Jeffrey LewisWill China's Deterrent Go To Sea?

I just love this illustration, by Istvan Banyai, that accompanied Keir Lieber and Daryl Press’ article in The Atlantic. This is the first of a three part series on issues surrounding China’s newest ballistic missile submarine. The other posts will discuss how capable the submarine could be and what the new SSBN will mean for US defense planing.

Does the JIN-class submarine indicate that China’s deterrent will go to sea?

The simple answer to that question is “I don’t know.” Anyone who claims to know is blowing smoke up your, um, air intake tube. Even Chinese policymakers who strongly advocate such a course of action would do well not to count their submarines before they launch.

Continuous Operation?

China is, obviously, building at least one submarine. The question, however, is whether China will send the submarine on patrols armed with nuclear weapons.

The Office of Naval Intelligence thinks that China will build five SSBNs to maintain a more or less continuous deterrent at sea:

While China only built a single XIA SSBN, a fleet of probably five TYPE 094 SSBNs will be built in order to provide more redundancy and capacity for a near-continuous at-sea SSBN presence.

I note with some interest that DIA, with perhaps different bureaucratic priorities, predicted in 1999 that China would by 2020 have two JIN-class submarines and the XIA for a total of three boats.

The ONI estimate is the latest of a long line of predictions that China would deploy a fleet of SSBNs large enough to sustain continuous deterrent patrols. In 1974, an NIE predicted four Chinese SSBNs by 1980. In 1984, DIA was suggesting those four boats would be available by 1994.

In the end, China built one SSBN, the XIA-class submarine, that as far as I know never went on patrol, nor was armed with the JL-1 SLBM.

I mention these past assessments not to mock or discredit the intelligence community — but rather to observe a kind of analytic phenomenon.

These assessments were based mostly, I suspect, on the idea that China would send its deterrent to sea because that was the rational thing to do, or to put a finer point on it precisely because that is what we did. Indeed, the other four NPT-nuclear weapons states all maintain SSBNs, including Britain which relies exclusively on “a fleet of four submarines to maintain one continuously on patrol and … assure the invulnerability of the deterrent.”

That China would build a fleet of SSBNs was not a bad guess, although it didn’t turn out that way. Why China didn’t send its deterrent to sea is part of a larger story, I think, about how bureaucratic politics often play a decisive role in defense planning.

Why Did China Build Just One Xia?

The real wonder, to me, is that China built an SSBN at all. Nuclear powered submarines were well beyond the capability of the nascent People’s Republic and the resulting product — the 092 or XIA SSBN — was a dreadful military system

The XIA SSBN and the JL-1 programs were deeply troubled programs, as document in John Lewis and Xue Litai’s China’s Stategic Seapower. The short version is that the submarine itself was extremely noisy, while the missile had a short-range and a troubled testing history. These limitations meant that operational concepts for the SSBN stretched credulity, as Zhang Aiping pointed out to the designers:

At a meeting of the First Academy in April 1975, Zhang Aiping belittled the idea that the PLA might send a submarine as far as the Arabian Sea to launch a missile. Even from there, the closest Asian location, for a sub firing on Moscow, the distance to the Soviet capital would be too far for the JL-1. Zhang concluded with the judgment julang shangan [the Giant Wave (JL-1) must go ashore], and all participants accepted his ruling.

Moreover, I suspect some folks had real concerns that placing nuclear weapons at sea would fundamentally compromise leadership control over nuclear weapons. China has typically exercised rather tight control over its nuclear warheads, keeping the warheads stored separately from its missiles, apparently at separate locations. A sea-based deterrent force would upset those careful arrangements, requiring a much greater degree of trust in unit commanders than Beijing has demonstrated to date.

Of course, countries build stupid or ineffective weapons programs all the time. The XIA’s troubles did not play out in a vacuum, but rather within the competition among bureaucracies. And those bureaucratic interests may have been sufficient to doom even a modestly capable system.

What is now General Armaments Department (then a quasi-autonomous fiefdom eventually called COSTIND) played a central role in advocating for the development of nuclear weapons and maintains, I suspect, substantial influence in decisions about nuclear weapons policy, posture and doctrine. Any concern that the Central Military Commission might have over loosening command-and-control arrangements by placing nuclear weapons on submarines should overlap with a bureaucratic desire by those who currently maintain custody of the nuclear weapons — either General Armaments Department, the Second Artillery or both — to preserve their control and, presumably, influence.

