Thanks for reading! The best way to make sure that you don’t miss any of these conversations is to subscribe to The Good Fight on your favorite podcast app.
If you are already a paying subscriber to Persuasion or Yascha Mounk’s Substack, this will give you ad-free access to the full conversation, plus all full episodes and bonus episodes we have in the works! If you aren’t, you can set up the free, limited version of the feed—or, better still, support the podcast by becoming a subscriber today!
And if you are having a problem setting up the full podcast feed on a third-party app, please email our podcast team at leonora.barclay@persuasion.community
Mark Lynas is an author, campaigner and speaker. He is science advisor with the Climate Vulnerable Forum and policy lead with WePlanet, a pro-science environmental advocacy network active in over 20 countries. His latest book is Six Minutes to Winter: Nuclear War and How to Avoid It.
In this week’s conversation, Yascha Mounk and Mark Lynas discuss how likely a nuclear war is, what a nuclear winter would look like, and how to avoid it.
This transcript has been condensed and lightly edited for clarity.
Yascha Mounk: You always have cheerful topics in store for us—from climate change to nuclear war. Nuclear war is one of those things we all learn to worry about at some point in childhood. It’s clearly this strange, looming danger hanging over humanity. And yet, it’s usually in the background of discussions. Not since the Cuban Missile Crisis has it really been at the absolute forefront of the news. Where would you rank the risk of nuclear war relative to the other big existential risks facing humanity?
Mark Lynas: Wow, there’s a lot in that. I agree that thinking—or worrying—about nuclear war feels a bit retro. I grew up with nuclear war being the thing—we were all ready to die at pretty much any time—until the Berlin Wall came down and suddenly the Cold War was over. Suddenly, the risk of nuclear Armageddon seemed to have disappeared completely. We were friends with Russia, and people stopped worrying about it. I think people were very glad to be able to drop it from the list of existential risks they had to spend any time worrying about. And then, of course, climate change came. I’ve contributed to the sense of existential crisis attached to climate change through some of my writing. My book Six Degrees talked about climate as very much a catastrophic risk, not just to humanity but to the planetary biosphere. So that then became the preeminent concern of this generation—certainly of younger people. You had Greta Thunberg, and you had the Fridays for Future movement. That generation grew up with climate as their existential risk.
Of course there are others. There are books about which existential risks are greater than others—whether it’s AI, pandemics, asteroid collisions, or whatever. We can get into ranking them, if you like. I actually had a piece in The Wall Street Journal recently, co-authored with Ted Nordhaus from the Breakthrough Institute, making the point—as gently as we could—that climate is an existential risk of a different nature from nuclear war. The impacts of climate change manifest over decades or even centuries—you’re talking about the melting of the large ice sheets and multi-meter sea level rise, and so on. Whereas, of course, as we all know, nuclear war can destroy the biosphere in pretty much half an hour. It doesn’t take very long to get ICBMs in the air and destroying the major cities of the Northern Hemisphere.
Mounk: One of the things that’s interesting here—and it’s a gruesome thing to talk about, but I think it’s actually important to be intellectually rigorous about it—is that the distribution of outcomes seems, at least intuitively, to be somewhat different. Climate change is a very serious risk to humanity. I think the likelihood that it leads to a significant number of deaths, to a significant number of people’s lives being impaired, is perilously high. All of those are very good reasons to worry about it and to make the fight against it a priority. But it is harder for me to see a scenario in which climate change actually destroys humanity—or even destroys, broadly speaking, our level of technological civilization. It’s easy to imagine hundreds of millions of people dying—which is obviously a very, very bad thing—but it’s harder to see how climate change leads to either the death of 99% of humanity or to us being unable to sustain the level of technological advancement we currently have as a species.
Whereas the risk of nuclear war is perhaps lower in the sense that a bad outcome is less likely. You might disagree with that, but I think it’s easier to imagine a world where we somehow get through the 21st century without a nuclear war and there’s just no immediate downside from having those weapons. But if the negative case is activated, it’s much easier to imagine a scenario in which the vast majority of humanity dies, in which we’re basically back in the Stone Age.
