Enjoyed the episode? Want to listen later? Subscribe here, or anywhere you get podcasts:

John F. Kennedy had ordered a stand down to all U-2 flights during the Cuban Missile Crisis, because he knew this was exactly the kind of thing that could get out of control. And he did not have total control, but a U-2 flight still went off and somebody got shot down.

And Kennedy’s response, which is like peak Kennedy, was, “There’s always some son of a bitch who doesn’t get the message.” And I think that’s a really good way of looking at dynamic complex systems.

Jeffrey Lewis

America aims to avoid nuclear war by relying on the principle of ‘mutually assured destruction,’ right? Wrong. Or at least… not officially.

As today’s guest — Jeffrey Lewis, founder of Arms Control Wonk and professor at the Middlebury Institute of International Studies — explains, in its official ‘OPLANs’ (military operation plans), the US is committed to ‘dominating’ in a nuclear war with Russia. How would they do that? “That is redacted.”

We invited Jeffrey to come on the show to lay out what we and our listeners are most likely to be misunderstanding about nuclear weapons, the nuclear posture of major powers, and his field as a whole, and he did not disappoint.

As Jeffrey tells it, ‘mutually assured destruction’ was a slur used to criticise those who wanted to limit the 1960s arms buildup, and was never accepted as a matter of policy in any US administration. But isn’t it still the de facto reality? Yes and no.

Jeffrey is a specialist on the nuts and bolts of bureaucratic and military decision-making in real-life situations. He suspects that at the start of their term presidents get a briefing about the US’ plan to prevail in a nuclear war and conclude that “it’s freaking madness.” They say to themselves that whatever these silly plans may say, they know a nuclear war cannot be won, so they just won’t use the weapons.

But Jeffrey thinks that’s a big mistake. Yes, in a calm moment presidents can resist pressure from advisors and generals. But that idea of ‘winning’ a nuclear war is in all the plans. Staff have been hired because they believe in those plans. It’s what the generals and admirals have all prepared for.

What matters is the ‘not calm moment’: the 3AM phone call to tell the president that ICBMs might hit the US in eight minutes — the same week Russia invades a neighbour or China invades Taiwan. Is it a false alarm? Should they retaliate before their land-based missile silos are hit? There’s only minutes to decide.

Jeffrey points out that in emergencies, presidents have repeatedly found themselves railroaded into actions they didn’t want to take because of how information and options were processed and presented to them. In the heat of the moment, it’s natural to reach for the plan you’ve prepared — however mad it might sound.

In this spicy conversation, Jeffrey fields the most burning questions from Rob and the audience, in the process explaining:

  • Why inter-service rivalry is one of the biggest constraints on US nuclear policy
  • Two times the US sabotaged nuclear nonproliferation among great powers
  • How his field uses jargon to exclude outsiders
  • How the US could prevent the revival of mass nuclear testing by the great powers
  • Why nuclear deterrence relies on the possibility that something might go wrong
  • Whether ‘salami tactics’ render nuclear weapons ineffective
  • The time the Navy and Air Force switched views on how to wage a nuclear war, just when it would allow them to have the most missiles
  • The problems that arise when you won’t talk to people you think are evil
  • Why missile defences are politically popular despite being strategically foolish
  • How open source intelligence can prevent arms races
  • And much more.

Get this episode by subscribing to our podcast on the world’s most pressing problems and how to solve them: type ‘80,000 Hours’ into your podcasting app. Or read the transcript below.

Producer: Keiran Harris
Audio mastering: Ben Cordell
Transcriptions: Katy Moore


Ways of thinking about nuclear deterrence

Rob Wiblin: As I understand it, a lot of people, including me, think about nuclear issues at a pretty high level of abstraction — where we’re going for the sort of pure game theory that I might have learned in second-year microeconomics. That’s a very natural way for me to think about this issue: if X does this, then Y will respond that way. And it’s quite a clean analysis.

A different way of looking at it would be to say that if you have a much thicker understanding of the powers that different individuals in the system have, and the different committees that something would go through, and how the information flows from the radar to the president, then you can instead think about it at the institutional level: How is a decision actually being made here? And given that, what decision is likely to be spat out by the system? And that’s at a much lower level of abstraction than some of the purest game theory.

Jeffrey Lewis: I think that that’s, again, the joy of dealing with both Tom and John at the same time. Because Tom really liked those game theoretical situations, but he was also aware of the ways in which they break down in reality.

