Jeffrey LewisTrial Operational Deterrence

Yeah, that’s a weird title, huh?  Well, it’s a weird idea.  Bear with me.

Over the course of my writing, teaching and corresponding about China’s nuclear forces, I kept stumbling over the same question: Why does China insist on calling the Second Artillery, China’s nuclear-armed missile force, the “core force” for China’s deterrent?

“Core force” — and other similar phrases — imply the possibility of nuclear operations by the Chinese Air Force or  Navy, even if China relies on the Second Artillery’s land-based missiles for the old minimum means of reprisal (or assured retaliation, if you prefer).  Indeed, some training materials like Science of Second Artillery Campaigns make reference to joint nuclear operations. It seems impossible to imagine this is entirely anticipatory.

But the United States intelligence community pretty clearly thinks China’s lone Xia-class nuclear submarine never became operational and that no Chinese aircraft have nuclear delivery as a primary mission. Of course, this may change once the new Jin-class SSBN is operational. I think nuclear-armed H-6 bombers, on the other hand, are less likely, but can’t rule them out.  But, again, these haven’t happened yet, and it is hard to believe that China would use language like “core” force in an entirely anticipatory sense.

So, I’ve always wondered: Why does Beijing uses the “core force” language and the refer to “joint” nuclear operations?

I’ve come to think that the answer lies in the Chinese notion of “trial operational deployment.”  Which sounds strange, and informs the title of the blog post.


Trial Operational Deployment

In December 1980, China deployed two DF-5 ICBMs in silos. Yes, all of two. The full deployment of three brigades totaling 18 DF-5s didn’t occur until the mid-1990s. (May 1995, according to one source.)

Why bother with two measly silos? In an alarming situation, two ICBMs are better than nothing. And Beijing viewed the international security situation in the 1970s as alarming. “To Beijing, the situation in the late 1970s was alarming,” John Lewis and Hua Di wrote, “The Soviet Union seemed to be on the offensive and prevailing, while the United States was retreating and losing. On October 30, 1979, Marshal Nie Rongzhen … directed the urgent deployment of all available strategic weapons systems, saying that though a bit backward in performance, [the DF-4 and DF-5 missiles] would still be better than ‘millet plus rifles’ in fighting a war.”

The phrase “millet plus rifles” is a reference to favorite remark of Mao about the relative unimportance of weaponry in deciding the outcome of a conflict, especially when compared to ideological factors. Here is an example from Mao in 1955:

We have an expression, millet plus rifles. In the case of the United States, it is planes plus the A-bomb. However, if the United States with its planes plus the A-bomb is to launch a war of aggression against China, then China with its millet plus rifles is sure to emerge the victor. The people of the whole world will support us. As a result of World War I, the tsar, the landlords and the capitalists in Russia were wiped out; as a result of World War II, Chiang Kai-shek and the landlords were overthrown in China and the East European countries and a number of countries in Asia were liberated. Should the United States launch a third world war and supposing it lasted eight or ten years, the result would be the elimination of the ruling classes in the United States, Britain and the other accomplice countries and the transformation of most of the world into countries led by Communist Parties. World wars end not in favour of the warmongers but in favour of the Communist Parties and the revolutionary people in all lands. If the warmongers are to make war, then they mustn’t blame us for making revolution or engaging in “subversive activities” as they keep saying all the time. If they desist from war, they can survive a little longer on this earth. But the sooner they make war the sooner they will be wiped from the face of the earth. Then a people’s united nations would be set up, maybe in Shanghai, maybe somewhere in Europe, or it might be set up again in New York, provided the U.S. warmongers had been wiped out.

Hence the image atop this post, which is a modern interpretation of a painting by the artist Shi Lu, entitled Millet Plus Rifles.
Mao’s point, and Nie’s, is that, as you know, you go to war with the Army you have. According to Lewis and Hua, the Chinese had a name for this notion of initial operational capability —  “trial operational deployment.”


Nuclear-Capable Aircraft

China conducted nuclear weapons tests in the atmosphere through 1980, something that caused Beijing a lot of political grief. (In fact, China’s opposition to the Limited Test Ban Treaty is a large part of how “no first use” came to be a major diplomatic position for Beijing. But you’ll have to buy the book for that story.)

