United States-China relations are complex and encompass both cooperative as well as competitive elements. It is not a simple zero-sum game. The issue that best illustrates the complexities is North Korea.
US President Donald Trump has justified the easing of his approach towards China by the need to enlist Beijing's help to stop Pyongyang's nuclear and missile programmes. China certainly has an important role to play, but while American and Chinese interests overlap, I do not think their interests are similar and neither has sufficient leverage to stop North Korea.
Pyongyang has the initiative and has no reason to relinquish it. We are on the cusp of a very significant shift in the North-east Asian strategic equation.
North Korea does not yet have nuclear-armed missiles capable of reaching the continental US; it has probably not yet weaponised its nuclear devices to make them deliverable by missiles. But Pyongyang is determined to acquire survivable, nuclear-armed ICBMs (intercontinental ballistic missiles).
While electronic interventions such as those that then US President Barack Obama reportedly ordered three years ago may delay missile development, I do not think they will deflect Pyongyang from its goal.
Pyongyang is convinced that the Democratic People's Republic of Korea (DPRK) and the current regime will not survive unless it acquires survivable ICBMs. Dissuading a state from proceeding with a course of action it considers existential is impossible.
A balance of mutually assured destruction in North-east Asia will not be a satisfactory situation for anyone. But it will not necessarily be unstable - in fact, it may well be more stable than the current situation.
Pyongyang must be persuaded that the cost of proceeding is higher than the cost of not proceeding. But if you believe that the cost of not proceeding is the end of your state and regime, then you might as well go ahead because the cost of proceeding will always be lower.
The failure rate of North Korean missile tests seems higher since these interventions were said to have begun. But it has also conducted successful tests.
Pyongyang will persist and it will eventually succeed - unless the US and its North-east Asian allies are willing to fight a full-scale war to stop it.
I do not think they are prepared to pay the price. China certainly does not want war. Since the DPRK already has nuclear devices, all it has to do is detonate them on its own territory near the demilitarised zone, which is only about 60km from the South Korean capital Seoul, to raise the cost of war to unacceptable levels. It could conceivably also do so on the border with China. Even if such extreme contingencies could be pre-empted, Pyongyang will certainly retaliate in any way it can, and there is no way of ensuring that all of its retaliatory capabilities can be simultaneously neutralised.
Seoul is within range of conventional artillery, and South Korea and Japan are within range of North Korea's existing missiles. Japanese Prime Minister Shinzo Abe has warned that North Korean missiles could be armed with sarin, a nerve agent. Unilateral military action by Washington would thus impose serious direct risks on its allies at a time when the US itself does not yet face a direct threat.
If the US acts unilaterally, it will, in effect, force its allies to immediately bear the very heavy costs of mitigating threats to itself that are still theoretical or putative as far as the US is concerned. This would cause grievous political damage and could permanently undermine trust in America well beyond North-east Asia.
Only if North Korea already had nuclear-capable ICBMs and there was credible intelligence that a launch targeted at the US was imminent, would unilateral action be politically justifiable or at least understandable.
But the point is to prevent North Korea from acquiring such a capability in the first place. Sanctions do not work. North Korea is already one of the most heavily sanctioned countries in the world. Yet, the speed of its advances in missile and nuclear technology has surprised experts. There is room for additional sanctions and for tightening the implementation of existing sanctions. But this will, at best, only buy time.
WHEN PYONGYANG CAN HIT SAN FRANCISCO
Time for what? North Korea is a brutal but functioning state. Repeated predictions of its demise have proven premature. We cannot assume that it will conveniently collapse before it achieves its goal.
China cannot stop North Korea.
Beijing is certainly very angry with Pyongyang, whose nuclear and missile development programmes have created new risks for China, for example, through the deployment of the US missile shield Thaad (Terminal High Altitude Area Defence) in South Korea. Beijing has signalled its displeasure by stopping coal imports and allowing an unprecedented public debate on whether it should continue with its present approach towards North Korea.