So perhaps we can start to dig a little bit into that. What would it look like if we had a global nuclear war anytime soon? Say there’s a nuclear confrontation between the United States and China that somehow also draws in other nuclear powers—a worst-case scenario. Is that the end of humanity? Is that the end of our technological civilization? What would that mean for the planet and for humans?
Lynas: I’m a student of mass extinctions. I’ve covered them a lot in my book—the “big five” geological mass extinctions that have happened since the Cambrian, over the last half-billion years. There’s a 100% probability of climate change happening, because it’s already happening. Whereas nuclear war may never happen. So it’s a different thing altogether. The "kill mechanisms"—to use the mass extinction terminology—aren’t really there for climate change. There are many more ways to adapt to slow-onset events, which are the kinds of impacts most likely to result from climate change.
Whereas with nuclear war, the kill mechanisms are very obvious. It’s not just the explosions, the blasts, and the burning. It’s the nuclear winter. That’s really the central kill mechanism—one that could certainly destroy civilization, possibly destroy humans as a species, and almost certainly destroy the majority of life in the biosphere.
Mounk: I feel like “nuclear winter” is the kind of term that everybody has probably heard, but that many people have trouble fully understanding—both in terms of the mechanisms that lead to a nuclear winter, and what that would actually mean. So if you have a significant number of nuclear weapons deployed between two superpowers in the world, why would that cause a nuclear winter? What would that nuclear winter look like? Tell us some of those gruesome details.
Lynas: An India–Pakistan war with a hundred nukes or something wouldn’t deliver a planet-wide nuclear winter. It would make things slightly cooler for a bit, like a large volcano does. A major missile exchange between the superpowers would lead to a nuclear winter, if you burn most of the large cities of the Northern Hemisphere. And the mechanism is very well established now. You can see it happening in wildfires and some of the firestorms that have happened in human cities. Hamburg didn’t deliver a nuclear winter, but it did produce a pyrocumulonimbus—these big thunderclouds which are the mechanism by which soot, these dark particles, are transported into the stratosphere. That’s above the troposphere, where the world’s weather happens. If enough soot enters the stratosphere, it can’t be removed by rain. And because it’s black, it gets heated up by the sun and rises higher and higher into the stratosphere. It essentially becomes a veil that wraps around the Earth and blocks sunlight from reaching the surface with the same intensity.
In the worst-case scenario—outlined in a paper I referenced in my book, published in Nature Food in 2022—it involves about 150 teragrams of soot. That’s enough to make it essentially fully dark at the planet’s surface. Temperatures drop below zero practically worldwide, and they don’t rise above zero in the mid-latitudes. Somewhere like Iowa stays below freezing for about 750 days, I think. So, there are no more harvests. There’s no possibility of humans surviving—even with stored food supplies—in any significant numbers at all. A nuclear famine would inevitably result from that.
Mounk: Take us through the stages of this. You have, let's say, China and the United States going to nuclear war with each other, having this very significant exchange of nuclear weapons. How many people would die as a direct result of that exchange of nuclear arms? How many people globally would be likely to survive the ensuing nuclear winter?
Lynas: The majority of people survive the initial war. The death toll from an exchange of, let’s say, 1,500 nukes each—so 3,000 total—at one megaton per explosion, just for the sake of argument, would probably kill between a third of a billion and three-quarters of a billion people. So the majority of the population is still left alive.
But if you look at this Nature Food paper, then you have to consider what proportion of the population can survive the ensuing food crisis. There’s actually a table in the paper that goes through all of the countries’ populations. I think for the United States, 99% die. Some countries it’s 100%. But virtually all countries in the world lose the majority of their populations. I redid the figures for a world of 8 billion, and it’s getting close to seven billion people who would die in the nuclear famine, according to these model extrapolations. That’s based on estimating what food might remain and distributing it among the population using a lifeboat ethics approach—so that whatever the proportion is that can have sufficient calories are able to survive, and everyone else then dies.