The PLA Navy might have been a weak, or perhaps unenthusiastic, proponent of a sea-based deterrent, given the somewhat strained operational concepts and the pressing need to spend money on the modernization of the Navy as a whole.

Looking back, it is not difficult to see the XIA SSBN as a sort of “science project” dreamt up by an ambitious weapons bureaucracy that looked rather less appealing in the cold light of morning. In any event, as Deng Xiaoping consolidated power in the late 1970s and early 1980s, defense spending declined and more emphasis was given to conventional weapons. Plans for a second submarine were delayed and, eventually, canceled.

Things Could Be Different Now

The JIN-class submarine is probably much more capable than the XIA SSBN. How capable is a subject for later this week, but at the very least I don’t expect to hear stories about how sailors couldn’t sleep because the submarine made so much noise.

ONI is predicting five submarines based, as best as Hans Kristensen can figure out, on the idea that this is what would make sense — again, assuming that we are the ones making the decisions.

But we — that is to say American analysts, China watchers and arm-chair nuclear strategists —are not making the decision, rather Chinese military officials and party cadres are. And the interests of the Chinese participants are likely to reflect their own bureaucratic positions rather than an impartial defense analysis.

That is true for ONI, too, which is not without its own bureaucratic perspective on the growth of the Chinese navy. This is not to say that ONI’s employees are either cynical or wrong, but just that where one stands often depends on where one sits.

The idea of keeping a submarine at sea with a dozen or more nuclear weapons is a very different posture for China, one that would likely only occur with substantial changes in who (or Hu) is making decisions about strategic forces in China. If China keeps its new land-based mobile missiles say “assembled and in the garage” with its warheads in another location, then maybe China will also keep its ballistic missile submarines in port and away from US ASW platforms looking for a little training. One could imagine a force of one or two submarines that rarely, if ever, patrol.

That force would be very vulnerable, but it would replicate the basic balance we see currently in China between readiness, on the one hand, and, on the other, leadership control.

Of course, some of the bureaucratic structures that have underpinned China’s no-alert posture have undergone substantial change as the country has modernized. The old COSTIND, which was corrupt and rife with nepotism that it would make Dick Cheney blush, has been transformed into the General Armaments Department as part of a long running effort by the PLA to re-exert control over the quasi-autonomous unit. The degree to which GAD competes or cooperates with the Second Artillery isn’t really clear to me, although things may be less fractious than when COSTIND looked at the PLA much like a wolf looks at sheep.

The Second Artillery is vastly more professional today, as is the Navy. That professionalism may make possible a greater delegation of launch authority, or a much more capable effort to wrest control of the country’s strategic programs.

On the other hand, letting relatively junior officers out into the field with real missiles and real nuclear weapons isn’t something the Second Artillery has ever done — save for a 1966 test launch of a live nuclear warhead on a DF-2.

Absent evidence that China is patrolling sending nuclear weapons out into the field on a day-to-day basis, it is hard to see why a new system would — in and of itself — result in a change in posture. It would seem to me that sending the deterrent to sea would be part of a much larger change, let’s call it a sea-change, in how Chinese leaders look at nuclear weapons.

Wait For the Evidence

All this is a plea for patience. We can’t, I think, assume that China’s operational patterns will resemble our own just because we are that awesome.

Chinese leaders have their own concepts, petty bureaucratic feuds and asses to cover. Everything to date suggests that the technological imperative to develop systems has not resulted in Chinese forces that kept on alert.

In particular, I am interested in DIA’s suggestion that China might continue to operate the XIA submarine along with a fleet of two JIN class submarines. A colleague of mine has observed that China maintains a kind of “artisan’s” approach to defense procurement — building a few missiles, making some small changes, building some more — that is different from our defense procurement system.

Of course, China may choose to build a small fleet of submarines, sending the boomers to prowl the depths of the Pacific while DF-31s tear up the local highways. A more alert Chinese nuclear arsenal would require policymakers in Washington to think very differently about China’s nuclear weapons — something else that I’ll talk about later in the week.


  1. None (History)

    Well argued. It is a big leap to put armed, ready-to-use nuclear warheads on missiles on patrol for China.

