International Arms Control in a Disordered World

Panel Presentation to World Peace Forum, Tsinghua University, Beijing, 4 July 2026

Nuclear arms control, on which I will focus my remarks, is not the only international arms control issue demanding the attention of policymakers. Ensuring the effective enforcement of the Chemical and Biological Weapons Conventions is a continuing concern; more work needs to be done to fill gaps in the Outer Space Treaty; the new risks associated with hypersonic weapons, cyber warfare, lethal autonomous weapons systems and artificial intelligence generally have barely begun to be tackled; and effective conventional arms control in a world awash with more than a billion such weapons, including small arms, continues to face a mass of challenges.

But it is nuclear arms control that I believe demands our most urgent attention. Not just because nuclear weapons remain the most indiscriminately inhumane ever devised. Not just because any significant exchange of such weapons would pose an existential threat to life on this planet as we know it, even more than other weapons of mass destruction and destructive new technologies at their present stage of development. Not just because, even if these weapons are never used with deliberate aggressive intent, there remains a huge risk of stumbling into catastrophe through accident, human error, system error or sabotage, with it being nothing more than sheer dumb luck that such a fate has so far been avoided. An urgent new focus on nuclear arms control is necessary not only for all these reasons, but because this is the one area of international arms control in which in recent years we have actually been going backwards.

The big arms control agreements of the past which, at least between the US and Russia, banned certain systems outright, set constraints on deployments and build confidence through transparency, are now all dead: the Anti-Ballistic Missile (ABM), Intermediate-Range Nuclear Forces (INF), Open Skies, and Strategic Arms Reduction treaties. With the expiration of New START in February 2026, Russia and the United States will be without any limits on their nuclear forces for the first time in over fifty years.

Moreover, crucial multilateral treaties are on life support. The Comprehensive Nuclear Test Ban Teaty (CTBT), though operating in practice, has still not been ratified, and the Treaty on the Non-Proliferation of Nuclear Weapons (NPT) is in as fragile a condition as it has ever been, with the last three five-yearly review conferences all concluding without any consensus agreement. The loss of confidence in Washington's willingness to deliver on its extended nuclear – or even conventional – deterrence commitment to close allies has led to serious discussion in South Korea, and even in some quarters in Japan, and in Europe, about the need to acquire their own nuclear weapons. As to Iran, I remain unconvinced that it has ever decided to or would actually acquire nuclear weapons, as distinct from demonstrating its capacity to do so. But should it so decide, and succeed, there is every prospect that the non-proliferation dam will finally burst, in the region and beyond.

The dramatic reduction in global nuclear weapons numbers that occurred after the Cold War ended has now stalled, and is being reversed. Every nuclear arms state is now modernising or increasing its arsenal, especially China, whose inventory has doubled in a few short years to some 600 weapons, with new land and seaborne delivery systems. The nine nuclear-armed states possess between them over 12,200 nuclear warheads, with a combined destructive capacity of more than 145,000 Hiroshima bombs. Some 9,000 of these are militarily active or deployed. Alarmingly some 2,000 US and Russian weapons remain on high alert ready, to be launched within a decision window for each president of four to eight minutes.

More troubling still, the longstanding taboo against the use of nuclear weapons seems to be weakening, with Russia's President Vladimir Putin in particular talking up this prospect in the Ukraine war in language not heard since the height of the Cold War. A number of states are considering using nuclear weapons – especially so-called ‘tactical’ weapons – not just for deterrence but for warfighting.

While nuclear arms control has never been more necessary, and never more difficult to achieve, it is crucial that in all three of its dimensions – risk reduction, non-proliferation and outright elimination – the effort continue to be made.

The case for outright elimination remains compelling. Nuclear weapons have proved at best of minimal, and at worst of zero, utility in maintaining peace, whether the context be deterring war between large nuclear-armed powers, or deterring attacks by non-nuclear armed states against states with such arms. Nor have they ever been likely to help protect weaker states against conventional attack: weapons that would be manifestly suicidal for a state like North Korea (or Ukraine if it had retained its Soviet stocks) to use are not a very credible deterrent.