Mounk: Would there be some way of preparing for this? If the United States is spending enormous amounts of money on national security—if someone at the Pentagon took the risk of nuclear war and nuclear winter very seriously—would it, in theory, be possible to start storing up enough preserved food so that the American population could get through that kind of nuclear winter? Is that just one of the precautions humanity should try to implement at that scale? We can obviously produce all kinds of foodstuffs that are sufficiently nourishing and have a shelf life of over two years. Actually, storage wouldn’t be a problem. It’s going to be so cold, you could just store things outside. You could have canned tuna for two years, whatever.
The problem is that we haven’t done that. But presumably, would it in theory be possible to prepare for nuclear winter in that way, and thereby significantly boost the number of people who would survive it?
Lynas: That’s an interesting question. It’s a bit like terraforming. It’s as if you suddenly find yourself on an uninhabitable planet and you’ve got 8 billion people to feed—or, if you're thinking at the level of a single country, maybe a couple of hundred million. How do you do that for three years? I guess it’s not fully inconceivable that you could store enough foodstuffs. But you’d need to be able to produce food without using photosynthesis. So you'd have to find other ways of synthesizing fresh food and be able to store the ingredients for that. This applies to any kind of sunlight-blocking catastrophe—a supervolcano or a very large asteroid would have the same effect. So it's a conversation worth having: what are the survival mechanisms if you want the species to survive? Elon Musk’s plan of going to Mars is probably the least reliable way of doing that. I mean, Earth will still be better than Mars under any plausible scenario.
Mounk: Presumably it's less about figuring out how to grow food during the nuclear winter and more about pre-producing the food and storing it in such a way that you have enough reserves to get through the nuclear winter.
Lynas: You’d need to be able to stop people you haven’t prioritized survival from accessing it or from seizing it by force.
Mounk: Well, unless you have enough food—enough that people know they’re going to get through it. In your book, you talk in quite vivid detail about how, because there would be such an extreme famine, you’d get these roving gangs and all the violence that would ensue as people try to procure the food they need.
But if over the next 10 years we simply say: let’s increase U.S. agricultural production by a certain amount in order to produce vast stores of food that could get us through three years, you avoid not only the starvation, but also the conditions that produce the need for that gang warfare—for those roaming gangs of people desperately searching for food and using violence and weapons to get it.
Lynas: Yes, so you obviously need to be able to maintain and defend mechanisms of centralized political authority. You have to be able to distribute food. If none of it is being produced independently on farms, you need some kind of distribution system—and that system needs to be defensible. It needs to function in a way that allows you to maintain social stability. How you do that in a freezing world of total darkness, for at least several months, is a huge challenge.
Precipitation levels don’t return to pre-war levels either. It doesn’t rain or snow to anything like the same extent, and that lasts for at least a decade. So we’re talking about a very, very long time. I think, at the moment, it’s inconceivable to store the amount of food that would be required to sustain the current global population.
Mounk: Let’s assume that this scheme I just made up doesn’t work. There’s this huge famine. Presumably, there are also a lot of people dying from the aftereffects of radiation and so on throughout this period. The medical system would presumably break down. So what would the level of illness, pestilence, and sickness look like during this nuclear winter?
Lynas: Immediately after the bombs, there’s no triage. I don’t think there’s any conceivable medical system that could support that level of death and destruction. So everyone who’s injured is going to die—especially if they’ve got acute radiation sickness. Most of them would be dead within a month or two. After that, radiation becomes a fairly trivial concern. There might be a couple of percentage points’ increase in cancer incidence in the decades to come, but really, that’s the least of your problems. So I wouldn’t be too worried about radiation as long as you can get people into fallout shelters for a couple of months in the hottest zones initially, and keep them alive and well-supplied in that scenario.
I can’t believe I’m talking about how to survive a nuclear war—the whole point is not to have one—but you did ask the question. But yes, I think food is really the big one in this issue.
