The Nuclear Taboo Is Beginning to Fracture
For eighty one years, no nuclear weapon has been used in war. Now arsenals are being modernised and political and military leaders are once again considering circumstances in which the bomb might actually be used. The danger lies not only in the weapons themselves, but in the return of an idea the world had spent generations trying to suppress.
A nuclear war would continue long after the last missile had landed. The first hours would bring the destruction familiar from Hiroshima and Nagasaki, magnified by weapons many times more powerful: cities consumed by blast and fire, hospitals destroyed, electricity and communications severed, populations exposed to radiation. What followed could be quieter and more extensive, spreading far beyond the countries that had gone to war.
Researchers studying nuclear winter have tried to understand that second catastrophe. Cities and industrial areas burning simultaneously could send immense quantities of soot into the upper atmosphere, reducing sunlight and temperatures around the world and shortening growing seasons at precisely the moment when transport, trade and governments were struggling with the destruction left by the war.
A major study published in Nature Food in 2022 modelled several possible nuclear conflicts. In one large regional war, researchers estimated that disruption to food production could eventually lead to more than two billion deaths. In their most extreme scenario, global crop calories fell by around 90 per cent three to four years after the fighting, creating the possibility of famine on a scale modern civilisation has never experienced.
The figures are models, not predictions, but they make one point difficult to escape. A government can choose where a nuclear weapon explodes. It cannot decide where the consequences end. A war begun over territory or political power could eventually kill people thousands of miles away who had nothing whatever to do with the dispute.
And it is against this background that nuclear weapons are becoming imaginable again.
Earlier this month, former Republican congresswoman Marjorie Taylor Greene alleged that high level discussions were taking place in Washington about the possible use of nuclear weapons against Iran. The allegation has not been independently substantiated, but its appearance in mainstream political discussion is itself remarkable: the possibility that an American president might contemplate nuclear use during a conventional war no longer sounds as though it belongs exclusively to the nightmares of the Cold War.
For generations, that possibility was constrained by something more than military calculations. Nuclear weapons acquired a stigma unlike that surrounding conventional arms. Governments could possess them, point them at one another and threaten retaliation with them, but their actual use came to occupy a different moral and political category. The political scientist Nina Tannenwald called this the nuclear taboo.
Its power depended partly upon something very simple. Nuclear weapons were understood to be unusable because nobody could confidently know what would happen after the first one exploded.
That understanding is beginning to weaken.
Making the unthinkable thinkable
The change is not that governments suddenly want nuclear war. It is that they are increasingly preparing for circumstances in which nuclear weapons might have a role within a war.
The Stockholm International Peace Research Institute reported this summer that nuclear armed states are placing greater emphasis on their arsenals as instruments of national power. All nine nuclear armed countries are modernising their forces. At the beginning of this year the world possessed an estimated 12,187 warheads, around 4,000 deployed with operational forces and more than 2,100 maintained on ballistic missiles at high operational alert.
For much of the period after the Cold War, those weapons remained largely outside public consciousness. Terrorism, insurgencies, cyberwarfare and conventional conflicts occupied governments. Nuclear weapons remained aboard submarines and inside silos, but the fear that once surrounded them faded.
Now the vocabulary is returning, although in a different form.
The most revealing word may be tactical. It is an oddly reassuring adjective to place before the most destructive class of weapon humanity has invented. A tactical nuclear weapon sounds smaller, more precise and more controllable than the enormous strategic warheads associated with the destruction of cities.
There is a military logic behind this. If an adversary believes that the only possible nuclear response to an attack would be an apocalyptic exchange, it may doubt that any political leader would order one. A smaller nuclear option makes the threat more credible and, according to its advocates, strengthens deterrence.
But a weapon made more credible as a threat also becomes more credible as something that might actually be used.
Pope Leo XIV warned earlier this year that tactical and increasingly miniaturised nuclear weapons risk making nuclear use appear more viable and therefore less improbable. Once governments begin asking whether a smaller weapon could destroy an underground installation, halt an army or demonstrate resolve without provoking a wider exchange, the question has changed. It is no longer simply how to prevent nuclear war, but whether a particular nuclear war might be contained.
No one knows.
A government ordering a single strike might intend it as a warning. The government receiving it would have no way of knowing whether another was coming. Forces might be dispersed and weapons prepared for launch before they could be destroyed, movements that could themselves be interpreted as preparations for retaliation. The distinctions made in planning rooms between tactical and strategic, limited and general, would suddenly depend upon frightened people making irreversible decisions with incomplete information.
The illusion of control
Technology encourages another dangerous possibility: that nuclear war is becoming more manageable.
Satellites provide extraordinary surveillance, conventional weapons can strike with increasing precision and artificial intelligence can process military information faster than human analysts. These technologies may improve warning and prevent mistakes, but they can also create confidence that events can be understood and controlled more precisely than before.
The problem is that there has never been a nuclear war between nuclear powers. There is no historical experience showing how governments would behave after the first explosion, whether a “limited” strike would be understood as intended or whether leaders confronted with the possibility of losing their remaining weapons would wait to discover what came next.
That uncertainty is the reason the nuclear taboo mattered.
For eighty one years, nuclear powers have fought wars, suffered defeats and confronted one another without using the bomb. The record is extraordinary, but it is a record of the past rather than a guarantee about the future.
If the taboo breaks, it is unlikely to happen through some dramatic announcement that nuclear weapons have become acceptable. It will happen more quietly, as the language surrounding them changes. A weapon becomes tactical, a strike limited, a response proportionate and escalation manageable. Each qualification makes the bomb sound slightly less like the end of politics and slightly more like another instrument of it.
Yet the world beyond the planning room would recognise none of those distinctions. Missiles might fly for minutes while their consequences lasted generations, cities burning in one part of the world while harvests failed in another. People who had never heard of the quarrel that began the war could die thousands of miles from its battlefields.
That understanding created the nuclear taboo. The danger now is that we may be forgetting why we learned to fear the bomb before we are forced to remember.