China tried reasonably hard to move testing underground as soon as feasible, but despite a number of underground nuclear tests, China continued atmospheric testing throughout the 1960s and 1970s. As a stop-gap measure, China conducted  atmospheric nuclear tests using retrofitted bombers and specially manufactured fighter aircraft. Dropping nuclear weapons from aircraft introduces some risk into tests. A test in 1979 failed when the parachute failed to open; here is a pretty cool first-person account in Chinese.  On the other hand, airbursts reduce the amount of radioactive fallout, which helps reduce the political fallout.  For Beijing, the added risk of testing from aircraft was worth the political benefits.

China Today: Aviation Industry, an official account of Chinese aircraft industry from the 1980s, describes in some detail the program to modify aircraft to serve as assets for the nuclear weapons testing program. China retrofitted  one H-5 bomber, (probably) one H-6 bomber, and several Q-5 aircraft.  Here are some comments on each program:

  • “In order to cooperate with the development and test of the atomic bomb the Bureau of Aviation Industry assigned the Xi’an Aircraft Factory a task in 1963 to retrofit a H-6 aircraft assembled by the Harbin Aircraft Factory in 1959 into a nuclear weapon carrier.”
  • “In September 1967 the government assigned a task for retrofitting the H-5 into a nuclear carrier which could be used both for nuclear test and operational missions.”
  • “In order to support nuclear test the Nanchang Aircraft Factory completed the manufacture of several nuclear weapon carriers which were derived from the basic Q-5 in 1970.”

There is a curious little phrase — “a nuclear carrier which could be used both for nuclear test and operational missions.”

Operational missions? Really? Just one aircraft?  (The reason I think it was only one aircraft is that the Harbin Aircraft Factory completed the retrofit with “intense work over half a year.”)

Who the hell was going to fly that thing to Moscow, Major Kong?  (I know, the range is a stretch.  Work with me here.)

I find it curious that China would consider a lone H-5 bomber as an operational capability.  Then again, I suppose in an emergency,  just like two ICBMs, one bomber would be better than “millet plus rifles.”

The H-5 conversion predates the term “trial operational deployment,” which didn’t come around until 1974, according to Lewis and Hua, but it’s the same idea.


Xia-Class SSBN 

That brings us to China’s lone SSBN for much of the past thirty years — the Xia-class SSBN. (China has a new fleet of three SSBNs just waiting for their shiny new SLBMs.)

In October 2013, Xinhua showed off nuclear submarines, including the Xia-class SSBN.  This struck me as strange, since the Xia-class submarine rarely leaves port and the United States has never considered it an operational platform. To bring targets into range, the Xia would have to sail long distances. Zhang Aiping, according to John Lewis and Hua Di, ridiculed the notion that it would sail as far as the Persian Gulf to bring Moscow into range. After all, Major Kong is already assigned to the H-5 bomber.

Yet we have the Xia-class SSBN in Xinhua, accompanied by approving remarks by military analysts. “China says it has a no first use nuclear weapons policy,” Xinhua quoted a military officer named Yin Zhuo saying, “Nuclear submarines can effectively deter and fight back against those who want to launch nuclear attacks on China. It can reduce the danger of nuclear war.”

Just like the two ICBMs in the early 1980s or the one retrofitted H-5 bomber, the one  Xia-class SSBN is better than millet plus rifles.


Trial Operational Deterrence

So, this was a pretty anemic triad in the early 1980s — a couple of ICBMs, one submarine and one bomber. Sure, China had larger theater forces, but the overall force was very, very small.

The Chinese built a number of test assets that neither we nor they consider “deployed.” The difference is that the Chinese think they get some measure of deterrence even from test assets.

The notion that test or developmental assets  — whether it is two ICBMs in silos, an H-5 bomber or an SSBN that never leaves port — might confer some measure of deterrence might seem very strange. In Western academic literature, we tend to think the period when new nuclear forces are under development is the moment of maximum danger for a nuclear aspirant — a small number of provisional assets might be said to invite attack, not deter it. The Chinese, on the other hand, seem to think that even the most limited capability helps out. You just add it to the millet plus rifles and go.

From an American perspective, we would place little reliance on test assets. “You fight how you train” is a popular bit of wisdom. Sending forces with little training and no operational experience on a long, one-way journey to retaliate against a nuclear attack would seem, to a American perspective, like a fool’s errand. Our arms control treaties even make special provisions for test assets because, you know, they aren’t part of the “real” force.

Then again, maybe the Chinese are right. Neither the United States nor the Soviet Union attacked China, despite Beijing’s prolonged period of vulnerability. Looking at declassified US intelligence estimates, we gave the Chinese credit for far more nuclear bombers than they probably manufactured. Perhaps the Chinese are right to think that deterrence depends far less on details like readiness, training, and operational plans than we think in the United States.

