There has been a lot of prominent discussion lately (in the Washington Post and Wall Street Journal, among other places) about the size of China's nuclear arsenal, based on a study by Georgetown University professor Phillip Karber, "Strategic Implications of China's Underground Great Wall." The study considers the question of why China has built a vast network of tunnels -- often called China's "underground great wall" -- that stretch for some 3,000 miles. Karber's report suggests that the tunnels could hide as many as 3,000 nuclear weapons.
A top national security strategist during the Cold War, Karber recently led a group of his Georgetown students in a study of the underground system. The three-year study was sparked, Karber said, by the devastating 2008 earthquake in Sichuan, when some of the tunnels caved in and radiation teams were dispatched to the area, leading to speculation that the tunnels held nuclear weapons.
The existence of the underground great wall was no secret. In December 2009, a report by China's state-run CCTV mentioned the tunnels, built by the Second Artillery Corps and used mainly to shield China's nuclear strategic missiles. Under Karber's guidance, the students took an Internet-based, open-source approach to understanding the underground great wall, using Google Earth, blogs, online journals, video clips, satellite imagery and photos, and fictional TV docudramas about Chinese artillery soldiers as sources for their research. Based on that research and several controversial assumptions, Karber's report argues that China's nuclear arsenal could be thousands of warheads larger than previously believed, stirring a debate about Chinese intentions.
Karber's report, however, fails to answer key questions and leaps to unwarranted conclusions.
Perhaps the largest deficiency involves the amount of plutonium and highly enriched uranium (HEU) that China has available for weapon production. Newly available public information indicates that China has an existing military inventory of about 1.8 tons of plutonium and 16 tons of weapons-grade HEU. China stopped production of highly enriched uranium in 1987 and had cut off plutonium production by 1990. All of its military HEU and plutonium production facilities have been closed or converted to other uses, or are being decommissioned, and there are no reports of new production. China could well have the smallest military stockpile of HEU and plutonium available for weapons among the five acknowledged nuclear weapons states.
How many weapons might the Chinese fissile material inventory support? Calculating from average numbers for US and Russian warheads, which contain about 4 kilograms of plutonium in their primary stage and about 20 kilograms of highly enriched uranium in the secondary, 1.8 tons of plutonium could produce about 450 warheads; those warheads would also use about nine tons of HEU in their secondaries. The remaining seven tons of highly enriched uranium in China's stockpile might produce another 230 or so warheads (assuming 10 kilograms of HEU for the primary stage of the weapon and 20 kilograms for the secondary). So the Chinese stockpile of fissionable material would support perhaps 680 thermonuclear warheads.
To put it another way: Even if China's entire fissile material inventory were used, it would not support an arsenal of more than 1,000 warheads. In actuality, it is likely that part of China's fissile material stocks would be held in reserve for future needs. The other four of the declared nuclear-weapons states have converted half or less of their fissile material stocks into weapons. If the same percentage held for China, the upper bound on its arsenal would be around 500 warheads. In fact, Western intelligence agencies and non-governmental organizations estimate that China has approximately 200 nuclear warheads, 40 or fewer of which can reach the continental United States.
Karber's report contains 357 PowerPoint slides but only a couple of charts and a few sentences that actually relate to the much-publicized claim that China's underground great wall of tunnels hides 3,000 nuclear warheads. That claim seems to be based on questionable reasoning. One of the report's bases for this conclusion is a US intelligence projection -- made late in the 1960s -- that China would have 435 nuclear weapons by 1973. Using a constant rate of growth above that estimate, Karber extrapolates that China could have built 3,000 weapons by 2010. This analysis is, to say the least, simplistic. US intelligence estimates of China's nuclear arsenal made much later reflect significantly lower weapons levels. In a declassified document, the CIA estimates China's total stockpile at between 200 and 300 warheads in 1996. In 2006, the US Defense Intelligence Agency estimated that "China currently has more than 100 nuclear warheads."
To support its contention about the number of warheads in the Chinese arsenal, Karber's report also mentions that "PRC data in 1995 gave the figure at 2,350." It should be noted, however, that this PRC figure was based mainly on a rumor originally asserted in a Hong Kong amgazine some 15 years ago, and official government documents and professional Chinese analysts do not back such a large warhead estimate.
In addition, Karber's report argues that China could have 3,000 warheads because it has deployed "more missiles" in recent years; the country must, therefore, have "more nuclear warheads." China does, indeed, have a large number of missiles, but most are armed with conventional weapons. The People's Liberation Army possesses as many as 1,000 non-nuclear ballistic and cruise missiles, according to estimates by the Nuclear Threat Initiative.
