The International Security Podcast

1 - A ‘Nuclear Umbrella’ for Ukraine?

Episode Summary

In this episode, Matthew Evangelista discusses his recent International Security article, “A ‘Nuclear Umbrella’ for Ukraine? Precedents and Possibilities for Postwar European Security.” How can Ukraine and Europe deter further Russian aggression after the Russo-Ukrainian war ends? The conventional wisdom holds that Ukraine should join NATO to shelter under the U.S. nuclear umbrella. Evangelista challenges this assertion, arguing that nuclear deterrence was never tested during the Cold War. He proposes an alternative way to reestablish European security: a non-offensive, confidence-building defense that does not rely on the threat of nuclear war.

Episode Notes


Matthew Evangelista is President White Professor of History and Political Science at Cornell University.

International Security Article:

This podcast is based on Matthew Evangelista, “A ‘Nuclear Umbrella’

for Ukraine? Precedents and Possibilities for Postwar European Security,” International Security, Vol. 48, No. 3 (Winter 2023/24), pp. 7–50,

Originally released on March 20, 2024.

Episode Transcription

Jeff Friedman: Hello everyone and welcome to the International Security Podcast. We are produced by International Security, a quarterly journal edited at Harvard Kennedy School's Belfer Center for Science and International Affairs and published by MIT Press. Each episode of the podcast will highlight an important piece of research from the journal, drawing out its implications for understanding the theory and practice of international politics.

I'm Jeff Friedman, Associate Professor of Government at Dartmouth College, and our guest today is Matthew Evangelista. Matt is the President White Professor of History and Political Science at Cornell University. His recent International Security article is titled “A Nuclear Umbrella for Ukraine? Precedents and Possibilities for Post-War European Security.” The article gives us a broad reassessment of the relationship between nuclear weapons and European security, from the Cold War through the current conflict in Ukraine. It's a big argument with significant policy implications. Matt, thanks so much for joining us today.

Matthew Evangelista: Thanks for having me.

Jeff Friedman: Okay, so your article takes aim at conventional wisdom about the role nuclear weapons have historically played in NATO's defense. Can you just start by telling us what that conventional wisdom is and how it shapes contemporary debates about defending Ukraine and Europe writ large?

Matthew Evangelista: The conventional wisdom is that nuclear weapons during the Cold War prevented the Soviet Union from invading Western Europe. So nuclear deterrence--it's called extended nuclear deterrence—it's a threat by the United States to defend an ally if necessary by attacking the country that invades it with nuclear weapons.

It's also colloquially known as the nuclear umbrella. And the conventional wisdom is it worked. There's a problem with that conventional wisdom because we know now that the Soviet Union did not intend to attack Western Europe. So in effect, nuclear deterrence wasn't tested. But it's still very popular. People still think this is the way to protect Ukraine in the future after the current war with Russia ends. Maybe it would've protected Ukraine if Ukraine had nuclear weapons before the invasion. That's basically it.

Jeff Friedman: So the article goes into some detail, as you say, into looking into the Soviet archives and indicating that it's unclear that the Soviets ever had a positive plan to attack Europe. If that's the case, then what were the risks of defending the continent with nuclear weapons? If the Soviets weren't going to attack anyway, what liabilities did that expose the West to?

Matthew Evangelista: So the main risk is that a conflict could escalate to the use of nuclear weapons. A crisis, for example. Both sides had strategic nuclear weapons that could attack the other side, intercontinental missiles and the like, but they also had thousands of tactical nuclear weapons. In Europe, the U.S. had about seven thousand. The Soviets had maybe eight or nine thousand, something like that. And there were plans to use them. The most dangerous such crisis, the one that I write about in the article, was over the status of West Berlin during the Cold War.


Because you'll remember, although many of my students wouldn't have any way of knowing if I didn't tell them, that West Berlin was a very anomalous situation. It was deep within Eastern Germany—the German Democratic Republic, a Soviet ally, surrounded by hundreds of thousands of Soviet troops. Yet because of the way the war had ended with four-power occupation of Germany and of the capital city of Berlin, there was a western zone, linked to the Federal Republic of Germany and the United States, and NATO vowed to protect it. They had access to it, but they had to drive or fly across much of Eastern Europe to get there. So they were always worried that the Soviets might encroach on western Berlin. And then what would they do? And they relied on nuclear weapons—extended nuclear deterrence—ostensibly to protect West Berlin.

