Command and Control in North Korea: What a Nuclear Launch Might Look Like

September 15, 2017

For special access to experts and other members of the national security community, check out the new War on the Rocks membership.

A new nuclear state, in a major crisis with a conventionally superior nuclear-armed adversary, contemplates and prepares to move nuclear assets in the event it has to use them. Who controls the nuclear forces? Who decides when they might be assembled, mated to delivery vehicles, moved, and launched? Who has nominal authority to order those decisions? Who has the physical ability to implement them even without proper authorization? How experienced are the relevant units in these operations? What could go wrong?

These were the questions that bedeviled Pakistan in the 1999 Kargil War and again in the 10-month standoff with India in 2001-2002. They are the same challenges and issues that confront North Korea today.

As the mountain of dust settles after North Korea’s purported thermonuclear bomb, intermediate-range ballistic missile, and intercontinental-range ballistic missile (ICBM) tests this summer and it becomes an increasingly operational nuclear state, one of the many deadly serious challenges it faces is how it manages its nuclear forces, or what command and control arrangement it erects. These arrangements are the transmission belt that makes a state’s nuclear strategy operational — how and when nuclear weapons are managed and might actually be employed. As a nuclear weapons power, North Korea now has to think about how precisely it wants to implement its “asymmetric escalation” strategy. And so does the United States, since these arrangements have very real implications for when nuclear weapons might be used intentionally — or unintentionally — in a conflict.

Command and control architectures consist of the procedures, physical arrangements, and technical features states develop to manage their nuclear forces during peacetime and crises or war. In a classic article, Peter Feaver outlines the balance of what is known as positive and negative control over nuclear forces in emerging nuclear powers. Positive control is the set of features and procedures that enable nuclear forces to be released when the proper authority orders it; negative control refers to the features that inhibit their use otherwise. Ideally, a nuclear state will strike a balance between positive and negative command and control, such that only an order to release nuclear weapons by the proper authority will result in a launch. But we do not live in an ideal world and new nuclear states often do not have the organizations, reliable personnel, and hardened and robust command and control architecture to implement balanced positive and negative control. Instead, they often lean one way or another. States that favor negative control are designed to more likely fail-safe — if something goes wrong, the weapons won’t launch — whereas states that skew toward excessive positive control may have a tendency to fail-deadly.

Nations in more permissive security environments can afford to favor negative controls, such as technical use-control devices and assertive civil-military structures that emphasize highly centralized control, so that their systems are designed to fail-safe. Think India, and probably China. In contrast, states in severe security environments that require credible nuclear threats to deter conventional invasions are tempted to select arrangements that favor positive control and therefore fail-deadly if and when the balloon goes up and conflict erupts. This can be a deterrence-reinforcing arrangement, and is often a feature, not a bug. Think Pakistan, at least during a crisis or conflict. And probably North Korea. Here, we outline what little we know about North Korea’s command and control system based on public information, and the many questions and concerns about the challenges it faces, especially in a crisis or an actual conflict.

Peacetime versus Crisis Command and Control on Land

Even states that face existential threats can lean toward a fail-safe posture during peacetime. Pakistan, for example, with its otherwise severe security situation, has centralized command and control during peacetime. Its weapons are believed to be centrally managed and controlled, with no physical predelegation of nuclear systems to the military. In a crisis, however, that likely shifts. At some point, Pakistan’s tactical nuclear weapons systems may be flushed out and integrated with conventional forces to deter an Indian attack. It is unclear what physical inhibitions exist to prevent the units in possession of nuclear weapons from launching them if they choose. Once flushed out in a crisis or conflict, some subset of Pakistan’s nuclear forces may be designed to fail-deadly.

North Korea’s security pressures are even more acute than Pakistan’s, since Kim Jong Un already has good reason to believe that regime change is the ultimate aim of the United States and South Korea. The first question, then, is whether North Korean command and control makes a distinction between peacetime and crisis/war at all, or whether it is always on a crisis footing. At the very least, it almost certainly has various equivalent “alert levels” (References in this article to “peacetime” therefore just mean the baseline alert level, not necessarily a condition of “peace”).

