Is it possible to “win” a nuclear war?

Following their first meeting in Geneva in 1985, US President Ronald Reagan and Soviet leader Mikhail Gorbachev issued a historic joint statement stating their shared belief that “a nuclear war cannot be won and must never be fought.”

The maxim lived on. The Geneva summit turned out to be a key milestone in the beginning of the end of the Cold War arms race. Nearly four decades later in 2022, leaders of the world’s five main nuclear powers — the US, Russia, China, France, and the UK — issued another joint statement, affirming that “a nuclear war cannot be won and must never be fought” and that their arsenals are meant to “serve defensive purposes, deter aggression, and prevent war.”

The thinking behind the phrase is that these weapons are so destructive — with potential consequences that include the literal destruction of human civilization — that it makes no sense to talk about “victory” in a nuclear war.

It’s a powerful idea. But do the nuclear powers really believe it?

As the world marks the 80th anniversary of the bombing of Hiroshima and Nagasaki this week, it’s clear that the world is entering a new nuclear age, characterized by increasing tension between superpowers, China’s growing arsenal, and the rising possibility that more countries will acquire the bomb.

And judging from the nations’ actions and strategy documents — as opposed to their declarations at summits — we are also in an era in which nuclear powers do believe they can win a nuclear war and want to be prepared to do so.

Recent years have seen threats of Russia using a “tactical” nuclear weapon in Ukraine and a military conflict between India and Pakistan that US officials believed could have gone nuclear. The governments making these threats aren’t suicidal; if they were contemplating nuclear use, it’s because they thought it would help them win. In response to growing threats, the United States has been updating its own doctrine and arsenals to provide more options for a so-called limited nuclear war. Looming over it all is the danger of war between the US and China, a conflict that would be fought under the nuclear shadow.

The idea that there can be a winner in a nuclear exchange rests on several assumptions: that the conflict can be contained, that it won’t inevitably escalate into an all-out exchange that sees whole cities or countries wiped out, and that there will be anyone left alive to claim victory.

Some experts claim that as long as the potential for nuclear war exists, we’d be foolish not to plan for how to win one as quickly and with as little destruction to ourselves as possible. Others say the idea that a nuclear war could be kept “limited” is a dangerous notion that only makes such a war — and the risk that it could escalate to something not so limited — more likely.

The bombings in Hiroshima and Nagasaki killed somewhere between 100,000 and 200,000 people, depending on estimates, but both cities are once again thriving metropolises today. Despite the fears of some of the scientists involved in developing the bombs, they did not ignite the atmosphere and kill all life on Earth. They did play a significant role — though there continues to be a debate about just how significant it was — in ending World War II. The only time nuclear weapons were used in war, the side that used them won the war.

But the difference then was that only one country had the weapons. Today, there are nine nuclear-armed countries with more than 12,000 nuclear weapons between them, and most of those are far more powerful than the ones used on Japan in 1945. The W76 warhead, the most common nuclear weapon in the US arsenal, is about five times more powerful than “Fat Man,” dropped on Nagasaki.

When most people imagine what a war using these weapons would look like, images of armageddon — annihilated cities, radiation fallout, nuclear winter — come to mind. Popular depictions of nuclear war, from Dr. Strangelove to the Terminator movies to last year’s chilling quasi-novel Nuclear War: A Scenario, soon to be adapted into a film, tend to focus on the worst-case scenarios.

The apocalyptic possibilities have, for decades, motivated global campaigns to ban nuclear weapons and haunted many of the world leaders who would have to make the decisions that would set them in motion. That includes Donald Trump, who has described what he calls “nuclear warming” as the “biggest problem we have in the whole world.”

If there could be a silver lining to the fact that humanity has built weapons capable of destroying itself, it’s that this fear has made those weapons much less likely to be used. “Mutually Assured Destruction” (MAD) has never actually been officially US policy — the RAND Corporation analyst who popularized the term back in the 1960s meant it as a critique — but nonetheless, the idea that nuclear war would be suicidal for both sides is arguably what kept the Cold War from getting hot. The logic continues to operate today: Joe Biden preemptively ruled out responding with direct military force to Russia’s invasion of Ukraine because of the potential consequences of war between the two countries that account for 90 percent of the world’s nukes.

