A Serious Game

00:00:15
Speaker 1: Bushkin. In C. S. Lewis's famous book The Lion, the Witch and the Wardrobe, the Pevensey Children happen upon the magical land of Narnia, entering through a secret passage in an English country house. When I spoke with Uniwong about her own adventures, that story was the first thing she brought up until recently.

00:00:40
Speaker 2: The path into war gaming is similar to the path to getting to Narnia. You stumble upon it by accident. It's not of your own design. You don't even know it exists.

00:00:50
Speaker 1: Narnia an imaginary world that teaches us something important about the real world. Uniwog zoom background is a steal from Star Wars R two D two, standing next to C three po. She's part of the group of intellectual scholars PhDs strategists who spend their careers helping the Pentagon think about how to better wage war and more importantly, how to better avoid war, and how do they do that by playing games.

00:01:29
Speaker 2: Someone said I should go to the Connection's worgaming conference. So that was in twenty eleven. But I walked into this room of middle aged men in baseball paths, hunched over miniatures, rolling dice, and I suddenly had this sense that I had come home and these were my people, and this is what I was supposed to have been doing all along, and I didn't know it.

00:01:50
Speaker 1: Why was it You walk in that room and you say, oh, this is where I belong. Can you explain why you had that feeling?

00:01:57
Speaker 2: I think it's the this is the It's the geekdom, right, Like geeks recognize each other at a deep level. So I recognize this brand of geeks at a deep level that I didn't realize existed.

00:02:15
Speaker 1: I'm fascinated by the idea that there's a certain kind of insight that you can only get from a simulation. Can you explain why is that? Why can't Why can't I look at a scenario with a bunch of smart people and without a simulation, come up with all of the different angles in a dance just on our own. Why do we need a wargame?

00:02:33
Speaker 2: I think we will invoke Thomas Shelling as is appropriate for nearly all occasions related to deterrence and wargaming and anything I think truly imaginative.

00:02:42
Speaker 1: Thomas Shelling was an economist, a Nobel Prize winner, and the spiritual godfather of the wargaming world.

00:02:49
Speaker 3: Game theory is merely the study of how rational people interact when they know that their decisions impinge on each other, when people have to anticipate what another will do or what another is already doing, or when people try to influence the choices of others through threats and promise and various devices to change their expectations of how one will behave.

00:03:14
Speaker 2: He once wrote that no one, no matter how intelligent, can make a list of the things that would not occur to them.

00:03:21
Speaker 1: And if we've learned anything recently, it is surely that there is an ever expanding list of things that have not occurred to us. My name is Malcolm Gladwell. You're listening to Revisionist History by podcast about things overlooked and misunderstood. This episode is about the idea that the best way to prepare for the unexpected is not to make plans or predictions, but to play games, not board games. Grand theatrical productions costing many thousands of dollars, planned over months, stretching for days, experts and specialists playing each role, each with the passion of an actor on a Broadway stage, with the hope that the rest of us will learn from what they found in their elaborate imagine worlds. When I was young, we lived in a house by the highway with a very long, narrow backyard one hundred and fifty yards deep, lined with what must have been more than one hundred tall pauper trees. In the fall, when the leaves fell, we all realized that raking the leaves would be an overwhelming task. So my father through a party. He was a math professor and decided to invite all of his graduate students in the afternoon. He told them they would deal with the leaves. In the evening, there would be a big dinner and then party games of the sort that eccentric math professors consider appropriate for their graduate students. My father students were all from overseas, from the top colleges of India, Nigeria, Saudi Arabia, Israel. I don't think any of them had ever raked leaves before, but my father believed strongly that a day of year hard work and playing games was as good as life. God and his students picked up on his enthusiasm. The leaf raking turned into sport. The games were hilarious, they convoluted, and the party proved such a success that Every year, a new crop of students would show up, and the leaf breakers from previous years would invite themselves back, until the party became huge, a united nations of math geeks in our backyard, creating huge piles of leaves calling out to each other in strange accents. My working definition of smart at the age of eight or nine was knowing things. But when I listened in on the dozens of Africans and Indians in my backyard, it seemed to me that they defined themselves by what they didn't know. They would never say I just solved such and such problem that was boring ancient history. They would say, I'm working on a problem. There is something that I don't understand that I wish I did, and an even larger group of things I don't yet know that I don't know, with the clear implication the pleasure of finding out the unknown was why they got up in the morning. I hadn't thought of the leaf raking party for a long time until I started talking to the wargamers, when I realized, oh, these people all belonged to the same tribe.