Mounk: If we don’t take significant additional precautions beyond what exists today—if this happened tomorrow—and, as you're saying, up to a billion people die from the direct impact of the nuclear war, and then up to 99%, perhaps more in some places, of the population starves to death in the years after that, where does that leave humanity? If this nuclear war happens in 2025, what does the world look like by 2035? What about 2055 or 2075? Are we basically back in the Stone Age? Or do you think enough humans survive with enough knowledge that some of the technology we’ve developed over the last thousand years is retained, and humanity starts to rebuild?
Is this potentially a literal extinction event—where humanity disappears altogether—or is it an event that, while killing the majority and leading to devastation in all kinds of ways, still allows those who somehow survive the bottleneck to rebuild something like the civilization we know in the decades and centuries to come?
Lynas: I think it’s the latter. I don’t think anybody foresees this as a total human extinction event. But the humans who survive would be living in a devastated biosphere, where all the trappings of civilization—from electricity to internet connectivity to world trade—will have ceased to function, because most of the major countries will have been wiped out.
I keep referring to this Nature Food paper because it’s the one that actually quantified the death tolls in different places. There were some countries that had 0% deaths: I think Argentina, Australia, i.e. countries with very large land areas that currently produce large agricultural surpluses for export, which obviously wouldn’t be exported any more, and that also have relatively small populations. So, it’s not as if in this scenario everybody dies. But I think the post-apocalyptic society that would emerge is very difficult to visualize. We don’t really have many historical analogs—not even from famines. In the book I talk about cannibalism, because that’s actually a fairly common occurrence in famines, and even in things like shipwrecks and plane crashes. So it’s quite likely that humans would eat each other—that might be the main remaining food source in years two, three, or four.
But what happens in the rebuilding scenario—I really have no clue. Humans have rebuilt very successfully from past cataclysms. If you think of the Second World War, look how quickly Japan and Germany rebuilt from that. Their cities were basically just flattened ruins. So, I think the least likely outcome is total human extinction. But it becomes more possible, obviously, if you have drastically fewer humans left on the planet in a severely devastated biosphere.
Mounk: I have a personal question for you: Would you rather survive, or would you rather die?.
Lynas: If you read Cormac McCarthy’s The Road, that’s sort of the conundrum there: what’s left worth living for? There’s this love story between the father and the son, and the father obviously tries to stay alive for the sake of the son—but what kind of world is the son going to be living in? That was in the back of my head as I was writing this book. Our parents used to say, I’d rather be under the first bomb. Having written this book, I can see why, and I think they were probably right.
I don’t even know. The closest parallel I can think of is, in a way, the Holocaust—the struggle for survival in a place like Auschwitz, where all morality has vanished. The best you can hope for is some kind of degenerate existence, where you’re desperately scrabbling for grains of food on the floor and fighting against other people—grandmothers, children—to get to it. So is that a world I’d want to try and survive in? I’m not really sure it is, no.
Mounk: Okay, so we’ve done the really gruesome stuff at the beginning of this conversation—actually trying to inform people and think through what this would look like.
I think the question now is: that’s not what any of us want to happen. So how do we avoid it? Let’s backtrack a couple of steps. Tell us about the history of the development of nuclear weapons and why it is that, so far, we’ve been able to avoid nuclear confrontation. There’s the famous Doomsday Clock which is always five minutes to midnight. It’s been five minutes to midnight for more than five minutes, and we haven’t reached Doomsday yet. So why is it that we’ve been able to avoid this really gruesome outcome so far?
Lynas: Nuclear weapons were famously developed in the Manhattan Project during the later stages of the Second World War. As everyone knows from Oppenheimer and just general knowledge, two atomic bombs were used in Japan—on Hiroshima and Nagasaki—on the 6th and 9th of August, 1945. The Americans had more bombs planned. They were preparing to drop them on Kyoto and other Japanese cities that had been spared from the massive firebombing campaign, which had already killed hundreds of thousands of people—including the infamous raid on Tokyo on the 9th of March 1945. That, by the way, resulted in the largest single death toll of any air raid, bigger than Hiroshima.