I think the simplest example of this is: how do you even know you’re under attack? It’s not a video game. It’s not that you’re told that there’s an attack, and you’re told this accurately, and you’re given the correct number, and you’re given it in a prompt and predictable way. You know, we have radars and we have space-based sensors that look for infrared plumes. So that kind of data that comes in, there are often false alarms. There are things that are missed, there are ambiguities. It’s handled at an extremely low level, and it’s reported up a chain of command. And by the time the president’s told it, the president’s not looking at the radar image. The president has a person conveying words that are meant to describe this complex phenomenon that’s out there happening in the world.

And so when you realise how little control each individual really has in that process, on the one hand, it’s kind of a bummer, because it’s like, “Oh my God, we’re all going to die.” But on the other hand, it becomes a really rich and complex story.

And John F. Kennedy got this really well when he had ordered a stand down to all U-2 flights during the Cuban Missile Crisis, because he knew this was exactly the kind of thing that could get out of control. And he did not have total control, but a U-2 flight still went off and somebody got shot down. And Kennedy’s response, which is like peak Kennedy, was, “There’s always some son of a bitch who doesn’t get the message.” And I think that’s a really good way of looking at dynamic complex systems.

Rob Wiblin: Right, yeah. So I suppose in theory, it seems like you could try to incorporate all of this stuff into a game theory matrix, but by the time you’ve done that, it’s so enormously confusing that common sense and thinking through the process without a table might serve you just as well.

Jeffrey Lewis: I think you have to do both. I do not in any way, shape, or form reject game theoretic models to try to understand how decision-makers might act. I just think it is important to be familiar with the particular animals in the zoo. Particularly because I see a lot of predictable decision-making pathologies.

How the US will dominate Russia in a nuclear war

Rob Wiblin: OK, I don’t quite get how [“Mutual assured destruction”] isn’t a thing. Maybe on paper it’s not in the policy anymore. Maybe it never was called that in the official policies exactly. But isn’t it still the case that when Russia and the US are fighting a proxy war in Ukraine, the fact that they could both destroy one another — and that if things got to some point, they would both bomb one another and both of their societies will be ruined — doesn’t that, and their risk aversion, kind of constrain their actions and affect what they do?

Jeffrey Lewis: That’s just deterrence. The whole idea of mutual assured destruction is you accept the possibility of reciprocal deterrence. And our policy is not to do that. Our policy is to be able to prevail.

Rob Wiblin: I see. I thought that the US policy to some degree was to accept mutually assured destruction, and maybe build into the treaties with Russia that there’ll be a sufficient arsenal that both countries would be wrecked. But actually that’s not what’s going on. The US, does it still have an idea that it could win a nuclear war with Russia?

Jeffrey Lewis: Yes. Yes. Go look at the declassified OPLAN.

Rob Wiblin: How would that play out?

Jeffrey Lewis: So the idea is you deter, and if you failed to deter, I believe the next phase is “Shape.” Actually, I have a declassified OPLAN in front of me. I can look this up. Sorry, this is so goofy. I really hate the way they phrase these things.

There’s phase 0, which is “Shape.” Phase 1 is “Deter” — that’s where you’re attempting to deter a threat.

Rob Wiblin: So “Shape” is kind of shape the situation through your general foreign policy all the time. And then at the point where there’s a conflict on the table, then you’re engaged in deterrence.

Jeffrey Lewis: Exactly. Phase 2 is “Seize the Initiative,” which is, I don’t know, Deterrence+. And phase 3 is “Dominate.”

Rob Wiblin: And how exactly would the US dominate Russia in a nuclear war?

Jeffrey Lewis: So that is redacted. [laughs] So all we have is, “During this phase, USSTRATCOM, in collaboration with the applicable GCCs, conducts operations, integrated across all mission areas, in order to defeat adversaries and return to phase 0 on terms acceptable to the US. The USSTRATCOM, as directed by the President — redacted.”

The danger of silently defecting

Jeffrey Lewis:As I said, I believe that presidents silently defect from this [official US strategy]. I think they get the briefing and they say –“That doesn’t seem very sensible to me.” But because it’s a political loser, because it’s abstract, because it’s complicated, they allow the plans to be written, they spend the money to acquire the forces, they select senior officials who believe in the mission. And what I think they have concluded is that because they are silently defecting, they know they’ll never do it.

