When the atomic bomb was first revealed to the world, through the destruction of two Japanese cities, there was immense speculation about what this new invention would portend for global affairs.

Several of the scientists who had contributed to its creation had been weaned on the science fiction of H.G. Wells, whose books imagined that a super-weapon wielded by an enlightened bomber corps could be used to create a global police state, and enforce a global peace.

Many of the statesmen involved considered it a boon for American power and hegemonic interests: here was a uniquely American weapon, created by American scientific and industrial might (augmented, of course, with some international cooperation and refugees from Nazism, such as Enrico Fermi and Niels Bohr) that could serve to be the basis of a new Pax Americana.

But it didn’t work out that way. It turned out that building up a reliable supply of nuclear weapons took a non-trivial amount of effort. Atomic bombs were, for the last half of the 1940s, rare, costly, and still fairly crude. Furthermore, the President of the United States, Harry Truman, was simply not that keen on using them. He kept almost the entire stockpile out of military hands throughout his presidency, fearing that if he did not, they would use them rashly. When the military implored him to give them access to them in 1948, he shot them down: ‘I don’t think we ought to use this thing unless we absolutely have to. It is a terrible thing to order the use of something like that, that is so terribly destructive, beyond anything we have ever had. You have got to understand that this is not a military weapon.’

If the atomic bomb is not a military weapon, then what is it? It is a weapon of politics, perhaps: a weapon whose primary value derives from what happens when you merely threaten to use it. This is how the strategy of nuclear deterrence evolved. Nuclear weapons might have tremendous political power, but you lose that the minute they start being actually used in battle. The threat of nuclear attack might deter an enemy from doing something you don’t want them to do (like invade a neighbour), or you might try to use it to compel them to do something you would like them to do.

Paradoxically, the threat needs to be ‘credible’ for it to have any weight (the enemy has to feel like an attack is a realistic possibility), but invoking the threat often and frivolously reduces its credibility (because nobody believes you are going to start nuclear war over minor disagreements).

The American nuclear monopoly turned out to be very short-lived, in any event. By 1949, the Soviet Union had tested its first atomic bomb, and over the course of the 1950s and 1960s several more states joined the ‘nuclear club’. The US poured trillions of dollars into developing new categories of warheads, such as hydrogen bombs, or thermonuclear weapons, as well as new methods of delivering them to their targets — silo-launched missiles, truck-launched missiles, submarine-launched missiles, big bombs, little bombs, air-launched missiles, nuclear torpedoes, nuclear bazookas, nuclear landmines, and so on. They were dispersed into bases on foreign soil, and put on to boats and under the seas, ready to be used at a moment’s notice.

And there were always some who thought: yes, they should be usable. What’s the point, otherwise? What are they good for?

But the politicians always came to the conclusion that the costs of using nuclear weapons, whether in the form of a risk of nuclear retaliation, international outcry, or moral hazard, would not be worth whatever military gains might hypothetically be achieved by them. And there were those who whispered that no country benefited from a high-bar to nuclear weapons use more than the United States, because we put our cities on the coasts, centralised our military into easy-to-target bases, and we, in some sense, had the most to lose, in terms of population, economy, and martial strength.

Over time, the rhetoric of American nuclear strategy has, by and large, hardened around a simple idea: that the existence of these weapons is justified by their deterrence value. That they work by making our enemies hesitant to use their own nuclear weapons, or to cross other invisible red lines that might be deemed too risky. And that even efforts that might make nuclear weapons seem more usable, like new low-yield submarine missile warheads ordered by the Trump administration, are really just there because we want to make these threats seem more credible, by adding more possible responses between ‘nothing’ and ‘Armageddon’.

But the Russian invasion of Ukraine has caused some interesting and subtle reframing of how nukes are being talked about in the United States. For one thing, there has been a perceptible increase in fears of a Third World War breaking out. There have been twenty-first century nuclear threats before — lest we forget, the North Korean crisis of 2017-2018, which probably came closer to some kind of nuclear weapons use than we yet have public knowledge of (what exactly were the serious thoughts inside the heads of Donald Trump, Kim Jong Un, and their closest advisors during that tense time?) — but the bold atrocity of the Russian attack against Ukraine, and its proximity to NATO nations, and the reaction from NATO and the United States, has intuitively brought home to millions the possibility of what the theorists call ‘escalation’, and on a scale not contemplated since the end of the Cold War. The standard response from nuclear experts about the risk of nuclear war today, up until this year, was always along the lines of: there are many nuclear threats today, and some of them are perhaps more complex than the threats that emerged during the Cold War, but at least the odds of a war between the United States and Russia seem lower than in the past. Now, perhaps not.

But it’s not just fear that I see being expressed; there’s another emotion, a more surprising one: frustration. There’s a sense that nuclear weapons are stopping the United States, and NATO, from helping Ukraine against the Russian attacks. That Russian nuclear weapons put limits on peacekeeping actions, such as imposing a No-Fly Zone over Ukraine. As the former top NATO commander General Philip Breedlove put it in April: ‘We are deterred.’