Jeff Friedman: Yeah, I thought one of the most striking parts of the article was just seeing how plausible the paths to escalation were in that crisis. And I think you drew that out really well. There are a lot of differences between Europe in the 1960s and today. One of them is that Vladimir Putin has demonstrated a willingness to launch a major war in Europe. Do you think that makes NATO's nuclear deterrent more or less important moving forward?

Matthew Evangelista: Yeah, it's a tricky question. It…what was unusual about the Cold War, and we see this better in retrospect, is it was, according to Robert Jervis, a genuine security dilemma, that neither side really wanted war. There was a lot of misperception in the structure of each side's forces, and that's what made the Berlin crisis so dangerous, that the United States thought it could use nuclear weapons as a demonstration, a single use of a nuclear weapon or something like that, and the Soviets would back down and allow access to the city. But in fact the Soviet war plans, and again, we have good information about this mainly from Vojtech Mastny and his project on the Warsaw Pact, from the Warsaw Pact archives, that the Soviets would've responded with massive preemptive use of nuclear weapons to try to destroy the NATO ones.

So that kind of uncertainty and risk was the main problem. That—we don't have uncertainty, in a way, now about Russia's intentions because as you said, Putin has already attacked Ukraine and we don't know what else he would be willing to do. So a lot of people would say once this war is over, whatever the territorial settlement turns out to be—it probably won't be with Ukraine restored to its pre-2014 borders—whatever that settlement will be, maybe Ukraine should join NATO, should benefit from the U.S. nuclear umbrella, the threat of attack against Russia, if it invades again. I think that's wrong. I think it's too dangerous and I don't think it would necessarily work.

Jeff Friedman: It's interesting that the article talks about the security dilemma. We often think of that as a situation in which both sides need to be reassured, but in this case with Russia seeming so aggressive, why would we think that Moscow needs reassurance from NATO? Why do we think that the security dilemma is a valuable way of thinking about the risks and benefits of various forms of defense posture today?

Matthew Evangelista: I think it really depends on the nature of the settlement that ends the war. It's likely that neither side is going to be fully satisfied. If Crimea, for example, remains part of Russia, if some of the Donbas remains part of Russia, the Russians could expect that Ukraine would want to retake that territory. The Ukrainians could expect that Russia might want to resume the war and seize more of Ukrainian territory. Nuclear weapons doesn't solve that situation. What would help would be a robust defense that would not make the other side think that it would break the agreement and attack.


So it's strange to say that we want to reassure Russia, because Russia was clearly the aggressor. But if we don't want Russia to resume the war then we want to make sure that Ukraine can defend itself, but can't violate the terms of the territorial settlement in a way to provoke Russian aggression again.

Jeff Friedman: One of the interesting aspects of the article, I thought, was you have a debate at the end about just how risk acceptant we think Vladimir Putin is, and whether his invasion of Ukraine reflects an appetite for risk as opposed to being misinformed about the potential costs of the invasion.

But I wonder even if we think that Putin is very risk acceptant, why would he expose Russia to a risk of nuclear escalation? And apart from Ukraine, what do we think are some of the other plausible places in which European security could potentially break down?

Matthew Evangelista: Yeah. So I have a section called “The Narva Nightmare,” and I'm not the first to call attention to this. Narva is a city in Estonia. It's on the border with Russia, it's literally across the Narva Russia [River] from the sister city of Ivangorod, and it is very vulnerable for a number of reasons. Physically, it's vulnerable because the Russian troops could just march across the river or over the bridge and seize it.

But politically it's vulnerable because 95% of the people who live in Narva are Russian speaking. Russian is their first language. 88% are ethnic Russians. So we talk about the Russian-language issue as being related to the causes of the war in Ukraine. This is a much more serious problem.


And where the risk comes in is—so Estonia is a member of NATO. The Republican politician Newt Gingrich once mentioned that it's located in the suburbs of St. Petersburg, and he didn't think NATO should commit to defend a country that's in the suburbs of St. Petersburg. But that is NATO's commitment under Article Five, that if Narva, if Estonia came under attack by Russia, NATO would be committed to come to its defense. And how would it do that in a city so vulnerable? It would do it by treating it as a trip wire and the trip wire would lead to a nuclear attack. That's what extended nuclear deterrence means.