00:06:37
Speaker 4: I think that what wargaming does is by putting people in these simulated positions, these realistic but unreal realities. It encourages them to and exposes them to things they maybe haven't bought up.

00:06:55
Speaker 1: That's David Shlaypak, who was one of the founding directors of the Center for Gaming at the RAND Corporation. RAND is the big think tank out in Santa Monica that's been running wargames for years. Yuniwong worked at RAND, so did Thomashef, author of The Wargamer's Creed. One of the things Shlaypack started doing in his wargames was to make the element of chance tangible. You have one army against another, you make a list of the ten different things that could conceivably happen, and then you roll the dice.

00:07:26
Speaker 4: And the first time we did that, I was terrified. I was thinking, when we go in front of a bunch of people who have experienced combat, who have actually been fighting real wars for the last fifteen years, and we show them that we're going to use dice, they're just gonna laugh us out of the room. And we had all sorts of arguments prepared about why that was legitimate.

00:07:47
Speaker 1: He was going to play the wargame with senior military officers and he didn't know how they would react.

00:07:53
Speaker 4: We were ready to defend it, and nobody batted an eyelash.

00:07:59
Speaker 1: The military guys playing Schlapeck's games didn't want things to be predictable. Predictable they could do on their own. What they wanted was to be forced off the narrow path that logic and prior experience would have them follow. Schleepek was once involved in a game to test what would happen if the Russians invaded the Baltics, Estonia, Lithuania, Latvia, three small countries right next to Russia, all members of NATO. If Russia attacked, it would face the wrath of Western Europe plus the United States. On paper, the wargame seemed to be based on a dumb question. Who would win? We would?

00:08:41
Speaker 4: NATO has thirty five times Russia's GDP. Russia is just emerging from this demographic crisis that is more severe than any that's ever been inflicted on an advanced nation. Their military has been on the downslope for twenty five years. This is a no brainer. Of course NATO's going to win.

00:09:03
Speaker 1: But the first time they played it, NATO didn't win. Russia did now mean Russia would actually win in a battle against the Baltics. No, this was a game, an active imagination, but it did show everyone involved that an absurdly remote possibility had to be taken seriously. Schlapek ran the game again and again. Russia always won.

00:09:28
Speaker 4: Russia's always winning, and in fact we eventually started telling people, look, we've done everything except go out on the street and grab random individuals and ask them to play. And no one's done better than X. Can you do better than X?

00:09:46
Speaker 1: The expectations and predictions of experts are useful, but only up to a point. The war game is where you put those expectations to the test. A few years ago I was invited to join Rand's Board of Trustees. It requires two meetings a year, one at RAND headquarters in Santa Monica, the other at their office near the Pentagon in northern Virginia. Half of every board meeting is devoted to the kinds of things that nonprofits worry about, budgets, fundraising, hiring. The other half is research presentations from RAND staffers, like the electricity grid in Puerto Rico has just been wiped out by Hurricane Maria. How much would it take to fix it? Or do American school children get enough sleep. But my favorite presentations are always the war games. Like this one. The Army comes to RAND with a question.

00:10:57
Speaker 5: They asked us to do a war game type of analysis to understand what it would take to secure loose nuclear weapons in a failed North Korean state.

00:11:07
Speaker 1: That's Colonel John Gentia, retired Army former tank commander. He did two tours in Iraq. He's a doctorate from Stanford and taught history at West Point. He's now a senior historian at RAND, and he gave a presentation to the board about his North Korean game. Wargames always start with a path to war, an imaginary but plausible scenario that explains why a conflict might start. In this case, the scenario was a sudden spike intentions between North Korea and the United States.