But given current global geopolitical realities, and the psychological comfort that nuclear weapons seem to continue to give their possessors, complete disarmament does seem for now to remain a very distant dream. New heart has been given to anti-nuclear campaigners by the Treaty on the Prohibition of Nuclear Weapons (TPNW) agreed by 122 UN member states in 2017, and in force since 2021, which bans nuclear weapons outright. But there is less to the Ban Treaty than meets the eye. It is aspirational rather than operational, with serious gaps when it comes to verification and enforcement. And there is no prospect for the foreseeable future of it being signed or ratified by the states that matter – the nuclear-armed states, or the ‘umbrella’ states (like Australia) believing themselves, sensibly or not, to be sheltering under their protection.

For now, while we should never abandon the goal of ultimate elimination, it makes most sense to focus nuclear arms control efforts on holding the line on proliferation, and on nuclear risk reduction. As to non-proliferation, more cohesive middle power diplomacy in forums like the UN and G20 might be helpful in increasing the pressure on wavering states, but the enterprise will remain on a knife-edge so long as all the present nuclear armed states maintain their blatant hypocrisy in demanding that others forego weapons that they are totally unwilling to give up themselves.

When it comes to nuclear risk reduction, and this I believe should be our main policy focus,1 the critical priorities are what can be described as the ‘4D's’. They are Doctrine (getting universal buy-in for a No First Use [NFU] commitment; Deployment (drastically reducing the number of weapons ready for immediate use); De-alerting (taking weapons off high-alert, launch-on-warning readiness); and Decreased numbers (dramatically reducing the global stockpile).

In many ways the most game-changing of these measures would be the universal embrace of No First Use, meaning that each nuclear armed state would make an explicit declaration that it will not use nuclear weapons either preventively or pre-emptively, against an adversary, nuclear-armed or not, and keeps them only for use or threat of use as retaliation following a nuclear strike against itself or its allies.

China and India are the only states currently committed to NFU, and it is crucial that their support continue. Russia abandoned an earlier pledge in the 1990s; France has long maintained a first use posture; the United Kingdom, Pakistan and North Korea have not ruled it out; and Israel, as ever, refuses to confirm that it has nuclear weapons. The US came close to adopting a ‘sole purpose’ declaration (the functional equivalent of NFU under Presidents Obama and Biden but, unhappily, was talked out of it by their allies in Europe and the Asia-Pacific.

But there continues to be widespread support for NFU among advocates for nuclear sanity, and for good reasons. Retaining a first-use option is dangerous, because it runs the risk of an adversary misreading its intentions and, fearing decapitation, launching a pre-emptive strike, precipitating an avoidable nuclear war. And it is unnecessary, because the major nuclear powers all have immense conventional firepower, amply sufficient to deter or respond to chemical, biological or other non-nuclear attack.

China has the capacity to play a real leadership role in nuclear risk reduction given the credibility it has acquired over many years for its ‘minimal deterrence’ posture and general restraint on nuclear issues, if not for its transparency. It has been reluctant to do so while the size of its nuclear arsenal remains so much smaller than those of the US and Russia, but on any view the present PRC stockpile and deployment, including its second strike capability, is ample for the country’ protection, and China would make a huge contribution to the restoration of global nuclear sanity if it were prepared to join the US and Russia now in negotiating a new Strategic Arms Reduction Treaty.

The most crucial immediate need in re-energising nuclear arms control is to restore between the key players the dialogue – and the trust and confidence that flows from such contact –that has been a central value of past arms control measures. Even the most bitter antagonists have a shared interest in avoiding nuclear confrontation. It is important, not least at times of crisis, that senior officers and officials know each other and understand each other’s concerns, and have easy and open lines of communication. On this front, unhappily, around the world, we are currently a long way from where we need to be.