Nevertheless, North Korea has explicitly signaled that Kim Jong Un, and Kim Jong Un alone, can order a nuclear strike. With nuclear tests, recent ICBM launches, and even satellite launches, North Korean propaganda has made an effort to show Kim specifically signing orders and calling for various actions. This suggests highly centralized and assertive command and control during the peacetime-equivalent, favoring negative control. This is not surprising for a leader that is highly paranoid and assertive over his military. The notion of standardized predelegation — as the United States has — would likely be a non-starter for Kim. And unlike Pakistan, which has to eventually integrate tactical nuclear weapons with its conventional defenses to deter or defeat an Indian attack, Kim’s first use of nuclear weapons would be against either more distant targets such as Guam or other regional military targets that still require “strategic” nuclear weapons on medium- or intermediate-range missiles. So, during peacetime, there is no need to involve the conventional military in North Korea’s nuclear forces. Kim can rely on a handful of trusted Strategic Rocket Forces command personnel to ensure the centralization and control of the country’s nuclear weapons.

Given the current technical state of North Korea’s nuclear weapons and delivery systems, it is possible that the land-based nuclear weapons are not even maintained in an assembled state during peacetime. This inherently generates a procedural negative control, since it would require many people — scientists, engineers, and military personnel — to coordinate the assembly and launching of nuclear weapons. There may be rudimentary use-control features in North Korean nuclear weapons — early U.S. nuclear weapons, for example, essentially employed simple physical locks to inhibit full assembly of a weapon without the proper combination. (As late as the late-1990s, Britain secured its nuclear gravity bombs with what effectively amounted to bike locks.) More sophisticated coded-control features, such as permissive action links (PALs, which are specific coded-control features designed to inhibit and delay nuclear use) are unlikely to be installed, though this is presently all speculative.

For states that rely primarily on assertive procedural negative control during peacetime, the stress on command and control intensifies acutely during a crisis, if and when nuclear weapons have to be assembled and prepared for use. This is where North Korea’s nuclear command and control system, which appears to be a black box, is likely quite rudimentary.

Imagine that a crisis breaks out and Kim orders some nuclear weapons to be assembled and moved to assigned ballistic missile operating areas. First, units that may be inexperienced with handling nuclear weapons are being tasked to do so in the midst of an intensifying crisis, under incredible psychological stress. Second, even if Kim has nominal authority to release nuclear weapons, what physically prevents these units from simply doing so without his orders now that they are fully ready? Even if units do not consciously go rogue — a risk that exists during peacetime, should weapons exist in a mated state with some predelegation — the United States and its allies would presumably apply a range of measures to disrupt North Korean communications networks with electronic warfare assets. With communications disrupted, the Korean People’s Army units tasked with nuclear operation would come under intense use-it-or-lose-it pressure without knowledge of whether Kim and the Supreme Headquarters remained intact or had been decapitated.

This brings us to the third point: The Kim regime has likely designed the wartime command and control system around fears of a decapitation strike by the United States and South Korea. How can the regime make sure the military is able to respond with nuclear weapons if Kim is killed during the onset of hostilities? Given Kim’s overarching concerns about regime change attempts, one has to assume that the system is designed to fail-deadly — making sure his units will launch nuclear weapons if the regime is believed to have been decapitated is the only logical deterrent posture. This means that the actual decision to use nuclear weapons in a crisis or conflict may rest no longer with Kim but with the commander in physical possession of the system. Thus, although North Korea may have centralized and assertive peacetime command and control, largely due to disassembled components and procedures that make it almost impossible for anyone to use nuclear weapons without Kim’s express authorization, all that may shift sharply in a crisis or conflict.

