-based early-warning sensors may have been disabled by an attack). There are many strategies a competent and motivated adversary would use and that are available to any actor able to build a nuclear weapon and an ICBM. In the film, none of them are used, so it was artificially easy.
But the only tests of the system have been simple tests, and the record isn’t great
That leads to the second part of the answer, that the only kind of threat the GMD system has been tested against is the simplest and least likely. This is a critical weakness of the testing program. We just don’t know how well it would work in a realistic scenario because it’s never been tested in one. Any credible missile defense needs to be able to work in realistic conditions, not just the simplest to test conditions.
The Missile Defense Agency’s memo reportedly says that today’s interceptors “have displayed a 100% accuracy rate in testing for more than a decade.” (Note that there have only been three tests in the last decade.) I’ve spent a lot of time looking at the GMD’s test record over the years. Over the last 25 years, there have been 20 intercept tests, meaning tests where an interceptor is launched to destroy a target. But even so, under simplified conditions, the system failed about half the time. (There are different ways to count this; for example, a test in 2006 was considered a “hit” by the Missile Defense Agency but the Pentagon’s independent testing agency scores it as a “no kill” because the system only achieved a “glancing blow” and didn’t destroy the target. Most people would consider that a failure.) Some problems were quality control or equipment malfunction, some were design flaws. One of the interceptors in A House of Dynamite fails because of a specific malfunction that happened in three of the GMD test failures, so that is certainly realistic.
There are many reasons for the program’s failures. One is that the system was built on a political rather than technical readiness timeline, as laid out in detail in a 2016 UCS report. But fundamentally, strategic missile defense is a really hard problem. In the film, they call intercepting an ICBM in flight akin to “hitting a bullet with a bullet.” That’s because the warhead and the GMD’s “kill vehicle” are both about the size of a file cabinet and they are each flying through space at around 30 times the speed of a jet, and the kill vehicle needs to maneuver itself into a direct collision with the target. This is not an easy engineering problem, and clearly the Pentagon is still working out how to do that reliably, but it is actually not the hardest part of the problem to solve. The most challenging issue is providing defense when the adversary is taking measures to confuse, overwhelm, attack, and evade defenses using countermeasures. That is the well-known Achilles heel of the system and one that has not been addressed seriously
No tests have included countermeasures that would challenge the system and show it works in the real world. As longtime critic and former director of the Pentagon’s Department of Operational Test and Evaluation Dr. Phil Coyle used to often say, the tests have been “scripted for success.”
The GMD test program is like T-ball practice when the real world is major league baseball. It’s irresponsible to say the GMD system provides effective protection when it has not been demonstrated to work in conditions expected in a real conflict, and that doesn’t look like what we see in A House of Dynamite. So whether missile defense worked in the film isn’t really the right question.
Why is the Pentagon so sensitive about this?
Certainly, the Pentagon wants unprecedented control over all messages about its work. But there has always been an incentive to give the rosiest picture of the GMD’s system’s capabilities, and the stakes are even higher today, as one of the president’s marquee projects (beyond a golden ballroom and the undermining of US democracy and rule of law) is the Golden Dome missile defense system. The vision behind Golden Dome is a comprehensive missile defense system that would defend against all types of missiles from any adversary. Congress has already allocated $25 billion for this project and the administration wants another $150 billion, though the costs of such a fantastical system are likely to be many, many times higher than this. There is a lot on the line for those who want to sell this idea.
The Pentagon and White House officials have a long history of overstating the system’s effectiveness without good evidence to back it up. In 2003, the Pentagon’s undersecretary of defense for acquisition, technology and logistics predicted—before any interceptors were in the ground or any tests of anything besides prototypes had been conducted—“effectiveness is in the 90 percent range.” When asked about this, the then-director of the Missile Defense Agency said, “There are a lot of things that go into [determining] effectiveness. Everybody can be right.” Perhaps that was because at that point North Korea had not tested a nuclear weapon and wouldn’t test a missile able to reach the United States for almost 15 years. But this makes it clear that to talk about effectiveness, you have to be specific: effectiveness against what?
What should we do instead?
It’s important not to come away from A House of Dynamite with the idea that we’d be safe from the threat of nuclear war if we just spent a little bit more on missile defense. It’s especially critical to understand that a massive endeavor like Golden Dome is more than a waste of money—it incentivizes an arms race and is a distraction from opportunities to pursue real security. The Iron Dome (before it became Golden Dome) executive order upends decades of bipartisan policy that the United States would not use missile defense to protect the US against the Russian or Chinese arsenals, which have the size and complexity to defeat such defenses. This new policy provides incentives for an arms race that reduces security for all. China and Russia have been concerned for years about the US development of strategic missile defenses, and seem to be taking steps to offset them by expanding or diversifying their nuclear arsenals in response, for example, Russia’s development of a nuclear-armed, nuclear-powered cruise missile. And Golden Dome envisions putting space-based missile defense interceptors into orbit for the first time, which would start a dangerous new chapter.
Instead, we should be taking known, sensible steps toward reducing the dangers of nuclear weapons. One that could be taken right now is for the United States and Russia to agree to continue abiding by the limits of the New START treaty while entering into nuclear arms control and disarmament negotiations with the seriousness required of this existential threat.