As tensions between the United States and China have worsened, it has become common to claim that we are entering ‘a new Cold War’. In truth, there are countless differences between the US-China situation today and the freeze that was maintained between the West and the Soviet Union from 1947 until 1991 – not least, the intense economic connections between China and the rest of the world. Relations are chilly, but not as frigid as in the heyday of Le Carré’s Smiley.

Nonetheless, we should all regret that two crucial features of the original Cold War do not prevail today: the political consistency that made both sides’ behaviour predictable, at least after the Cuban crisis of 1962; and the nuclear-arms talks and agreements that made both sides’ arsenals transparent. Today’s ‘not-really-Cold-War’ may in fact remain more dangerous than the old post-Cuba Cold War, until these two lessons have been learned by both sides.

The clue to the most important fact about the Cold War is contained in its name: there was no ‘hot’ conflict between the superpowers during that long confrontation. Their ideological differences were fundamental, their suspicions of each other powerful and there were proxy wars aplenty. Nuclear weapons would have meant that if a war between the West and the USSR had taken place, the result could have been the destruction of much of the planet.

Once this was understood, two big things happened: both sides built up their military forces and tried to get ahead of the other in terms of military technology and to seek advantage in all sorts of ways; but also, the two sides negotiated a series of agreements and treaties designed to make this competition more transparent, more constrained and, above all, better understood by both sides.

The clearest point of all was the mutual realisation that the biggest danger lay in a misunderstanding by one side of the actions, motivations, or likely responses of the other. Hence the shared recognition of the need for clarity, even while both sides were engaged in all sorts of lower-level subterfuge, secrecy, and skulduggery.

That requirement remains valid today, even in a world that is more complicated and in which, thanks to satellite and other digital technology, the United States and China know a lot more about each other than did America and the Soviet Union in the early years of the nuclear age. What they don’t understand well is each other’s motivations, ambitions, or likely responses.

Part of the reason is that motivations, ambitions, and responses are matters of psychology, not of statistics or pure military strengths. The other part is that psychology is changeable, especially when political leaderships change.

The harsh reality about the US-China confrontation is that the atmosphere may not be frigid but the stakes are high. Unlike in Stalin’s time this is not an ideological contest, but it is one for regional domination in the Indo-Pacific and thereby for global leadership. This is old-fashioned great-power rivalry, but with nuclear weapons as well as enormously destructive conventional forces.

The contest is centred on Taiwan and the surrounding South China and East China seas, but not limited to it. Taiwan holds historical as well as territorial significance for the Chinese Communist Party: capturing it would represent a belated victory in two wars: the civil war between the Communists and Nationalists that ended in Chiang Kai-Shek’s escape to Taiwan; and the Sino-Japanese war that led to Japan colonising Taiwan in 1895.

Whatever Communist propagandists might say, China’s claim over Taiwan is not strong: the mainland ruled the island only for two centuries up to 1895, far less than, for instance, England ruled over the now-independent Ireland. Yet the island matters for more than mere historical reasons: in strategic terms, China would like to gain maritime control over the whole Western Pacific, which possession of Taiwan would potentially give it.

The United States would very much like to prevent a Chinese seizure of Taiwan, because it would probably destroy all its security alliances in the region, would push it back to Guam, Hawaii and even to its mainland, and would put in danger the free and safe movement of ships, aircraft, and people in what is among the world’s busiest shipping routes.

It would also push a successful democracy, one which America has been supporting for three-quarters of a century, into the hands of a dictatorship and would place America’s long-time allies, notably Japan, South Korea, and the Philippines, in danger. In practice, this would mark the end of the United States as a global power.

The good news is that such a potentially catastrophic war is preventable, chiefly by convincing both sides that the costs of initiating a conflict would be enormous. The Biden administration has rallied a group of countries inside and outside the Indo-Pacific to build up their conventional deterrent forces, and to work to convince China that those forces would be deployed in the event of an attempted invasion. At the same time, China’s own military build-up, in effect, is an attempt to deter America and its allies from intervening.