This facet of North Korea’s command and control structure has implications for how and when Kim might have to consider using land-based nuclear weapons in a potential crisis. With the stresses and concerns about maintaining communication with the Strategic Rocket Forces command — surely one of the first things an attack would try to sever — and the concerns about decapitation and preemption, when an attack starts, does one really think Kim will wait to see whether his conventional forces can repel the United States and South Korea before considering using nuclear weapons? Almost certainly not. Given the time it takes to assemble, fuel (while North Korea relies primarily on liquid fuel missiles, at least), and move nuclear forces out to their operating areas, and their possible vulnerability as this is being done, if Kim wants to truly retain assertive control over his nuclear forces, weapons would have to be launched early in a crisis. Once the artillery shells start flying, Kim probably cannot be confident that he can retain control over, and communication with, his Strategic Rocket Forces units — forcing him to worry about unauthorized use. And if he waits too long, he may lose the window of opportunity to launch nuclear weapons successfully at all if they start getting destroyed. This was the concern with Pakistan in the early 2000s, and it is even more acute with North Korea today so long as it has a relatively primitive command and control architecture.

Peacetime versus Crisis Command and Control at Sea

If all this is not terrifying enough, there is the added complication of North Korea’s submarine-launched ballistic missile (SLBM) forces, which would have to be fully assembled and ready if and when they are put out to sea. Managing nuclear weapons at sea is incredibly challenging, and even states that favor fail-safe arrangements have to make significant compromises at sea because fully assembled weapons are physically located on the submarine. Moreover, given the difficulties of mating separated warheads with delivery vehicles in launch tubes, the weapons would almost certainly be pre-mated before going out to sea. The very reason states go to sea is to have a survivable leg of their nuclear forces, allowing them to credibly threaten retaliation even if all their land-based systems are destroyed. North Korea is likely especially concerned about land-based survivability given its geography and persistent U.S. reconnaissance, so its interest in a sea-based leg is not surprising. To date, its submarine force is modest and not thought to be operational, though it is undergoing active development. The force consists of the Pukguksong-1 solid-fuel SLBM and a lone developmental Gorae-class ballistic submarine. Another submarine may already be in development as well. In August North Korea also showed off a diagram for a successor SLBM, the Pukguksong-3, which may or may not already be ready for flight-testing.

All this raises questions of how North Korea might manage this leg of its nuclear force. States have two basic choices when it comes to managing nuclear forces on submarines: a bastion model (like the Soviets), which would keep the vessels close to port until they are sent out during a crisis, or a continuous-deterrent patrol model (like the United States and United Kingdom), which would keep some submarines constantly out on deterrent patrol. At least for the foreseeable future, North Korea does not have enough ballistic submarines for a continuous deterrent patrol model. That means it has little choice but to favor a bastion model, in which the submarine would be under highly centralized, assertive control, and probably unarmed during peacetime, but flushed out likely fully-armed and ready during a crisis or early phases of a conflict. Here we should note that the lone Gorae-class vessel is not known to have ever sailed too far outside of its holding pen at the Sinpo shipyard. Another possibility, thus, is that a Pukguksong-1 would simply be launched from the Gorae from the Sinpo littoral against regional targets, without ever leaving North Korea’s territorial waters. But in a crisis or conflict, it is anyone’s guess as to where the submarine may try to venture.

Even a bastion model creates several terrifying command and control vulnerabilities. First, the submarine is vulnerable to being sunk as it exits from presumably known ports, fully loaded with nuclear weapons. Second, the submarine crew would be asked to run a deterrent patrol when it has very little experience doing so, under the stress of a crisis and in the confined and psychologically demanding space of a submarine. This could be more stressful than the situation faced by the land-based counterparts. Third, what procedural or physical negative controls exist to prevent the crew from releasing their missiles? Again, one has to presume the system is designed to fail-deadly — the SLBMs would be North Korea’s last line of defense after a decapitation attack or one that disarmed North Korea of its land-based nuclear forces. There may be nominal procedural controls involving the so-called two- or three-man-rule that would require several officers to concur before releasing nuclear weapons. Still, the physical ability to launch the weapons rests with the submarine. Fourth, we see no evidence that North Korea has invested in the communication arrays that would be necessary to reliably communicate with ballistic submarines at sea. If North Korea doubts its ability to communicate with the submarine in a crisis or conflict, there may be no physical features that inhibit the crew from releasing the Pukguksongs if they choose to do so. Imagine a terrified submarine commander flushed out to sea for the first time, being hunted by advanced and aggressive American, South Korean, and Japanese antisubmarine warfare assets, unable to communicate with Pyongyang, assuming the worst has happened. Yeah. This is what keeps us up at night.