00:11:42
Speaker 5: That led to a number of other events where the US starts to mobilize forces in the US ground forces starts putting him on ships, you know, in the ports of Texas in California. North Korea is aware of all of this, tells the US to stop, and it doesn't. And I think at some point in that road to war scenario, we had the North Koreans firing a small nuclear round and just know it was a conventional missile that hit the US island of guang.

00:12:21
Speaker 1: Just imagine that this had actually happened, a North Korean attack on American soil. The world will be in crisis. The US would have to respond. But the problem is that North Korea is a country with an expanding nuclear program, and if a war broke out, a whole lot of exceedingly dangerous nuclear material would suddenly be floating around in the midst of a whole lot of chaos. The army one to know what would happen next, what would it take to round up all those stray nukes. Rand spent months blending the game. Then Genteel and his team flew to South Korea to play it out for the army a big conference room a map of the Korean peninsula on the table.

00:13:09
Speaker 5: In that game, we did use counters like the little square you know wargame that this is a tank brigade, this is an artillery battalion or whatever, with numbers on them.

00:13:19
Speaker 1: Do you have one person playing each actor?

00:13:23
Speaker 5: I think we had two separate teams, and we may have run two different games, but we had the Army colonels play the Blue Military Forces.

00:13:33
Speaker 1: Blue team in wargames speak, is the home team the United States and its allies. The colonels were playing themselves. One of rand's top East Asia specialists, Mike Mazar, played North Korea genteel, and some of his fellow RAND staffers served as referees. In the first run, the US and South Korean forces attacked across the demilitarized Zone, the border that separates North and South Korea, maybe the most dangerous border in the world. Land mines everywhere. The result was catastrophe losses. In the next version of the game, the US and South Korea imagine that they could broke or a deal with a faction of the North Korean Army to let them across, which got them to the other side. But now they had an even bigger challenge.

00:14:24
Speaker 5: I mean, it's freaking hard as hell. I mean, a failed North Korean state with a nuclear weapons writ large. You know, parts of a nuclear device, delivery systems, records, documents of scientists, all those kinds of things to track those down in the chaos of a North Korean state. It was pretty clear.

00:14:44
Speaker 4: What was your conclusion?

00:14:45
Speaker 1: Was it possible?

00:14:47
Speaker 5: Oh, yes, but it won't take weeks, it won't take months. It's gonna take years.

00:14:54
Speaker 1: In the game, it took thousands of soldiers to secure even a single North Korean installation.

00:15:00
Speaker 5: It took that long because you just you have to secure the outside of it, which requires forces to do that. Then you have to move inside of it. Potentially there could be North Korean forces that still wanted to fight to protect it.

00:15:14
Speaker 1: How many sites do we think they have, oh.

00:15:18
Speaker 5: More than one? And then let's not even take into account the chemical weapons, potential biological weapons that North Korea has. But we're talking scores and scores and scores of nuclear sites. Remember from my Board of Trustees talk, I made the point that core nuclear components to one of those nuclear warheads, you could put it in the size of your backpack and so then you could scurry it off on a road with a civilian carrying it. So it just presents all kinds of problems.

00:15:48
Speaker 1: It is talked to the ran boat about the game. Gentiel also spend a surprising amount of time on traffic jams. Because armies aren't the only factor in a military crisis. North Korean refugees would start streaming across the border. South Koreans would start fleeing their capital, so the US and Japanese governments would order the citizens to leave the country at the same time as a huge army is trying to head towards the border, which adds up to the mother of all traffic jams. As I listened to Genteel that day at RAND, I remember thinking traffic jams as a factor in any war with North Korea. That had not occurred to me, and I assumed that was because I'm not a military expert. But then I looked around the room at the people who were particularly Harold Brown, the gray eminence of the American defense establishment, big Leonine Head, thick black rimmed glasses, even in his nineties, the smartest person in any room he walked into. He had a look in his eyes, and I realized, Oh, traffic jams had not occurred to him either. Now, what did the wargame teach the army. They already knew it would be hard to secure nuclear weapons in a failed North Korea. The wargame just told them how hard it would be.