This is an arms race as well as a contest of deterrence, with defence spending having risen rapidly in China, Taiwan and Japan, with American, Japanese and Australian postures becoming more forward-leaning and deterrent in intent, with logistical networks being reinforced and extended by the Philippines, Japan and the United States, and with military exercises and air- and sea-incursions being used in demonstrative ways by both China and the United States.

A significant part of those competitive efforts at deterrence is also taking place in the form of rival narratives: between the Japanese-initiated slogan of a ‘free and open Indo-Pacific’ and China’s idea of Asian security being just a matter for Asians; between Taiwan as an ‘internal matter’ for the Chinese people to resolve, and Taiwan as both a regional and global concern over stability, human rights and maritime control; between rival interpretations of history and of the weight it carries in international law; between differing conceptions of which powers are disruptors and which are stabilisers.

That contest of narratives is interesting and natural, but also exposes the acute danger of a misunderstanding about the most fundamental and obvious issue of all: that of whether a conflict over Taiwan can be defined and constrained as a local and conventional affair, or whether it needs to be thought of as unavoidably global and potentially nuclear.

The simplest truth about deterrence is that whether you dress it up as being by ‘denial’ or by ‘punishment’, whether you elevate it to the mutually assured destruction threats of the Cold War, in the end it is all about psychology. To cite Lawrence Freedman, the UK’s doyen of war studies, ‘it is all about setting boundaries for actions’; to set those boundaries requires the adversary’s psychology to be affected by your threats and requires the adversary to be convinced that you really intend, and really will be able, to carry out those threats.

Impressive as they are, the inescapable reality about all these deterrence strategies is they cannot be fully convincing. We can never be sure of what effect they will have on the calculations of an adversary whose motivations and perceptions of their own interest we cannot be sure of.

This applies most clearly to the calculations of China and to the thinking of a long-serving dictatorial leader such as Xi Jinping, who, like Vladimir Putin, is steadily surrounding himself with loyalists while centralising power in his own hands. It also applies to Chinese calculations about America: efforts to gauge the priority ascribed to defending Taiwan, and assessing the political will held by whoever is in the White House at a moment of crisis, can only ever be educated guesswork. Moreover, this is not a danger belonging to a particular moment in time: as far as we can tell, it is a perpetual one.

In that sense, but only that sense, we are back in the conditions of the first decades of the Cold War, a time of mutual incomprehension and suspicion. In today’s world, this lack of mutual understanding, or at least of the clear communication that facilitates such understanding, gains particular importance if we start to think about the role of nuclear weapons in a potential US-China conflict.

As during the Cold War, it is possible to conceive of such a conflict remaining one that uses only conventional weapons, and it can even be argued that the mutual possession of nuclear arsenals might lead both sides to seek to keep any conflict limited, even in the case of quite a protracted war. But should these possibilities be relied upon, given the consequences of the conflict turning nuclear?

There are plenty of technical reasons for not relying on this notion: being sure of fighting a so-called limited war with another nuclear power raises all sorts of military difficulties. The far bigger reason, however, is political. Whichever way you look at it, the politics of deciding to fight a limited war against a fellow nuclear power looks perilous.

Who wants to be the political leader who chooses to fight a conventional war that then turns nuclear, destroying much of his or her own country? Who wants to be the political leader who loses a conventional war because of an unwillingness to use nuclear weapons? That leader would certainly think afterwards that it would have been better to have made a clear nuclear threat in advance to stop the war from starting in the first place. It may be conceivable that a future US-China war could remain limited and conventional, but is it likely? And what if that conception proves wrong?

Writing with the hindsight of 2005 but naturally not the foresight of the 2020s, the Cold War historian John Lewis Gaddis averred that:

Robert Oppenheimer… predicted in 1946 that ‘if there is another major war, atomic weapons will be used’. The man who ran the program that built the bomb had the logic right, but the Cold War inverted it: what happened instead was that because nuclear weapons could be used in any new great power war, no such war took place.

The critical deterrence question about a US-China war is that, to paraphrase Gaddis, since nuclear weapons could be used, can the terrible prospect for all sides of that use be turned into an insuperable obstacle to any such war ever taking place, as it was (with hindsight) during the Cold War?