And the reason it's a nightmare is we just don't know if Putin would take that risk, because what if he did attack Narva, in effect called NATO's bluff, and NATO decided it didn't want to blow up the world to save Estonia? He would totally undermine the credibility of the NATO alliance, and that's a big prize. I just don't think that's a risk worth taking. Either you undermine the NATO alliance or you get a global nuclear war. That's some choice.

Jeff Friedman: Yeah, and another part of that discussion in the article that struck me was just how hard it is to make those security guarantees clear that…I think you lucidly point out that there are many ways in which Russia could try to salami slice or send in the little green men in order to undermine Narva’s security in a way that might make it quite difficult to trip that wire. So I thought that was effective.


Okay. So far you've mentioned that rather than relying on nuclear weapons, NATO needs some way of defending Ukraine, Narva, and other interests that relies on conventional defense. That defense needs to be non-provocative and it has to be credible. That is—seems like a tall order.

What do you think would foot that bill? What kinds of alternative approaches do you have in mind and where do those ideas come from?

Matthew Evangelista: I don't think there's any easy solution to Narva. Gingrich is right in this point. It's geography makes it vulnerable no matter what. I think the best thing for Narva would be to make sure that those Russians and Russian speakers are not vulnerable to any kind of infiltration by Russia. Make sure that they feel loyalty to the Estonian state and then prepare them to resist an occupation in a way that would make Putin think twice, let's say, before trying it.


For Ukraine, they are in a better position having gone through this devastating war, but having been fighting to defend themselves from Russia since 2014. They have quite a strong army. The argument here would be to look to some of the proposals that emerged in the late Cold War from military analysts and peace researchers who wanted to lower the risk of nuclear weapons. So they developed these strategies called “defensive defense” or “non-provocative defense.” And the idea was to have a robust conventional defense, but without an offensive capability, without the capability to invade, seize, and hold foreign territory. So the idea is that you make it really tough for an invader. But you don't pose a threat to attack your neighboring country. And those have been revived lately especially by German scholars. One of the figures during the eighties was Lutz Unterseher, and some people have rediscovered his ideas. He's got involved again.


So one particular proposal I briefly described is called the “spider in the web.” That basically has a number of barriers to try to prevent an invasion, then to encircle the invaders, and then to strike. So the web and the spider. Anyway, there are lots of ideas like that, and what I'm arguing is they deserve a new hearing rather than just assume that we're going to rely on nuclear weapons, just repeat the mistakes of the Cold War.

Jeff Friedman: How would those…how would equipping a state like Ukraine for spider in the web, or I think another metaphor in the paper is the bristling porcupine—these are all very vivid—how would equipping them for that kind of defense differ from the kinds of military assistance that the West is currently providing in the ongoing war?

Matthew Evangelista: It would probably entail a more reliable supply of weapons than what we've seen with this politicized situation we're facing now. But it would eschew offensive weapons.


The proposal for the bristling porcupine actually has in mind to have various long-range weapons that could hit Russia. I can understand why Ukraine would want those to bring the war back to Russia and to have ordinary Russians notice that there's a war going on. But from the standpoint of the security dilemma, from the standpoint of the stable post-war settlement, it would be better to have defensive weapons.

So it would rely on a lot of things that the Ukrainians have now, especially sensors, because the idea is that you want to know where the invading troops are so they can be encircled and attacked.

Jeff Friedman: The next thing I want to talk about is just the political implications of these ideas. So, dating back to the early days of the Cold War, it was often thought that one of the benefits of the nuclear guarantee was that it had a political or psychological effect in stiffening European resistance and that was partly because European states found it quite challenging to build domestic support for the kinds of conventional defense that might have allowed NATO to protect itself without relying on the U.S. guarantee. So particularly now we see support for continuing to arm Ukraine flagging, particularly if the United States withdraws its nuclear guarantee, that could require a substantial increase in European states’ conventional military spending.

Do you think that's something that's feasible? To what degree do you think we could move towards this conventional defense in the short or medium run?