    Look in the military history of China, and you will find it is rife with examples where commanders defected to ‘the other side’ together with the entire armies and equipment under their command.

    Whereas such defections in the Western World meant troops going to another side that spoke different languages, have a different ‘nationality’, in the Chinese context, the defection can often mean switching from, e.g., the communist to the nationalists or vice versa.

    Likewise, look at the history of how Chinese have responded to invasions from outside of China by ‘barbarians’. In many instances, after the invaders won enough battles, the rest of China ‘goes along’ and the process begins to assimilate and transform the invaders into Chinese. This again, involve individual commanders making the decision to switch sides.

    The bottom line, loyalty to any particular Chinese regime / dynasty cannot be taken for granted by the central government.

    Giving them a boatload of nuclear weapons that can be as much as 1/4 of China’s deployed nuclear force is a hard bridge to cross.

  2. Andy (History)

    Excellent analysis.

    Personally, I believe US Naval planners would love for China to put its nuclear eggs into the SSBN basket. The qualitative technical and experiential advantage the US holds would make patrolling Chinese SSBN’s very vulnerable compared to mobile land-based missiles, at least in any conflict with the US. SSBN’s would make more sense against other potential adversaries, like India, which have no real ability to interdict them.

    The strategic calculus, therefore, may be to provide a diversified set of nuclear assets to deter a diversified set of potential threats. If this is the case, then China would not need the ability to continuously patrol that five subs would provide since a two-three sub fleet could be sortied in response to increased tensions, or, like the Russians apparently planned, their missiles could be launched while the subs are still pierside.

  3. Hallo84 (History)

    The modernization of PLA was more of broad spectrum rather than limited to weapons platform.

    The three pronged approach of combining PLAAF, PLAN, and 2nd Artillery’s ascendancy to CMC was a successful move by Hu to instigate new command structure that represented the increased involvement of the three components of PRC military.

    Chinese nuclear deterrence was never envisioned to be delegated to the navy. In truth PLAN did not hold precedence in Chinese politics to be taken seriously until very recently.

    The fact is China have already established a well dispersed rail based system under command of 2nd artillery. The role of SSBN will still be secondary in the near future.

  4. Anonymous

    Two thoughts/questions:

    (1) Will India beat China at deploying nuclear weapons at sea, either on its Dhanush ship-launched platform or eventually on the ATV?

    and (2) If India does, does that make a damn bit of difference as far as China’s strategic calculus goes regarding whether to deploy the weapons at sea in peacetime (which India could choose not to do either)?

  5. Vigilis (History)

    Enjoyed your article’s paradigm-absent views.

    Noted that you have also not broached the subject of Taiwan, yet.

    The primary counter to Chineese SSBN patrols by non-communist countries would be U.S. SSNs. About 60% of these are now WestPac dedicated.

    U.S. military preparedness enables freedom-loving Americans and Taiwanese to sleep well and exercise their liberties in open forums like yours, of course.

    Maintenance of life, liberty and the pursuit of happiness, it seems to me, calls for superb contingency planning. Reliance on your theory that China might not deploy SSBNs, although possible, seems tantamount to disarming city police in hopes criminals would behave.

    Throughout time, however, human nature has stubbornly confirmed that preparations for war are well-advised as the surest way of avoiding it short of unilateral surrender.

    Looking forward to parts 2 and 3.

  6. Another Anonymous

    Good article. But it still begs the question why the Chinese went through with the huge investment to develop the Type 094 and the JL-2 missile if they only envisioned a very narrow set of circumstances for deploying the sub and missiles. The greatest asset of a SSBN deterrent is its stealth and survivability. By limiting the deployment to times of crisis, you pretty much give up both of those by tipping off the enemy something is up when the sub deploys.

  7. johnwbragg (History)

    Followup to None:Given China’s policies of minimum deterrence and extreme ambiguity, maybe they’d send a boat out on patrol in the Yellow Sea without warheads, or even without missiles. We’d never know. We’d probably debate whether the SLBMs were MIRVed or not.

    Heck, if the Chinese can get the old Xia out to sea, half of the US intelligence community would start arguing that they must have retrofitted the Xia with the J-2 MIRV.

    Bluff, intentional or not, has served them well. How many decades did we assume that China had tactical nuclear weapons based on a military exercise where China used them?