A sea-change in the level of diplomatic and political clarity would be required to get us to that point. As things stand, there is no such clarity about whether whoever the leaders of China and the United States are at the time would be willing to use nuclear weapons in a war that begins over Taiwan, nor, as far as can be discerned, any channel through which a discussion of that issue takes place.

During the Cold War, the achievement of that clarity began, essentially, with the Cuban Missile Crisis of 1962, in which an episode rife with misunderstandings about each side’s intentions towards an island close to one of them gave rise to a process of diplomacy and negotiation.

That process, which began with the Limited Test Ban Treaty of 1963 and passed through the Nuclear Non-Proliferation Treaty (1968), the Strategic Arms Limitation Interim Agreement (1972), and onward into further arms-control agreements, combined hard engagement on the most critical issue between the two superpowers – nuclear weapons, in all their aspects – with a consistent political focus on the behaviour and apparent intentions of counterpart leaders. The consistency and evident existential importance of this engagement also contributed to the generally bipartisan and quite consistent approach to it across changing American administrations.

There were countless bad aspects of the Cold War, but that was the good aspect. Alongside the conventional deterrence strategies, what is needed now is the modern diplomatic and political equivalent of that ‘good aspect’. There need not be a new Cold War, but we need to learn from the old one about the value of such a diplomatic framework. Currently, no such framework exists.

American and Chinese officials and even presidents do meet from time to time, of course. They discuss lots of other matters, such as climate change, trade, and economics, too, and occasionally their military leaders meet. But there are two gigantic holes in the current diplomacy.

The first is that there are no arms-control agreements between China and the United States (or anyone else), covering either nuclear weapons or any form of missiles. China refuses to engage seriously in arms-control talks, perhaps because it perceives itself as the weaker party.

The second is that there has been no formal agreement between China and the United States about Taiwan since the ‘US-China Joint Communique of August 17, 1982’, which was the third in a spate of agreements that began with President Nixon’s historic opening to China in 1972. Yet the world has changed hugely in the four decades since then, including notably that Taiwan has become a democracy and China has transformed itself from a military and economic weakling into a giant on both counts.

In November 2023 the two sides held their first talks about arms control since the time of the Obama administration, but little of substance emerged. This is likely to remain the case unless and until the top leaders – ie, the American and Chinese presidents – succeed in convincing each other that there is something to be gained by putting lower-level officials to work at exchanging information about weapons, missiles and the rest, and exploring the scope for agreements, thus setting in train the sort of process that was a feature of the Cold War.

While China has long insisted on separating discussion of what it calls its ‘core interests’ – in which category Taiwan certainly falls – from discussion of all other matters, including strategic stability, the hope must be that eventually this will change if successive American presidents succeed in conveying convincing messages that what President Biden said on four occasions in 2021 and 2022 is now a firm and consistent US policy: that it would intervene militarily in an attempted invasion or blockade of Taiwan, with the implication that in the conflict that would ensue, the use of nuclear weapons cannot be ruled out.

Most likely, China will never accept publicly that Taiwan is anything other than ‘an internal matter’ and will seek to preserve the hope that reunification could either be peaceful or achieved by conventional means. Pure self-interest will start to dictate at least a private or implied realism about the need for dialogue with the United States the more that it becomes clear that a Taiwan conflict holds an unavoidable risk of turning nuclear.

We are all deterrers now, thanks to the dreadful exemplar of the Russian invasion of Ukraine. But only once it has become clear, consistently so, that the most important level of deterrence, namely the nuclear sort, has taken its rightfully primary position in strategic thinking and in diplomatic dialogue in the Indo-Pacific, will we be able to feel reassured.

Currently, war over Taiwan and hence between China and the US, can be described as ‘possible, avoidable, but potentially catastrophic’. This author will sleep more easily once it becomes describable by all sides as ‘inevitably catastrophic and therefore inconceivable’.

Bill Emmott, a former editor in chief of the Economist, is chair of the International Institute for Strategic Studies and author of ‘Deterrence, Diplomacy and the Risk of Conflict over Taiwan’ (IISS/Routledge, July 2024), on which this article is based.