But from the earliest days of the nuclear era, there have been prominent voices arguing that nuclear war could be kept within limited boundaries, and that it’s worth preparing to win one.

In the mid-1950s, President Dwight Eisenhower’s administration operated under a nuclear strategy that emphasized “massive retaliation,” meaning the US would respond to any Soviet attack with overwhelming nuclear force against Soviet territory. But Henry Kissinger — who at the time was a Harvard professor and up-and-coming security analyst, and later went on to become secretary of state and national security adviser — argued against “massive retaliation,” lamenting that “far from giving us freedom of action, the very power of modern weapons seems to inhibit it.” He wanted options between refraining from nuclear use at all and all-out annihilation. In 1956, Kissinger argued that the US should instead plan for fighting a “limited” nuclear war by emphasizing the development of lower-yield weapons and devising “tactics for their utilization on the battlefield.”

Herman Kahn, the RAND Corporation nuclear strategist who was one of the inspirations for Stanley Kubrick’s Dr. Strangelove character, envisioned a 44-rung escalation ladder for nuclear conflict, with what he called “barely nuclear war” kicking in at rung 15 and getting more serious from there.

If MAD stood for the idea that the only two options were avoiding nuclear war or global annihilation, the view that nuclear weapons could be used selectively with devastating but limited consequences came to be known as NUTS, or Nuclear Utilization Target Selection.

The debate never really went away, but it faded somewhat with the end of the Cold War when both the US and Russia substantially reduced their arsenals, and the risk of confrontation appeared to fade. Recently, however, the topic of limited nuclear war has been making a comeback.

“We have nine nuclear powers in the world today that are building nuclear weapons, not to put in museums, but for military and political use, and developing plans for their use,” Matthew Kroenig, a national security analyst at the Atlantic Council and Georgetown University, told Vox.

The United States is no exception. The 2018 US Nuclear Posture Review, issued under the first Trump administration, called for “expanding flexible U.S. nuclear options.” The 2022 review, issued under the Biden administration, included similar language. To provide those options, the US has begun production of a number of new lower-yield nuclear warheads, such as the 5-kiloton W76-2, which has been deployed on nuclear submarines. For reference, that’s about a third as powerful as the bomb dropped on Hiroshima, but more than a 1,000 times more powerful than the “massive ordinance penetrator” bomb the US recently used on Iran’s nuclear facilities.

Advocates for limited nuclear war planning are on the ascendance as well. Elbridge Colby, the current undersecretary of defense for policy, has attracted attention for advocating a shift in military priorities away from Europe and the Middle East toward what he sees as the more pressing threat from China. He’s also a leading advocate for preparing for limited nuclear war. In a 2018 article for Foreign Affairs, Colby argued that deterring Russia or China from using force against US allies requires developing the “right strategy and weapons to fight a limited nuclear war and come out on top.”

These advocates say that recent actions by America’s adversaries make it necessary to plan for fighting a limited nuclear war. US officials believe that Russia’s military doctrine includes a so-called escalate to de-escalate strategy, in which it would use a nuclear strike or the threat of one to force surrender, to compensate for disadvantages on the battlefield or to avoid an imminent defeat.

Russia’s war plans are classified, and some analysts are skeptical that such a strategy exists, but an example of the kind of thinking that keeps American strategists up at night is laid out in a 2023 article by Sergei Karaganov, a one-time adviser to President Vladimir Putin and one of Russia’s leading foreign policy commentators. Karaganov argues that Russia has “set too high a threshold for the use of nuclear weapons,” and that in order to prevent further US meddling in Ukraine, Russia needs to demonstrate its willingness to use a nuclear weapon. He reassures readers that nuclear retaliation by the US to protect a faraway ally is unlikely, and that “if we correctly build a strategy of intimidation and deterrence and even use of nuclear weapons, the risk of a ‘retaliatory’ nuclear or any other strike on our territory can be reduced to an absolute minimum.”