But Karber's key argument centers on the growth of China's tunnel system; he contends that more tunnels means more nuclear warheads. If the tunnels of the underground great wall were used only for storage of nuclear warheads, and if there were reason to think the tunnels were packed wall to wall with missiles, this logic might be reasonable. But there is another, far more plausible explanation for the size of the underground great wall. According to major Chinese media and other Chinese-language publications, China's underground great wall is not just a weapons-storage depot; China has moved its land-based missiles underground to protect them from a preemptive nuclear strike.
China has no reliable air-based or sea-based nuclear forces. Since 1980, when it initiated China's nuclear modernization, the Second Artillery Corps has focused on ensuring that the country's limited number of land-based strategic missiles can survive a first strike. With the development of the Soviet (now Russian) and US satellite surveillance capabilities and the increased accuracy of nuclear weapons, China became concerned about the vulnerability of its missiles, in particular its silo-based DF-5s and its cave-based DF-4s, which need to be pulled out of tunnels and caves and launched from surface sites. These liquid-fueled missiles require up to two hours of preparation for launch. In addition, unlike the United States and Russia, China does not have an early warning system, and its missiles apparently are not in a launch-on-warning posture.
Under its announced no-first-use doctrine, the Chinese government says it would launch a retaliatory nuclear attack only after it survived a nuclear strike. To assure the survival of an adequate number of weapons for retaliation, China has just two primary options: One is to build more warheads and launchers. But to survive a US preemptive attack that could involve as many as 1,000 warheads and extremely accurate targeting, China would need a huge nuclear arsenal. So rather than greatly expand its arsenal, China has chosen the second option: It has protected its small missile force by moving it underground. Many Chinese media outlets, including the People's Liberation Army's National Defense Daily, have reported that the engineering unit of the Second Artillery Corps began the underground great wall in 1985 and finished its first phase in about 10 years. By the mid 1990s, then, China had a reliable second-strike capability.
The road-mobile solid-fueled missiles (e.g., the DF-31 and DF-31A) China deployed around 2006 are less vulnerable than fixed-based surface missiles and silo-based missiles. But the United States is pursuing capabilities -- including long-range, precise conventional strikes and the monitoring of mobile targets from space -- that could make Chinese road-mobile missiles vulnerable again. So China continues to construct tunnels to protect its newer missiles.
The tunnel system is operated essentially as a missile-launch base, or an underground version of a ballistic missile submarine. Just as a submarine deterrent offers survivability, so too does this subterranean force. China builds many tunnels for the same reason a ballistic-missile submarine needs a huge ocean in which to hide.
The tunnels of the underground great wall are hundreds of meters underground, deep in mountain areas, and are difficult to detect from space. Details of the tunnels have not been publicized for obvious security reasons, but it is known that they are scattered across China and are not all connected to one another. They are designed to withstand nuclear and conventional attacks. Rail lines and trucks move missiles, related equipment, and personnel within the network. All the activities necessary for launch preparation can be done in the tunnels. Some of the tunnels may also be used for logistical support or to house command and control facilities.
Beijing's willingness to reveal the existence of the underground great wall – as it did in the 2009 CCTV report -- shows that it wants potential adversaries to know China has a real and reliable retaliatory counterattack capability. From a Chinese perspective, the underground great wall enhances the mutual deterrence between China and the United States, improving strategic stability.
While many security analysts are skeptical about China's no-first-use policy, the underground great wall and the country's small, deeply de-alerted nuclear arsenal are evidence that the minimum-deterrence posture is genuine. Top Chinese leaders, from Mao Zedong to the current leader, Hu Jintao, have publicly embraced the posture. As Mao stated a few months after China's first nuclear test, "We don't wish to have too many atomic bombs ourselves. What would we do with so many? To have a few is just fine." Similarly, Deng Xiaoping once emphasized that China's small number of nuclear weapons "is only to show that we also have what you have. If you want to destroy us, you yourself have to suffer some retaliation as well."
China's minimal-deterrence policy has proven to be effective and smart, providing savings that can be used on economic development. It is unthinkable -- in the opinion of many China experts, including me -- that China would change that policy to pursue extremely expensive weapons parity with the superpowers.
There is one circumstance that could push China to expand its arsenal significantly -- continued development of US missile defenses that might neutralize a Chinese second strike. In fact, such development could become a major driver that speeds China's nuclear modernization.
To discourage Beijing from significantly building its nuclear forces, Washington should accept mutual deterrence with Beijing and limit its missile defenses, so they do not threaten the potential effectiveness of China's small arsenal. For Washington and Moscow to move toward deep cuts in their nuclear forces, China may have to reassure both capitals that it will cap its arsenal at a low level (say, 200 warheads). So long as the missile-defense issue is resolved, that reassurance will probably be relatively easy to obtain, because -– as much evidence shows -- China has long pursued and maintained exactly such a limited nuclear arsenal, hidden underground.