00:17:07
Speaker 5: There's people on the road. If you're passing through the DMZ, chemical weapons may have been used, so the military forces have to put on protective gear to move through a chemically contamidated area.

00:17:23
Speaker 1: Everyone in the tanks has to be in full chemical protection gear, body suit, mask, rubber gloves, boots, what the US military calls MOP for protection.

00:17:36
Speaker 5: Can you imagine what it's like trying to move through this kind of terrain where a Bradley commander or a tank commander is buttoned up and in MOP for and can only look out through the little sites on the top of the turret or even through the optics. And then bring in the fact that potentially there's combat action where US tanks, North Korean tanks, infantry are being destroyed, killed, cries for help, blood, anger, all these kinds of thing things, the anguish of seeing people on the road freezing to death, can't do anything about it. And then there's just the anger and the hatred, the desire for revenge, to kill, And then you're trying to manage all of that and still carry out the mission. It would be easy for the North Koreans to build an atomic ied and to plot that in a choke point going through a steep valley, and to set that off. Then you have that, and then the whole larger operational political context of now a nuclear weapon has been used, what will be the US response? All those kind of things.

00:18:57
Speaker 1: The wargame added detail, color, emotion to the Army's understanding, and the army ordered up the wargame because they knew enough to know that many of those details would be on the long list of things that had ever occurred to them. These are the details that clear your mind of wishful thinking, of comforting delusions, of the false reassurance of overconfidence.

00:19:19
Speaker 5: Sherman said, war's hell, and you cannot refine it. And I've said this before. Look, I mean for the US Army, US military, Iraq and Afghanistan was not easy, right, and it was specially not easy for the populations of those two countries. I mean devastating in a lot of ways. But this kind of conflict war on the Korean Peninsula would make what we did in Iraq and Afghanistan look very, very simple in comparison.

00:19:48
Speaker 1: A wargame is a reality check, And I think we're willing to test our illusions with reality checks only if we practice, if we make it a habit, if we keep reminding ourselves that the world outside our imagination is much larger than the world inside it. This is what the leafreakers tod for all those years ago in my family's backyard. They took a chance on an afternoon of doing something that they had never done before and which did not, at least in the abstract, sound like fun. And then afterwards, a twenty something advanced math doctoral student from Bangalore would sit down with another math whiz from Legos and a third from the West Bank, and they would play Charades on a team together. Charades is a game that depends on shared cultural understanding, but here it was being played by three people with no shared cultural understanding, which is what made it fun. The experience of figuring it out, of learning what figuring it out felt like, was so intoxicatingly unpredictable that they would all come back the next year and the year after that, even after they had gotten their PhDs and moved far away, only to discover when they returned, of course, that Charades had been replaced by some brand new and fiendishly impossible game of my father's invention. All the leaf breakers knew is that they wouldn't know what would happen after the leaves were raked, and that was enough one final wargame, one that I still can't stop thinking about, designed by Uniwong Okay.

00:21:38
Speaker 2: So the wargame began with sort of China trying to exert creater control and signal that they were in control of the region and more dominant now in the region, and the United States and Japan were really resisting that.

00:21:50
Speaker 1: Yuniwog's scenario was set in the future at a time when the world's major militaries have gone all in on artificial intelligence.

00:21:58
Speaker 2: There was this premise that AI is going to fundamentally change warfare. I propose the idea because I was like, well, what happens if all your dreams come true?

00:22:08
Speaker 3: Right?

00:22:08
Speaker 2: There's a lot of work asking whether or not the Pentagon can really integrate AI in the way they hoped to, But I'm like, well, what if you got everything you wanted.

00:22:18
Speaker 1: The game was held in a conference room at rand's Washington, DC office. Someone playing China, another person playing Japan, the US North Korea. A senior information scientist from rand's California office, a Shandi Shaba, flew in to represent the commercial AI sector. Yuniwang was part of what's called white team, part referee, part instigator, The road to war was deceptively simple, lots of muscle flexing between the US and China, leading to Japan and the United States launching a cyber attack on a Chinese aircraft carrier. The Chinese responded with a flyby near US vessels, and with that the game was off and running.