Mounk: That’s because so much of central Tokyo was made up of wooden houses. It wasn’t, I guess, entirely unlike the nuclear winter scenario in that sense. It wasn’t just the strikes themselves or the bombs themselves that killed most of the people, though they did kill a substantial number. It was the giant firestorm that resulted.
Lynas: The Americans designed their incendiary weapons specifically to cause a firestorm—that was the whole point of it. They were dropped on targets that were already burning. These were napalm bombs, meant to cause the maximum amount of damage to the particular type of housing that Japan had. There was a systematic program to wipe out city after city after city, night after night. It was believed that this would win the war—though it’s arguable to what extent that actually drove the Japanese surrender, versus the Soviet invasion, which was happening at the same time.
The history of nuclear weapons after that period is fascinating. The Soviets got them very quickly—thanks to intelligence obtained from Klaus Fuchs, primarily, during the Manhattan Project. He was a British spy who passed secrets to the Russians. So by the 1950s, you have an arms race underway, with the Soviets building up nuclear capabilities as fast as possible.
Then you get the Cuban Missile Crisis in 1962, when the world realized just how close it was to nuclear Armageddon. On “Black Saturday” on the 27th of October, there was a Soviet submarine that was forced to surface by the Americans. The Americans didn’t know the sub was carrying a nuclear-armed torpedo. They were strafing it and dropping depth charges, and eventually the sub was forced to surface. The captain was about to order a dive and launch that torpedo at the American Navy—which would have started a nuclear war over Cuba. But one of the Soviet signalmen got stuck on the conning tower while trying to descend—so they couldn’t shut the hatches and prepare for the dive. During that delay, another officer on board—someone who outranked the captain and had fairly high authority—noticed that the American Navy was signaling an apology. He cancelled the order to fire. It was that close. Seconds—just seconds. By that point, Khrushchev and Kennedy were already desperate to de-escalate. But facts on the ground had taken on their own momentum. The Soviets had nuclear missiles ready to launch from Cuba much earlier than the Americans had realized. So if the United States had gone ahead with an invasion—as many of the top military brass wanted—there almost certainly would have been nuclear war then, too. There were multiple moments during the Cuban crisis when the world came very close.
Of course, there have been many incidents since—whether it’s nuclear bombs accidentally dropping out of planes or blowing up, like that Titan II missile in Arkansas in 1980. Or false information about incoming strikes. That’s what’s so dangerous, really—not just deliberate use of nuclear weapons, but inadvertent nuclear war.
Mounk: How much should the concern be about nuclear proliferation? You say, obviously, the United States was the first country to get the nuclear bomb. The Soviet Union quickly followed and eventually China exploded the bomb, I believe in the 1960s if I'm remembering correctly.
Lynas: The UK had one earlier as well, in the 1950s.
Mounk: I believe there's some question today about whether the UK is actually physically capable of deploying a nuclear bomb without American assistance, which is rather striking.
Lynas: That's true and that's a question now with the Trump era. That's a serious question for the UK, which doesn't apply to France.
Mounk: Now, India and Pakistan have them too. Israel has never acknowledged having a bomb, but they evidently do. North Korea has nuclear facilities. Iran is trying to develop a nuclear bomb and appears to be quite close to it. How does the probability of nuclear war increase because of that proliferation?
Lynas: That’s a really interesting question, and I actually look at this mathematically in the book. The published figure is about 1% per year for an inadvertent nuclear war. That’s the estimated annual probability. Now, technically, probability isn’t cumulative in a simple way. It’s not like it goes 1%, 2%, and over 100 years it becomes 100%. But if you throw a set of dice many times, you’re more likely to get a certain outcome than if you throw it once. So in that sense, it is cumulative. The probability of a nuclear war over the course of a century is probably around two-thirds—just for an inadvertent nuclear war. The probability is even higher for an intentional one. So it’s a significant risk over long periods of time. You’re looking at more than 50–50 odds, probably. That’s what’s so scary. On any single day, we’re probably not going to have a nuclear war. Even in any given year, it’s unlikely. But over time, the risk accumulates to a very dangerous level indeed.