A Question of Doctrine

While the unanswered questions about crisis-time command and control are worrisome, North Korea has already revealed quite a bit about its overall nuclear posture and strategy through its state-run media and foreign ministry officials, as we have discussed before. There remains the question of whether North Korea will choose to release some sort of a formal doctrinal document to give the world some insight into the conditions under which it would employ nuclear weapons, as, for example, India has done. Pyongyang has at least shown interest in documented pronouncements of its nuclear status; in 2012, Kim presided over a constitutional amendment at the 5th Session of the 12th Supreme People’s Assembly that identified North Korea as a “nuclear-armed state.”

A codified North Korean nuclear doctrine, however, appears to be an unlikely prospect, though it may be objectively desirable in eliminating some ambiguity around questions of command and control. After all, ambiguity, for North Korea, is a feature, not a bug. As long as U.S. and South Korean war planners can’t be confident about what would happen to North Korea’s nuclear weapons in the aftermath of a decapitation strike, releasing a doctrine that would only undermine that deterrent effect is highly unlikely. Leaving some things unsaid, thus, abets North Korea’s overall deterrent.

Second, North Korea lacks some of the incentives that led other new nuclear states to release doctrines. The most notable example is India, which released a draft doctrine in 1999, one year after its first nuclear weapons tests, and finalized its doctrine a few years later. India’s nuclear do­­ctrine was meant to reassure the world that it would manage and employ its nuclear force responsibly. The incentive to be seen as a normalized and responsible nuclear state is less salient in North Korea’s case, as its sixth nuclear test should have made clear. India, too, is somewhat of a unique case; other nuclear states, including France, Pakistan, and China, leave it to observers to piece together a coherent nuclear posture from a range of official statements, much like what North Korea is already doing. Issuing a doctrine could underwrite North Korean notions of legitimacy as a nuclear state, but at the expense of an ambiguity that pays strategic dividends. North Korea simply doesn’t care as much about nuclear legitimacy — or acceptance — as observers might think.

The observation that the North Korean command and control system may be designed to fail-deadly in a crisis or war has serious implications for policy. The most obvious of these is that if the Trump administration continues to insist that “all options are on the table” with North Korea and that preemptive war remains one of these options, it must reckon with the likely possibility that eliminating Kim alone would not eliminate the risk of inadvertent or unauthorized North Korean nuclear use. More seriously, any policy solution that does not include the complete, verifiable, and irreversible denuclearization of North Korea would leave this specter looming. And so, if the unacceptable risks of preemptive war weren’t already clear from North Korea’s increasingly robust ability to retaliate against the United States, questions about wartime escalation and post-conflict command and control should also be weighed against the advisability of a preemptive war.

This analysis raises more questions than answers, but we need to start asking them — and urgently, now that North Korea is a state with operational nuclear weapons. The questions are no longer about how to stop North Korea, but how to manage a world in which it is a nuclear weapons state.


Vipin Narang (@NarangVipin) is an associate professor of political science at the Massachusetts Institute of Technology. Ankit Panda (@nktpnd) is a senior editor at The Diplomat and an independent researcher.

Image: U.S. Air Force photo/Alex Fox Echols III

We have retired our comments section, but if you want to talk to other members of the natsec community about War on the Rocks articles, the War Hall is the place for you. Check out our membership at!