Jeffrey Lewis: And I think that is a terrible mistake. Because all the time, we see presidents get boxed in because they overestimate their power and they underestimate the structural constraints that will exist in terms of time, their own stress level, the advice they’re getting.

Jeffrey Lewis: There’s a great example of this in Fred Kaplan’s other great nuclear book, where there’s a war game that’s being fought. And the war game occurs at two levels: the so-called “deputies” level and then the “principals” level. The deputies are the deputy secretary of state, the principal is like the secretary of state. So you do it once with the number twos and then once with the number ones — you play the game twice.

And in the game, Russia uses one nuclear weapon — which I don’t think is very realistic, but we’ll just roll with it. The deputies decide one nuclear weapon is not going to make a difference to the war. Terrible risk of escalation. We’re just going to go ahead and win the war. And the punishment for the Russians is they lose the war and everybody thinks they’re awful.

They play it again at the principals level. And the principals decide they must respond with a nuclear weapon. They feel that their credibility is somehow on the line, in a way the deputies don’t feel. The deputies can be more dispassionate in some sense. But they’re terrified to hit the Russians back with a nuclear weapon because the Russians using a nuclear weapon against Poland is one thing. Us actually hitting Russia with one of our nuclear weapons, potentially very escalatory. And so they flounder around for a while before deciding that they’re going to nuke Belarus, which is not even a party of the conflict in the game.

And to me, what that illustrates is when you are in a situation that you have not fully anticipated, and you’re kind of making it up as you go along, and if you’re feeling some stress, and if you’re getting some advice that’s maybe not great, it’s maybe hard to silently defect anymore when you have invested all of this time and energy in creating this system.

So that’s why I say, even though I don’t think mutual assured destruction is a thing, it’s not a policy, it’s something the US rejects. I think presidents, when you have them in a calm moment, they know that’s madness. I am not worried about what they’re going to do in a calm moment. I worry what they’re going to do in a not calm moment when that happens.

Why inter-service rivalry is one of the biggest constraints on US nuclear policy

Jeffrey Lewis: This is like the old joke about the two US nuclear labs. In the US, we have two design labs, and they used to say it about each other. It would go something like, the folks at Los Alamos would say, “The Soviets are the adversary, but Livermore is the enemy.”

We have seen consistently in the US very high levels of inter-service rivalry, and we think that’s generally a good thing, right? That’s our competition, free-market idea. The last real effort to unify those services was the Truman administration, which really wanted to weaken the Navy and abolish the Marine Corps, and that led to the very famous Revolt of the Admirals, which was a pretty severe crisis in civil-military relations. As it happens, I think Truman was wrong about the Navy, but right about civilian control of the military.

So no, each of our military services has their own unique camouflage patterns. What I would say, charitably, to those folks is that because they have distinct service cultures, they sincerely believe that their way of war and their contribution is the best one. And you know, you pick, right? So if you go into the Navy, it’s because you believe in the need.

Jeffrey Lewis: And you’re inculcated with a worldview. So I don’t think people are purely craven, but I do think that it’s very telling that the thing that I say that makes people the angriest is when I suggest the US is about to buy a new ground-based ICBM. We have a perfectly serviceable missile on our submarines that you could put in those silos. It would cost a fraction of replacing the ICBM with a new one. And the Air Force hates it. And I’ve had this argument over about 10 years now, and every two, three years, the Air Force has a new reason they can’t do it. It’s never the same reason.

I think what it ultimately comes down to is that it’s just this service pride and prerogative and autonomy that we see play out. I mean, honestly, we see it in counterforce. The reason that the Strategic Air Command embraced counterforce when it was presented to them was precisely because the Navy had started to get in the business of deterrence, and the Navy systems were inaccurate and they could really only hit cities. And so this was a moment in which Strategic Air Command could say, “See, we have something better.”

So you have all these Navy officers who talked about how terrible nuclear weapons were, you shouldn’t target cities. Until they get inaccurate missiles — and then they talk about how minimum deterrence is the only way to go, and we don’t need all these Air Force things. Until those missiles become accurate enough, and then they go right into counterforce. It’s hard not to be cynical.

Jeffrey's top 3 policy ideas

Jeffrey Lewis: There are three things I really want to do, and they go from little to big. Little is: I really think the president should say, “We would never use a nuclear weapon if a conventional weapon would work as well.”