But these are only symbolic actions. China may well go along with further sanctions, but I think Beijing's actions will always fall short of being effective because anything Beijing does that would inflict enough pain to change Pyongyang's behaviour would also jeopardise the stability of the regime.
At a time when the Chinese Communist Party (CCP) is itself feeling insecure, can Beijing be complicit in the destruction of another and neighbouring Leninist system without giving the Chinese people inconvenient thoughts about their own system? This is an unacceptable risk. The most vital of all the CCP's interests is its own survival. From this perspective, a nuclear North Korea is the less bad option.
Foreign policy professionals in the Trump administration know this. But statements by US Secretary of State Rex Tillerson and Mr Trump himself nevertheless seem to display a level of expectation that China is bound to disappoint. How the administration will react when disillusionment dawns is unclear.
NO GOOD OPTIONS
So, what is to be done? There are no good options. Denuclearisation is a pipe dream. Any realistic approach must accept that the DPRK is here to stay and will eventually have a nuclear-capable ICBM.
One approach is to give Pyongyang what it essentially wants - an assurance of regime survival. This could be done by negotiating a US-DPRK peace treaty in return for a verifiable freeze on warhead and missile development. China will, I think, go along with such an approach.
This approach, however, carries two major uncertainties.
First, at what level of warhead and missile development would Pyongyang feel secure enough to agree to a freeze with robust verification measures?
Second, would such a deal be politically saleable in Washington, Tokyo and Seoul? Unfortunately, the two factors contradict each other: What makes Pyongyang feel secure would probably be politically unacceptable.
What is left is the means by which every nuclear-weapon state has hitherto been dealt with: deterrence. North Korea may be very bad, but it is not mad. It is rational. Once it has acquired the survivable ICBMs it believes are needed for regime survival, it can be deterred since Pyongyang will then have no reason to court destruction.
However, deterrence has its own complications. When North Korea has nuclear-capable ICBMs able to threaten the US, the question is bound to be asked - will San Francisco be sacrificed to save Tokyo?
Since the answer is obviously "no", Tokyo will have to seriously consider its own nuclear options. Japan has the capability to develop an independent nuclear deterrent very quickly and has, in fact, been quietly developing this capability - with American aid and acquiescence - for 30 years or so.
North Korea nevertheless is more a catalyst than a cause. China is modernising its own nuclear forces and will sooner or later acquire credible second-strike capability vis-a-vis the US. Thaad may make this even more urgent. When China has credible second-strike capability, the same harsh logic will operate and may do so even before North Korea acquires nuclear-capable ICBMs. One way or another, therefore, American extended deterrence in North-east Asia will eventually be eroded.
The decision will be politically very difficult. But Japanese public opinion has changed very abruptly several times in modern Japanese history, and since the alternative is to accept subordination to China, I believe it is only a question of when and not whether Japan will become a nuclear-weapon state.
I do not think the US is eager to see Japan become a nuclear-weapon state. Neither do I think that Japan is keen to become a nuclear-weapon state.
But, for both, this will eventually be the least bad option. Where Japan goes, South Korea must follow since Seoul is bound to wonder whether it will be sacrificed to save Tokyo.
A balance of mutually assured destruction in North-east Asia will not be a satisfactory situation for anyone. But it will not necessarily be unstable - in fact, it may well be more stable than the current situation - and it may be of some small consolation to Washington, Tokyo and Seoul that the implications for Beijing are somewhat worse.
A balance of mutually assured destruction will freeze the status quo and is an absolute obstacle to Beijing's goal - which is implicit in the essentially revanchist narrative of the "Great Rejuvenation" of China by which the CCP legitimates its rule - of recreating an East Asian order with China at its apex.
If the "American Century" must eventually end, neither will its successor be unambiguously an "Asian Century" or a "Chinese Century". There will be no clear denouement, and we will all have to learn to adapt to structural uncertainty and navigate it for the foreseeable future.