Matthew Evangelista: Yeah, so this has always been the argument, that nuclear weapons provide more bang for the buck. But it's also led to things like Donald Trump's challenge not to maintain the U.S. commitment to NATO if countries don't pay enough. But I think there are a couple of misconceptions here.

One is that nuclear weapons are cheap. You remember that President Barack Obama gave a speech in Prague in 2009 where he looked to a future without nuclear weapons. He said it wouldn't happen in his lifetime. And then he approved a budget for modernization of nuclear weapons of $348 billion dollars over about a decade. That figure now for the period between now and 2032 has gone up to $756 billion according to the Congressional Research Service. So that's not cheap. That kind of money devoted to conventional defense of Europe would go a long way.


But I take your point. I think that the Europeans will have to think seriously about defense and Vladimir Putin has gone a long way in focusing their minds on that. There's another element though that I would mention, especially for those not enthusiastic about military defense. There's been a lot of research over the years on nonviolent civilian resistance to occupation as a deterrent. And there was an interesting Rand report that came out recently about how Ukrainians spontaneously—civilian Ukrainians spontaneously created barriers to the Russian invasion and occupation. And there's a scholar at Harvard, at the Kennedy School, Erica Chenowith, who has a big project on this topic. And it's worth exploring, I think, because it's also a way of deterring an aggressor if you say look, you won't be able to achieve your objectives in the occupation if nobody is cooperating with you. So that's also an element in addition to the military defense.

Jeff Friedman: One of the things that struck me about your article is that you're saying in addition to thinking about how to defend Ukraine right now, against the immediate challenges it faces, policymakers should be thinking about what comes after, where do we want European security to end up when the war, we hope, ends or reaches a ceasefire at some point? And one of the things your article allows us to do is think about how that potential end state affects policy right now.


You've already mentioned we might make some different choices in terms of what weapons NATO gives to Ukraine in order to craft that post-war settlement that might be stable. Is there anything else that you think the United States and NATO should be doing in the short term if it accepted your argument about where we want European security to go after that?

Matthew Evangelista: So I'm reluctant to tell Ukraine what to do or to recommend that the United States tell Ukraine what to do. They're the ones who are fighting for their survival. I think, as most people do, that there will have to be some kind of compromise territorial settlement before the war ends. But I'm not in a position to tell Ukraine what it should—what it should give up.


What the U.S. can do, and I think it would be helpful, is to stop being so preoccupied with nuclear responses to Putin's threats. Putin again, just recently in response to French President Macron's claim that French troops might join in helping the defense of Ukraine, started threatening nuclear retaliation again.

Again, I think the United States should call him out on that, because the UN charter itself prohibits not only the use of force, but the threat or use of force. There's a treaty prohibiting nuclear weapons. Neither the United States nor Russia has been very enthusiastic about it, but that also prohibits the threat of nuclear use.

So why not call out Putin as a criminal rather than talk about all the different ways that the United States is going to modernize its nuclear weapons and deploy them in Europe? I don't see that as helpful.

Jeff Friedman: Thanks for summarizing the article. I think readers will appreciate the strength of the argument, and agree or disagree with it, the piece really helps to frame our assessment of various trade-offs between different kinds of security outcomes that we might seek to achieve, political and military effects that those choices will have. And so I think the piece really does a lot to help shape policymakers’ choices moving forward.


Let me just ask you one last question. What else have you read recently in this field that you'd recommend to our listeners for thinking about how to make sense of these subjects?

Matthew Evangelista: Okay. That's an easy question because if your listeners read my article—it's the first one in the journal, if they can make it all the way through, if there's not too much history there for them, the next article is really excellent. It's by Simon Miles and it's on the decisions by the former members of the Warsaw Pact to join NATO or to press for joining NATO. And Miles is a polyglot. He's done an incredible amount of research in the East European archives, and so I would recommend that one very highly.

Jeff Friedman: Thanks to everybody for listening to the International Security Podcast, produced by International Security, a quarterly journal at Harvard Kennedy School's Belfer Center for Science and International Affairs, and published by MIT Press. If you liked the conversation, we'd appreciate it if you took a moment to rate and review us.

The executive editor of International Security is Jacqueline L. Hazelton. Our producer is Monica Achen. Associate producer and technical director is Benn Craig. Thanks again to our guest, Matthew Evangelista. I'm Jeff Friedman. See you next time.