And the reason I say this is, earlier I talked about how often there’s this kind of disconnect between what the president really thinks and what the plans are. And a lot of what we do is try to surface those tensions to resolve them in positive ways. I really think we’ve gotten to a point where nuclear weapons have an almost metaphysical aspect to them. They’re lousy weapons. The military tends to not want to use them in conflict, but we imagine that we’ll invoke them because they have this magical power to make the other side give up, which I think is crazy. I think that’s total non-falsifiable metaphysical mumbo jumbo. Would it? Maybe? I don’t know. There’s a bunch of unanswered questions.

What I really think the value of the nuclear weapons is is the destruction they do, and they should be treated like any other weapon. I don’t think Kim Jong-un cares if he gets killed with a nuclear weapon or a regular weapon. So by saying we’re going to compare these to other weapons, they have real serious political downsides, so we would only use them if they provided some weapon-like quality that was irreplaceable. My guess is the number of targets in the world would drop to 10.


Jeffrey Lewis: Number two is ratify the Comprehensive Nuclear-Test-Ban Treaty. The US, China, and Russia are all modernising our nuclear test sites. If you look at the open source information, it is busy, busy, busy at all three. All of them are preparing to resume nuclear testing in case the other ones do. I think we’re going to stumble into a renewed era of nuclear testing, which will accelerate the arms race. The best way to do that is to ratify the damn treaty that says you don’t, and to try to put in place some additional verification measures and confidence-building measures to stop that outcome from happening.


Jeffrey Lewis: Number three is a little bit conceptual. We’re back to Tom Schelling. Tom Schelling thought the greatest accomplishment that academics made, and the only good thing we ever did in arms control, was convince people that missile defences or ABMs were destabilising.

People think defences are good, and how can my defences be bad? But the reality is, if I have a shield and a sword and you have a sword, I’m going to hit you with my sword and then parry or block your thrust, right?

Jeffrey Lewis: So that view became dominant in the 1960s. It’s why we got the ABM Treaty, and it was the foundation for all the arms control that came after it.

The US left that treaty in 2003. The Chinese and Russians have made it clear that they cannot possibly imagine limiting their offensive systems unless there’s a corresponding limit on the defensive systems that they have to beat. And that makes sense, right? It’s two sides of the same equation. But in America that is political poison.

So if you give me a free hand, a magic wand, that intellectual change of accepting that a world in which we have defences and offences is a world in which they both have the same thing — and that’s a world of an unconstrained arms race. That’s bad. And if you want arms control to work, if you want to limit offensive numbers, you have to have some kind of corresponding limit on defences.

How the nuclear security field excludes outsiders

Rob Wiblin: Earlier you mentioned that people lie to outsiders from this community. Do they perceive themselves as lying because they have an agenda that they’re seeking?

Jeffrey Lewis: No.

Rob Wiblin: OK, can you elaborate on the lying?

Jeffrey Lewis: Well, let me say, some of them may. There is always a little bit of, “I’m smarter than you and I can get away with this.” But normally what it is is that there’s a ritual that must be performed in how one says things, and that ritual is typically enforced.

So one trivial example is if you say, “The United States is going to use nuclear weapons,” you will quickly be corrected. You will be told, “The United States uses nuclear weapons every day for deterrence. The word you’re looking for is ’employ’.” And indeed it’s called the “nuclear weapons employment guidance.”

So it’s not, I suppose, that people know they’re lying — although there are individuals I think who are dishonest — but there is much more than that. There is a kind of culture of, “This is how we talk about these things. This is how we phrase things. This is how we describe things” — and it makes it very hard to dissent and to disagree.

Rob Wiblin: I see.

Jeffrey Lewis: I recall I had an argument with someone once, where I was arguing for a new policy as an outsider, and the person kept telling me that that’s not how they do it. And finally I got frustrated and said, “I know that’s not how you do it. The way you do it is stupid. Do it a different way.” And that person looked at me like this was the most bizarre suggestion. So there really is a kind of orthodox practice that one is expected to adhere to.

And so when one comes in from the outside, I always get a chuckle. Not a negative chuckle. But I see in the EA language a lot of “de-proliferation” or “anti-proliferation.” And that’s the kind of stuff that would immediately get you treated to a very tedious lecture about why the phrase is “nonproliferation” and how nonproliferation really only refers to this very narrow area. And you would just be listening to this with your eyes rolling in your head, like, “This is good faith engagement. OK.”