Mounk: What is the chance of being able to contain a nuclear war in some way? Obviously, the United States deployed two nuclear weapons in Hiroshima and Nagasaki, and since then there hasn’t been any further use of nuclear arms. Is there some world in which, say, if an India-Pakistan war escalates, there’s an exchange of one nuclear weapon each—and then somehow politicians are able to stop further deployment?
Or is the existential threat—the moment you know your adversary has deployed a nuclear weapon against you—so high that you feel you have to immediately launch all of your nuclear weapons to try to degrade their ability to keep attacking you? In other words, once you have nuclear war, is it virtually by definition an all-out nuclear war?
Lynas: Well, it sort of is for the United States. The United States has a “launch on warning” posture. I called the book Six Minutes to Winter because, theoretically, the U.S. president has just six minutes to decide whether to respond to incoming missiles—when they’re just blips on a radar screen. That’s obviously an impossibly short window to make a decision about the future survival—or not—of civilization. But the theory has always been, and still remains, that you have to get your ICBMs out of their silos before any of the incoming missiles hit. Obviously, that raises the risk of inadvertent nuclear war, because you have to trust that the technology is correct—that the blips are real. There have been several occasions when those blips weren’t real, on both sides—both Russia and the United States. But what that suggests is that the leaderships—Russia likely has a similar posture, though it’s less transparent—believe they need to launch a full retaliation in response to any incoming missile threat. So it’s less likely that you get just two or three missiles exchanged and then people back off.
Another way to look at this is through wargaming. There have been lots of exercises, and they do tend to escalate, because if you’ve lost two million people, and maybe your capital city has been hit, you retaliate with several hundred missiles. One way or another, you get to the full exchange.
Mounk: What about the distinction between a regional war and a global war? Presumably, if the United States is involved in a nuclear exchange—particularly with Russia or China—that would effectively turn into a global nuclear war relatively quickly.
But what if it’s between India and Pakistan, or some of the other more regional nuclear powers? Is there an imaginable scenario where it stays contained to that region of the world? If so, what would be the impact and outcome of that?
Lynas: I looked at the nuclear arsenals of India and Pakistan recently. The most recent Indian figures were just published in the Bulletin of the Atomic Scientists. They probably have a couple of hundred warheads each. So, the scenario would be: let’s say they destroy each other—they destroy all of their major cities—and several hundred million people die. That would probably inject enough soot into the atmosphere to cause a degree or two of global cooling for a year or so. It would be a bit like Mount Pinatubo in 1991, but 3x. So it would affect global food production. And of course, it would be a cataclysmic shock to the world—politically and in many other ways—but it’s survivable for the rest of the planet.
So, a regional nuclear war is a survivable event, for sure—for the majority of the world’s population. Obviously not for the combatant powers. That’s different, of course, from an unsurvivable nuclear war, where it’s not just the combatant powers that are wiped out, but almost everyone else as well.
Mounk: Alright, so what do we do about all of that? Beyond thinking about it and scaring ourselves and telling horror stories to our children if we're inclined to psychologically torture them, what do you think the number one, number two and number three priorities are in terms of trying to avert what obviously would be a horrendous outcome?
Lynas: It sounds a bit trite, but we have to start talking about it. At the moment, you almost feel a little bit ashamed. When I told people I was writing a book about nuclear war, people laughed and said, that sounds a bit depressing, or, God, Mark, can’t you stop writing about depressing things? because my last book was on climate. So I think that’s one thing: we need to break the taboo and talk about it more.