  8. johnwbragg (History)

    Followup to Another Anonymous: Why go through the time and effort of developing capabilities that you don’t have much use for yet? You sound like my middle school students. To learn, and to prepare for future learning.

    Maybe it’s a long term play, minimal investment over, say, 50 years, so that they’ll have the option of a real 3-4 SSBN continuous-patrol second-strike deterrent in 2030.

    Compare it to the percentage of GDP we spend on blue-sky projects. It’s worth it just for long-term R&D.

  9. Another Anonymous

    johnwbragg – If the PLAN only wants to experiment and develop a doctrine for SSBN employment, they already have that in the Xia and the JL-1. The Type 094 and the JL-2 are an order of magnitude improvement in capabilities. To me, having the ability to conduct a survivable second strike against the US, but not developing the doctrine to effectively employ it, is puzzling.

  10. Robot Economist (History)

    Given the large number of technical and doctrinal complications associated with the program, maybe the PRC is pursuing the Type 094 because it is a status symbol—like their tiny manned space program.

  11. Analyst (History)

    I believe the test of the missile armed with a live nuclear warhead to which you refer happened in 1966, not 1964. The first Chinese atomic test was in 1964.

  12. Jeffrey Lewis (History)

    Ack! Yes. My brain crosswired CHIC 4 and 1966 into 1964.

  13. johnwbragg (History)

    AA: It’s possible that the Chinese have in one, er, Great Leap Forward, have duplicated the pretty-quiet workhorse Soviet subs (Delta III and IV) of the late Cold War. It’s also possible that they’re at the point on the learning curve where they’re still building target dummies for US SSNs (Yankee class) or maybe they’ve built a sophisticated system that breaks down for 50 different sophisticated reasons. (Typhoon/Akula?)

    Given China’s submarine history, maybe building a quality SSBN fleet is seen as about as big a challenge as building a workable missile defense is for the US. Doable in theory, but a lot of ground to cover between here and there.

  14. Daniel Kimerling (History)

    Great post. You bring up a very interesting point about the Chinese not sending there subs on tours. Without practicing against enemy subs, it would be very hard for American hunter killers to go after Chinese boats. The operational tempo of submarines during peace is not that different than that during war, or at least that was how it was during the Cold War. If the Chinese keep the Jin and Xia close to shore, we will never be able to get good at hunting them.

    The Chinese also keep their subs in under ground caves, so it would be very hard for the US to detect them going to sea. If their ssbns do go deep, we may not know they have left port. That provides the Chinese with a dangerous advantage.

  15. None (History)

    Based on the history of how the US was able to intercept Soviet communications by planting bugs on undersea cables often well inside Soviet territorial waters, it is a fair bet that all possible entrances and exits from the undersea caves have been sown with high tech sensors that record the comings and goings of the Chinese subs even when they are in territorial waters.

    Robotic drones are just too cheap and plentiful not to have them deployed en mass against the Chinese threats.

    Finally, even if they manage to get out of port, the geography of their home bases are such that it is virtually impossible for the Chinese subs to patrol far before they run into assets deployed by Korea, Japan, Taiwan, the US, and other powers in the region.

    To suggest that it takes lots of practice to get good at hunting these noisy submersible is an astonishing insult at the quality and capability of the American and allied SSN and ASW forces.

    What about the converse, which is because they never come out, they do not have the chance to learn the capabilities and limitations of their boats and the tactics used by America and its allies in countering them. Logic would suggest that in a contest between a highly trained, experienced, and capable American navy and allies with over a century of experience with submarines (since 1775) and ASW warfare, greatly favors the US, not China. To suggest otherwise is fantastic.

    Calling a few noisy boats with a desultory load-out of missiles (if they are even loaded) a ‘dangerous advantage’ is an interesting thought worthy of Jules Verne.

    If the Chinese SSBNs are so dangerous, then how dangerous must the British, French and Russian fleet really be?

    Is there not a small risk that these presently not unfriendly countries might end up as enemies in the US?

    How dangerous would their SSBNs be then? Just in case, shouldn’t the US be prepared against the British and French fleets just in case they switched sides?

    Maybe what the Chinese should really do is to build a replica of the Nautilus and send it to make practice runs at US warships outside of Pearl Harbor?

    Disrupt the American saltpeter supply, and that will be the end of the empire…