Misconceptions in the effective altruism community

Jeffrey Lewis: So an example of the kind of thing that popped out at me, that really struck me, is: in a lot of different things I’ve seen in the effective altruism community’s writings, there’s a huge emphasis on counterforce versus countervalue targeting — as though that is a meaningful distinction that can be analytically used to understand nuclear policy and targeting and pathways to war. And to me that’s just crazy. Because while these are real terms that have meaning, those terms are fossils from an era that doesn’t really exist anymore. There’s an entire history behind those terms, and if you know that history, you can use those terms — and I use them and other people use them — but that history is so complex and weird and specific that they’re really limited in what they mean.

And so just to explain that a little bit, or unpack that: if you look at what the United States says about its nuclear weapons today, we are explicit that we target things that the enemy values, and we are also explicit that we follow certain interpretations of the law of armed conflict. And it is absolutely clear in those legal writings that the United States does not target civilians intentionally, but that in conducting what you might call “counterforce,” there is a list of permissible targets. And they include not just nuclear forces. I think often in the EA community, people assume counterforce means nuclear forces, because it’s got the word “force,” right? But it’s not true. So traditionally, the US targets nuclear forces and all of the supporting infrastructure — including command and control, it targets leadership, it targets other military forces, and it targets what used to be called “war-supporting industries,” but now are called “war-sustaining industries.”

Rob Wiblin: So that would include all kinds of industry, potentially?

Jeffrey Lewis: Factories, yeah. And so there of course is this constant effort to scrub targets, and move things in and move things out, and reinterpretation. But there just is not this clear distinction. And my favourite example of this, which is now archaic — I mean, here I am talking about fossils — but when the Soviet Union deployed its nuclear-armed anti-ballistic missile system around Moscow, there was a particularly difficult-to-destroy radar that had several dozen warheads assigned to it, and it was in the suburbs of Moscow. So yes, we don’t deliberately target civilians, but on the other hand…

Rob Wiblin: But when bombs start landing in Moscow, it might not be a distinction that the Soviets were super interested in.

Jeffrey Lewis: And it’s really important to understand the reason that this distinction came about was because — and honestly the reason it popped into my mind, and I perhaps uncharitably compared the EA community (which contains a bunch of really nice people) to General Power, who nobody thought was very nice, so it’s a little unfair — but the whole reason that the idea of countervalue existed is because very early on, Strategic Air Command had a fairly simple idea, which is: in a war, we’re going to use everything we have and we’re not going to hold anything back.

And that horrified a lot of civilian strategists, including Bill Kaufmann — who’s sort of the hero of the book Wizards of Armageddon and was my advisor’s, John Steinbruner’s, advisor. And he argued that that’s suicide. If we start hitting Russian cities, they’ll start hitting US cities. So he was one of many analysts who started to argue that we should try to at least refrain initially from hitting cities in the hopes that the Soviets would do the same.

And it’s important to note, we didn’t know if they would do the same. We didn’t know if the same thought had occurred to them; it was pure speculation. And initially Strategic Air Command resisted it. And then they realised that if they embraced counterforce, their list of targets would grow very, very long and they could have all the bombs they wanted. And the McNamara Pentagon — which initially embraces new cities and counterforce, and it’s still with us today in that sense — ends up being not very happy about what they thought was a rather clever trick to rein in the generals.

Articles, books, and other media discussed in the show

Jeffrey’s work:

Arms control and nonproliferation:

Recommended books and articles:

Recommended movies and TV shows:

Other 80,000 Hours podcast episodes:

Everything else:

Related episodes

About the show

The 80,000 Hours Podcast features unusually in-depth conversations about the world's most pressing problems and how you can use your career to solve them. We invite guests pursuing a wide range of career paths — from academics and activists to entrepreneurs and policymakers — to analyse the case for and against working on different issues and which approaches are best for solving them.

The 80,000 Hours Podcast is produced and edited by Keiran Harris. Get in touch with feedback or guest suggestions by emailing [email protected].

What should I listen to first?

We've carefully selected 10 episodes we think it could make sense to listen to first, on a separate podcast feed:

Check out 'Effective Altruism: An Introduction'

Subscribe here, or anywhere you get podcasts:

If you're new, see the podcast homepage for ideas on where to start, or browse our full episode archive.