We also have to start formulating grassroots, bottom-up political pressure. The kind of movement that’s paid huge attention to climate change over the last 15 or 20 years has put it firmly on the agenda—to the extent that, even with the current backlash, all the major economies are now covered by net-zero targets. We have annual COPs, and huge amounts of attention are given to the issue. But none of that is happening with nuclear risk. There are annual meetings for the Treaty on the Prohibition of Nuclear Weapons—the TPNW—which came into force in 2021. It’s a treaty that’s been signed by half the world—more than 90 countries—and ratified by over 70. Yet, most people don’t even know it exists. When I talk to audiences about the TPNW, no hands go up. But I think it’s important, because there is a legal mechanism at the United Nations level for the full abolition of nuclear weapons. I think there are mechanisms within the TPNW to begin increasing pressure on the nuclear weapons-holding states—of which there are only nine—to force them to the negotiating table, to start getting serious again about arms control. But recently, the reverse has been happening. All the arms control treaties—the legacy of the Cold War—have started to fall apart.
Mounk: How realistic is it that we’re ever going to get to complete disarmament—without just doing all of this again? How realistic is it that the nuclear powers would willingly give up their arms?
There are two concerns here. One is that even a superpower with the ability to defend itself very effectively by conventional means is going to be deeply mistrustful of whether its geopolitical rivals would actually stick to the same deal. Take the United States—it’s not meaningfully at risk of physical attack from China. So perhaps it could say, we’ll take our chances. We can defend ourselves with conventional weapons if we need to—we can give up nuclear weapons. But the counterargument is that China would obviously dominate us—and Russia would dominate us—and any nuclear-armed power would dominate us if they kept their nuclear weapons and we didn’t. So if we give up ours, and those other powers retain their capability—either by cheating, not adhering to the deal, or somehow deceiving us about their arsenal—then suddenly we’re in an incredibly weakened position. That’s a pretty powerful argument for those powers not to give up their nuclear weapons.
Presumably, there’s a similar logic for smaller powers. The reason places like North Korea and Iran are so invested in nuclear arsenals is that they feel that’s the ultimate deterrent against potential attack. The moment you have nuclear weapons, they serve as your defense of last resort against conventional incursions into your territory. Of course, one of the ironies here is that we’ve made this strange Faustian bargain: the risk of human annihilation is also one of the reasons why we’ve had relatively fewer deaths in war—despite the horrible conflicts in Ukraine and the Middle East—than we did for much of human history. That’s precisely because so many of the major powers won’t go to war with each other, due to the risk of nuclear escalation.
Lynas: I think that latter point is arguable. I think Russia was emboldened to mount a full invasion of Ukraine in 2022 precisely because it had nuclear weapons and Ukraine didn’t. So the existence of nuclear weapons didn’t prevent war; in fact, it may have made it more likely. We have to find a way for humans to continue fighting wars—because that’s not going away—but without destroying the biosphere. Even reducing the number of nuclear weapons down to maybe 10 or 15 per country could make a difference.
But then the question becomes: does every country get 10 or 15? Or is it still just the current nine? And if one country goes to 20 or 50 or 100, then you’re right back in an arms race. So I think what’s crucial here is to abandon any thoughts of unilateralism. That’s where CND and the old peace campaigns of the ’70s and ’80s went wrong. The demand for unilateral disarmament was misguided and felt misguided even at the time. To borrow Trumpian language, the only “cards” you hold are the offer to disarm your own nuclear weapons if your opponent agrees to do likewise, in some kind of fully verifiable, stepwise process. You also have to maintain full control over all fissile materials and the civilian nuclear fuel cycle, because of the proliferation risks. The mechanisms for this do exist. This is exactly what the IAEA is meant to do, and does very successfully in many places. That’s also how we know the Iranians are pursuing a bomb. They’re enriching uranium to far higher levels than are needed for any kind of civilian reactor.
We’ll see how the world responds to that. The Israelis are more likely to attack Iran in order to prevent it from acquiring a nuclear weapon. Iran is very likely to keep pursuing one because, as with North Korea, it’s seen as the ultimate guarantee of regime survival. Something’s going to give. Both of those positions can’t be held indefinitely.
