The 1990s and early 2000s have been described as a ‘golden age of nuclear arms control’. But a steady increase in global tensions since then has led to the demise of agreements on arms control that provided a stabilizing base structure for the global security architecture. With some strategists coming to believe that agreements like the Anti-Ballistic Missile (ABM) Treaty or the Intermediate-Range Nuclear Forces (INF) Treaty were overly constraining, the US and Russia have both ignored, suspended, withdrawn from or violated a wide range of nuclear and conventional arms control treaties over the last 20 years. Other agreements have been allowed to expire without extension or replacement – most recently, the New Strategic Arms Reduction Treaty (New START).
The deterioration of the geopolitical environment has led some experts to suggest that this type of arms control is no longer feasible. Some even question whether its is desirable. However, as states worldwide are investing more money in their military capabilities, arms control is crucial if states want to reduce the risks of all-out war.
The evidence from the Cold War shows that arms control agreements help to create a sense of strategic stability and play a crucial part in stopping arms-race dynamics from taking hold. Without those agreements, uncertainty and a lack of transparency about the size and structure of nuclear forces significantly increase the risk of inadvertent escalation in a crisis. If risks increase further, and the economic pressures of arms racing become apparent, arms control may become an attractive prospect again.
Given the technological changes since the end of the Cold War and the more complex strategic relationships among nuclear-armed states in the 21st century, new arms control agreements will likely look different from traditional bilateral treaties like New START.
For that reason, this paper defines arms control broadly to also include instruments such as confidence-building measures, strategic stability dialogues and other mechanisms for technical exchange, as well as normative and behavioural agreements. A broader definition of arms control, based on that of Thomas Schelling and Morton Halperin, which moves beyond referring exclusively to legally binding treaties, has become more common among practitioners as the space for a multilateral approach has become more constrained.
An expanded conception of arms control could create space for diplomacy. States that are not ready to engage in negotiations leading to a new legally binding treaty may be more willing to commit to incremental steps. An example would be a technical-level dialogue on how new systems fit in with a country’s stated nuclear doctrine. Such platforms can, in turn, provide a basis for future negotiations on a more traditional arms control agreement, as the diplomats and experts involved in the earlier technical dialogues will have a greater understanding of each other’s threat perceptions and security interests.
An expanded conception of arms control could create space for diplomacy. States that are not ready to engage in negotiations leading to a new legally binding treaty may be more willing to commit to incremental steps.
Arms control has previously relied on agreements being verifiable, often via on-site inspections of warheads and delivery systems, and observing where physical assets are deployed. The integration of new and disruptive technologies into military systems will require a rethinking of verification, as the move to digital or dual-use defence tech makes verification far more difficult and invasive. States will, for example, likely be unwilling or unable to share a software codebase. Agreements aimed at behavioural constraints could provide a solution to this problem.
About this paper
The paper expands the discussion on arms control in multiple ways. First, it goes
beyond the traditional focus on the US and Russia to discuss the relationships
between other nuclear-armed states. Second, the paper also explains why China’s nuclear build-up complicates some options for arms control, but might enable others. And finally, its discussion of behavioural and normative arms control variations adds to the emerging literature on these topics.
It will assess the possibility for new arms control agreements in four relationships between nuclear-armed states. Those relationships are:
- The US and Russia;
- The US and China;
- The N5 group of states (China, France, Russia, the UK and the US); and
- India and Pakistan.
The state-to-state relationships discussed in the paper are those with the greatest potential for escalation, mistrust and proliferation, making them critical focal points for reducing global security risks. Any arms control agreements between the countries involved would immediately reduce the risk of nuclear escalation.
In addition, these are relationships where there is some prospect of the states involved engaging in some form of arms control. North Korea was excluded from our study, as the prospects of the Kim Jong Un regime engaging in dialogue were considered much lower than for the four relationships listed above.
This paper distinguishes between the P5 – i.e. the five permanent members of the UN Security Council – and the N5, which refers to the five nuclear weapons states signed up to the Non-Proliferation Treaty (NPT). The same five states currently make up both the P5 and the N5, though the composition of both groupings has varied in the history of the UN and the NPT. The overlap is further complicated through the existence of the P5 process, a dialogue process for the N5 in the context of the NPT. However, this paper does not want to add to the impression that nuclear weapons are required for permanent membership of the Security Council.
Each of the following sections of the paper will focus on one of the four relationships listed. These sections briefly outline the current state of the relationships, the stabilizing role that renewed arms control could play, and under what conditions arms control agreements might prosper. The paper closes with recommendations for how to arrive at those conditions.
The conditions under which arms control is most likely to take place have been constructed using data from a horizon-scan survey. Horizon-scanning aims to systematically identify and evaluate trends and opportunities for change, and determine whether specific factors are accelerating or decelerating trends. It draws on wide-ranging expertise to cover as many impressions as possible. Horizon-scan leaders then distil the most striking insights, contrasting and explaining diverging views and highlighting where experts agreed.
Our survey took a diverse group of arms control experts with specific expertise on at least one of the state relationships listed above through a series of questions on: 1) the state-to-state relationships in focus; and 2) the feasibility of reaching new arms control agreements generally over the next decade, with an emphasis on opportunities for change. The expert insights were then contextualized in the wider history of these state relationships, past negotiation experiences and current positions to reveal opportunities and make recommendations.