Regardless of who is right, Beijing’s references to other nuclear forces seem to refer to the possibility that it might press into service the Xia-class SSBN or aircraft modified for nuclear weapons testing in emergency.  It’s an interesting notion, one that colors how I think about the 1969 Sino-Soviet crisis — but that’s a conversation for another day.

For now, I think it’s sufficient to note the history of Chinese trial operational deployments, even if they don’t always formally go by that name. China does not have a capital-T Triad on this basis — they don’t think of it that way, nor do we — but understanding the Chinese attitude toward developmental systems helps illustrate a broader point about how Chinese leaders have thought — or at least acted as though they thought — about nuclear weapons.  Possession, or perhaps mastery, is the key, with other details like numbers, posture, or even readiness given less emphasis.  That’s a broader mindset that helps explain why Chinese leaders have tended to treat strategic stability as a foregone conclusion, despite their own relatively small and vulnerable forces.


  1. j_kies (History)

    Based on the historic demonstration where a teenager piloted Cessna pentrated the Soviet airdefenses ‘up the gut’ to Red Square; a single H5 with a fairly inefficient TN device flown by a dedicated airman sounds like a pretty credible deterrent.

    • Jeffrey (History)

      Provided Lei Feng pilots it one-way.

  2. Cthippo (History)

    It’s an interesting question: “When does a deterrent become credible?”

    Given the uncertainties involved, I would say that once a country tests a nuclear weapon it’s potential adversaries have to assume that it has more and can deliver them. This is probably an incorrect assumption, and everyone involved in making the assumption knows it, but the consequences of being wrong are too high. It’s an old axiom in intelligence that when you don’t know you assume the worst.

    Glad to have you back amongst the living, or at least communicating, Jeffrey!

  3. Juuso (History)

    Interesting footage from China’s last nuclear weapon test.

  4. John Schilling (History)

    Deterrence is based not on the weapons you have, but the weapons people believe you have. And really, what are the odds that people will believe you have exactly one nuclear-capable bomber? If you demonstrate one, they’ll believe you have at least a squadron. If you demonstrate one and say to the world, “we only have the one – come, look and see”, they’ll probably still believe you have at least a squadron. The necessary modifications are minor (arguably zero) and internal, and it’s not like the Chinese went to any great lengths to convince the world that what we saw was all they had.

  5. SQ (History)

    This is a pretty significant reframing of how we understand Chinese nuclear policy over here. More in the book, yes?

    • Jeffrey (History)

      I don’t see it as reframing, at least from what I’ve written in the past. I think its a strategic monad in service of minimum deterrence that flirts with an existential deterrent. What I have realized, and made explicit, is that idea that, if deterrence is so easy to achieve, even test assets contribute.

  6. JO (History)

    Another example of this approach: for four (unpredictable) years the South African nuclear deterrent consisted of a completed nuclear test assembly and an instrumented site. Then there was a step change to six deliverable bombs. Carefully thought out and documented in policy records.

    • Cthippo (History)

      In the South African case one has to wonder whether it was an effective deterrent since no one was sure they HAD an arsenal. South Africa made preparations to conduct a test, but never actually made the shot, and the only people who knew they were making the preparations were the CIA and/or NRO who kept a lid in on the information.

    • Carey Sublette (History)

      Responding to Cthippo:

      But who were they deterring, really?

      South Africa was the superpower of Subsaharan Africa, without any actual rivals. The only purpose of the nuclear arsenal was to influence Superpower behavior (blackmailing U.S. support; keeping Cuban troops out of South West Africa).

  7. Hippo1 (History)

    The concept of “trial operational deterrence” fits in nicely with the Rumsfeld Commission’s idea that the U.S. intelligence community should not assume that other countries will consider their missles operational beased on our concepts of “safety, reliability, accuracy and requirements for high volume production.” This was their primary reason for arguing that the U.S. was facing a greater ballistic missile threat and thus justifying their call in other venues for a robust BMD.

  8. Bob (History)

    This should probably be seen in context with the Chinese idea that uncertainty enhances deterrence. From that perspective, it makes sense for the Chinese to act like systems such as the Xia-class SSBN actually works. The adversary cannot be fully condident that it does not.

    In a way, the idea is sound: even if the chances of vulnerable nuclear assets surviving a first strike small, the potential cost for the adversary of missing an asset or two are vast. That provides some degree of deterrence.

  9. panterazero (History)

    Millet plus rifles are cheap and so is an even approximately credible trial deterrent. Follow the renminbi.