Obviously, Putin hasn’t done this yet in Ukraine, though he has made repeated threatening references to his country’s arsenal, and at one point, in 2022, Biden administration officials reportedly believed there was a 50-50 chance Russia would use a nuke.

Russia is believed to have an arsenal of more than 1,000 “tactical” or “nonstrategic warheads.” (The distinction between “tactical” and “strategic” nuclear weapons is a little vague. The former refers to weapons meant to destroy military targets on the battlefield rather than target an enemy’s cities and society. Tactical nukes are generally smaller and shorter range, though some are larger than the bombs dropped on Japan, and some observers — including former Secretary of Defense James Mattis — have argued that there is no difference between the two.)

The US has also accused Russia of developing capabilities to deploy a nuclear weapon in space, which could be used to destroy communications satellites in orbit. This would be a less catastrophic scenario than a detonation on Earth, to be sure, but still a dangerous new form of nuclear escalation. (Russia has denied the American allegations.)

Unlike Russia and the United States, China has an official “no-first use” policy on nuclear weapons. But the country’s arsenal is growing rapidly, and many experts suspect that in an all-out military conflict, particularly if the war were going badly for China and its conventional forces were threatened, its threshold for nuclear use might be lower than official statements suggest.

The argument from some strategists is that ruling out nuclear use entirely gives China an incentive to escalate to the point where the US backs down.

“If we are completely convinced that a limited war is impossible, and the Chinese believe that it is possible, then they will checkmate us every time,” Colby told me in a 2022 interview for Grid. “At some point, we have to be willing to fight a war under the nuclear shadow. My view is [that] the best way to avoid testing that proposition, which I absolutely don’t want to do, is to be visibly prepared for it.”

On the other hand, Chinese planners can think this way too. Lyle Goldstein, a professor at Brown University who studies Chinese military strategy, says that “Chinese scholars are talking openly about limited nuclear war now,” which they have not in the past. But when confronted about this shift by Americans, they tend to make the argument, “We’re discussing it because you’re discussing it.”

It’s not only the world’s top three nuclear powers that engage in this sort of thinking. Pakistan’s nuclear doctrine, also classified, is thought to emphasize “calibrated escalation” to deter strategic surprise by its rival, India. During the recent military conflict between the two countries in May, fears of nuclear escalation are reportedly what prompted the Trump administration to intervene diplomatically, after initially suggesting it was not a core US interest.

Since acquiring nuclear weapons, the two South Asian adversaries have proven adept at managing military escalation and de-escalation without letting things spiral out of control. But this was the most intense conflict between the two in years, and after it ended, Prime Minister Narendra Modi vowed that India would no longer succumb to Pakistan’s “nuclear blackmail,” suggesting that his country’s tolerance for nuclear risk was growing higher.

Advocates for preparing for limited nuclear war say the attention devoted to full-scale global thermonuclear war distracts us from the sort of war that we’re much more likely to get into.

“Any use of nuclear weapons in the future will be limited. There’s virtually no prospect whatsoever of a global thermonuclear conflagration,” said Kerry Kartchner, a former State Department and Pentagon official and coauthor of a book on limited nuclear war.

The most likely way a war would stay limited is if one side simply decided not to fight. “There is a very, very strong, very powerful incentive not to use nuclear weapons,” even when the other side uses them first, Kartchner told Vox.

In his book The Bomb, journalist Fred Kaplan reports that during the Obama administration, the National Security Council held a series of war games simulating the response to a hypothetical use of tactical nuclear weapons by Russia during an invasion of the Baltic countries. Officials differed sharply over whether the US should respond with a nuclear strike of its own or keep its response limited to conventional military and economic means in order to “rally the entire world against Russia.”