Deterrence has always been a two-way street in principle. The framing of Mutual Assured Destruction, one variety of deterrence that emphasises mutual vulnerability as a path to avoiding nuclear war, always made this extremely clear. But the idea of MAD was never popular with the military, and the idea of embracing vulnerability still sends a chill up most people’s spines. Even the strategists may speak coolly of the value of mutual vulnerability as a vehicle for stability, but you can see them bristle under its implications. There’s a logic to deterrence, but it is always coupled, in the end, with raw terror. And one has to have a lot of faith in the whole system working to feel comforted by such a vulnerability, and when these systems ultimately depend on the psychologies of a small number of people at the very top of their hierarchies, it is very easy to lose faith.

It is more comforting to most people to emphasise the form of deterrence that applies to avoiding the things one is afraid of — it keeps our enemies deterred — without thinking about the fact that the weapons keep us deterred as well. It is also comforting to hope that some kind of deus ex machina will resolve the problem. Maybe missile defences will save us? They won’t; even if they worked as designed, which is hardly clear in a real-life threat scenario, they aren’t designed to shoot down weapons of the volume or sophistication that Russia has at its disposal. Maybe all of the Russian nukes will be duds? Not a proposition worth betting the lives of millions on. Maybe the Russian military, or its oligarchs, will just forcibly retire Putin? Who knows, but don’t get your hopes up. Putin has spent 20 years securing his position. Maybe Putin is actually deathly ill? It couldn’t happen to a nicer guy. But, again, don’t get your hopes up.

At the other extreme, every time the fear of nuclear war goes up — this happened with the North Korean crisis as well — one starts to see a lot of nervous people suggesting that it might be worth the risk, if just to resolve the tension once and for all. Which is so obviously short-sighted that it’s hard to take entirely seriously. When people who ought to be more knowledgeable and responsible occasionally voice such thoughts, they are perhaps playing out the fact that since the end of the Cold War, most Americans have not given much thought to the nuclear threat, even though it never really went away.

This shift towards considering nuclear weapons as something holding us back might be a profound one. It’s a sign of two things. The first is that the US feels, probably dangerously, in a very strong position vis-à-vis Russia. It’s easy to see why: the Russian military has not, thus far, inspired much awe with their attack on Ukraine, having by some estimates lost more soldiers in the first months of the invasion than the USSR lost during the decade-long Soviet-Afghan War. If their attrition rates keep up this pace, they will within the year lose more soldiers than the US did in Vietnam, Iraq, and Afghanistan.

This is not the feared Red Army that led to the creation of NATO in the first place; it is, at least by many Western news accounts, an army of unwilling soldiers, of outdated rations, of easily-hit supply lines, and of sitting ducks. And they are fighting against an extremely motivated Ukrainian military which has been training for this very sort of conflict for years, and is being supplied with advanced weapons by NATO members, including the United States. Russia may still yet win this war — their military budget dwarfs Ukraine’s, as does the total number of troops they can commit — but the damage to their military reputation will persist.

For Americans who have been bruised by the experience of our Forever Wars, there appears to be a sense that maybe, now, we could actually apply our military spending on an enemy that seemed worth it, both morally and militarily. Except for the nukes — the nukes are still there. They check American hegemonic ambitions in a way that is probably for the best (for we have seen how astray such ambitions can go unchecked), but it can be hard to feel that when one sees the reports of the butchery inflicted against Ukraine. One wants to do something, and we are, but the Russian nukes keep us from doing enough.

There is still much danger here. There is no rule written into stone that says Russia will not decide to escalate further, or to even bring in some kind of nuclear weapon use. Anyone who insists categorically that such a thing could not happen is talking about their hopes and dreams, not possible futures. The decision as to whether to bring nuclear weapons into this conflict are going to be made by a very small number of human minds, potentially just one (Putin’s), and while it is possible to predict human behaviour in the aggregate fairly well, predicting individual choices is pretty hard. Putin is, as the experts say, a ‘rational actor’, but all they mean by that is that they don’t think he is suicidal. The problem with rationality, in general, is that it depends on your axioms: what you think the facts of the world are. Clearly for Putin, some of the facts that led him to invade Ukraine were based on falsehoods — it has not been a walk in the park at all. It is entirely possible that he believes the absurdities he asserts to justify his actions. With such a world view, who knows what he is capable of? Is a fear of our nukes holding him back, too? Who can say? Who can get inside his head?

So, what are nukes good for? They constrain us, within limits. They enable dictators, within limits. Those limits matter, whether they are the limits of what we would do to them if we could, or what they could do to us if they could. But as frustrating as those limits might be, it is worth considering the evils that would be inflicted should they fail, and the nukes are suddenly able to do the one technical thing they are actually good for: destruction.