But at the same time, we also have to keep in mind the long-term ambition of getting to what I call “global zero.” That’s the only way human civilization survives over the long term, and by long term, I mean decades and centuries from now.
Mounk: It’s important to distinguish your position from unilateral disarmament which obviously seems unrealistic. When you have serious geopolitical rivals who do have nuclear weapons, to say, we’re just going to disarm on our own, weakens your position in a way that very few leaders would be willing to do and very few electorates in democratic countries would be willing to support.
Whereas saying, we’re going to try and reach some kind of global deal where we all disarm at the same time, still involves the nuclear nine weakening their position relative to the status quo but it’s at least more imaginable. What are the prospects for some form of global agreement? It feels to me like two things are happening at the same time, at least in how I think about this topic compared to 10 or 20 years ago.
One is that we seem to be living in a more conflictual world, with leaders who are more self-interested, more assertive, and less concerned about global goods—like safety from nuclear weapons—than the leaders of several decades ago. It would have been somewhat easier to imagine the cast of leaders 20 or 30 years ago making that kind of deal than it is to imagine today’s leaders doing the same. So the prospects for “global zero” seem, to me, lower than they were 20 or 30 years ago.
At the same time—somewhat inversely—it also feels like the underlying risk is higher than I would have assessed it to be a few decades ago. The risk was always there and it’s always been real. You’ve spoken very compellingly about how close we came during the Cuban Missile Crisis, for example. But I think I had growing up—coming into political consciousness as a teenager in the ’90s and then as a student in the 2000s—a basic faith in the rationality of the world. That, in the end, political leaders made relatively rational decisions. That they didn’t want to destroy the world. That we had mechanisms in place to prevent worst-case outcomes. That when we worried about something like a global pandemic, for instance, we had public health authorities who were competent and prepared to stop it. That, by and large, those systems worked. I think that made it easy to assume that these are serious risks, but in the end, we have rational systems that will contain them. Things are going to be fine.
But in part because of the pandemic, in part because of how erratic some major world leaders now are, and in part because of the scale of recent wars—like in Ukraine—I feel less confident in that view. It’s always tempting to think, “If some horrible outcome hasn’t happened yet, maybe it can’t happen.” But that’s intellectually wrong. It feels like the preconditions for a truly catastrophic outcome are higher now than they were a few decades ago on both fronts: the capacity for coordinated disarmament has declined, and the likelihood of risk-realization has grown. That’s a pretty depressing conclusion, right? That the prospects for getting to global zero are worse than before, and our understanding of the risk is more severe because the systems we once trusted turn out not to be all that rational. So, is it still possible to remain hopeful in that situation?
Lynas: It’s obviously difficult to visualize something for which there is no substantial political constituency and no demand. But that’s why I use the analogy of climate change. I’ve been around long enough and campaigning on that issue long enough to remember what it was like when no one gave a toss. When EVs were milk floats, just a joke. When solar panels cost $10,000 per square inch. It was inconceivable that modern industrial civilization could function without fossil fuels. We were heading toward a world of five or six degrees of warming—another catastrophic outcome. But we’re not in that world now. We’re moving—fully committed, I think now—towards a clean energy transition. It’s going to take a long time, and we’re likely going to get to plus two degrees. It’s going to be a very challenging few decades on the climate front alone. But that change happened for both technological and political reasons. And I think technology, politics, and economics are all intertwined. But politics comes first. We need a significant political demand—at least in democratic countries—for a de-escalation of nuclear risk. That demand simply isn’t there at the moment. It’s something we need to organize—whoever we is.
I’ve tried to do my part to raise awareness and, in some small way, help put it back on the agenda. I just hope that enough people begin to realize the risk is still there. People have dismissed the idea of nuclear winter and thought it was debunked. But it really isn’t. I think that is why it’s so difficult for us to imagine a world without nuclear weapons: because no one is currently asking for it.
In the rest of this week’s conversation, Yascha and Mark discuss how the fight against climate change has changed over the years, and why we need nuclear power. This part of the conversation is reserved for paying subscribers…