Years later, when President Biden believed a real-world version of this scenario could be imminent, he declined to say how he would respond. Kroenig, of the Atlantic Council, has argued that the US should respond to Russian nuclear use with conventional force. But he also believes that even if the US used nuclear weapons to respond, it could keep the conflict limited.

“You can signal through the use of military force,” he said. “I think Russia understands the difference between a low-yield battlefield nuclear weapon going off on the battlefield versus a big ICBM heading towards Moscow.” He concedes that this type of signaling wouldn’t work with a “true madman,” but argues, “in most real-world cases, leaders don’t rise to run major countries without having some kind of ability to think rationally and to preserve their own survival.”

Others aren’t so sure. “Whenever somebody says, ‘we can control escalation,’ they immediately assume a whole bunch of things that seem unrealistic to me, like perfect information, calm, rational decision makers,” says Jeffrey Lewis, an expert on nonproliferation at Middlebury Institute of International Studies.

From Napoleon to Hitler, history is rife with examples of leaders making military decisions that led to the destruction of their regimes. Putin believed the war in Ukraine could be won in a matter of weeks and that the international response would be far more limited than it turned out to be.

There’s also no guarantee that adversaries would be able to communicate effectively during a nuclear crisis. During the 2023 incident in which the US downed a Chinese spy balloon that had drifted over US territory, Secretary of Defense Lloyd Austin reached out to his Chinese counterpart, Wei Fenghe, to explain US attentions and calm tensions, but Wei didn’t pick up the phone.

An infamous 1983 Pentagon war game known as Proud Prophet, simulating a US-Soviet nuclear war in Europe, provides a sobering warning: As the strikes between the two sides escalated, they were unable to communicate their intention to keep the conflict limited. “When we hit the Soviets, they hadn’t the slightest idea of what our limitations were,” one participant recalled. By the end of the game, Paris, London, Amsterdam, Rotterdam, and Brussels — and every major German city — were destroyed.

Including tests, there have been more than 2,000 nuclear detonations since 1945. One, or even a few more, will not literally be the end of the world, but there’s limited margin for error. In a 2007 study, a group of physicists estimated that a limited regional nuclear exchange “involving 100 15-kiloton explosions (less than 0.1% of the explosive yield of the current global nuclear arsenal)” could “produce direct fatalities comparable to all of those worldwide in World War II” as well as causing enough smoke to rise into the atmosphere causing “significant climatic anomalies on global scales.”

When it comes to nuclear wars, even limited ones, “You might be able to survive the first one or two,” said Manpreet Sethi, a nonproliferation expert at India’s Centre for Air Power Studies. “But after that, we’ll be pushing the envelope. It can’t be business as usual after you’ve done a ‘little bit’ of nuclear war.”

Advocates for limited nuclear war planning argue that by ruling it out entirely, the US is inviting adversaries like Russia and China to use their nukes without fear of retaliation.

Sethi’s concern is that “If you start preparing for a limited nuclear war, you increase the likelihood of fighting a war like that because you get into the idea that escalation management is possible.”

For now, the example of Ukraine and Putin’s failure to follow through on his threats suggests that the taboo against nuclear use — no matter how “tactical” or “limited” — remains in place.

“The important lesson from this war is that nobody really has confidence that escalation can be contained, said Pavel Podvig, an expert on Russia’s nuclear forces at the United Nations Institute for Disarmament Research in Geneva. Encouragingly, Biden administration officials say they believe China may have warned Russia against using its weapons, suggesting this may be a red line even for Moscow’s backers.

This year’s Hiroshima anniversary is a moment for somber reflection on the risks humanity has put itself under. But a more optimistic view is that the world is also marking 80 years without any other country actually using these weapons, something many leaders would not have predicted at the dawn of the nuclear age. As armed conflicts continue to proliferate, longstanding arms control treaties fall by the wayside, and the number of nuclear-armed powers continues to grow, getting to the 100th anniversary with that record intact may prove even more challenging.