00:23:04
Speaker 2: The U and I States in Japan then went off and then consulted with each other and came back announced they were holding a large joint military exercise around the Sankaku Islands.

00:23:14
Speaker 1: The US and Japan put their missile defense on full auto, meaning they programmed the artificial intelligence systems running the missiles to respond automatically and immediately to any hostile act. They did that to send a message to China, don't do anything rash. But then out of the blue, North Korea decided to join in by firing off a missile on its own, aimed at a major Japanese city. Luckily, the missile was intercepted before it could do any damage. What are they doing, They're signaling that they're standing with China.

00:23:58
Speaker 2: Yes, so sometimes I say China has one ally and it's the worst ally in the world, and it's North Korea. So that is a very complicated relationship. But in the game, they did not pressure them. This was just North Korea decided they would try to be helpful and this was the result.

00:24:14
Speaker 1: But remember in this wargame scenario, the US and Japan had their defense systems on full auto, programmed to respond automatically and immediately to any provocation. Did the US and China in the game. It just never occurred to them that if North Korea weighed in and they were on full auto, we might end up dragging North Korea into the conflict.

00:24:36
Speaker 2: Correct, it was the focus was on signaling to China and nobody thought about potential North Korean behavior.

00:24:42
Speaker 1: So North Korea does this thing, the AI fires back at North Korea.

00:24:46
Speaker 6: Yes.

00:24:47
Speaker 2: China then responded with a limited blockade of Japan. So they announced the blockade, and the AI sent a single destroyer to a single Japanese port to enforce this blockade. This is where the plays for South Korea and Japan said, well, that was a failure. But the United States team was really extremely confused and said, wait, we like, I don't understand what's going on here. I don't understand what the AI is doing.

00:25:12
Speaker 1: God knows what the North Koreans will do next. The US is running around trying to figure out what the Chinese artificial intelligence systems are up to. Japan is under a partial blockade. Oh and there's a cargo ship at the bottom of the sea. The Chinese blew you up right.

00:25:29
Speaker 6: The Chinese blow yet because they were trying to enforce a blockade on Japan.

00:25:33
Speaker 1: I believe that's RAND specialist Ashandi Shaba who flew into play the role of a commercial AI system running container shipping. So you just kept sailing into port. You didn't stop and turn away.

00:25:47
Speaker 6: No, no, I was trying to keep my market open, like serve everybody as much as possible. That that was the idea there, and I figured by being compliant, I could I could keep both sides happy.

00:25:57
Speaker 1: Every one of these wargames that people talked to me about, it's always a mess.

00:26:02
Speaker 6: The artists are strategic insight from the mess, and some people doing better than others. It's always a mess. Everything is a mess.

00:26:10
Speaker 1: At the end. The RAND team put out a report on the game with the following lesson In conclusion, current planning efforts have not kept pace with how to handle the potentially destabilizing or escalatory issues associated with these new technologies. End quote. The wargame said, don't fool yourself. There's going to be chaos, but with luck we can learn from the chaos. When you made your remark about how everything's a mess and yet we continue on, that's like the most Nigerian thing you could possibly.

00:26:50
Speaker 6: I grew up in the heart of chaos. Legos is the most chaotic place on Earth.

00:26:55
Speaker 1: I was going to say, you were well prepared for this.

00:27:00
Speaker 6: I have Legos to thank for lots of things that I've done in my life.

00:27:05
Speaker 1: Basically one moment from Uniwong's wargame, several days in the United States and Japan began hunting the Sea of Japan for Chinese submarines to figure out which ports the Chinese were planning to blockade next. Emotions were running high.

00:27:23
Speaker 2: The player for China said he couldn't tell whether the US and Japanese anti submarine warfare assets were looking for his submarines to just track them or destroy them, so he ordered a submarine to destroy an autonomous anti submarine warfare plane, and then player for Japan felt that they had shown restraints so far, but they could no longer show restraint if the player for China was going to take kinetic action. So the United States in Japan at this point respond and they sink a manned Chinese submarine. So these are the first human casualties of the game. And at this point it escalates significantly. So the players were Japan and the United States go off to a separate room, and then they're trying to figure out a way of offering the Chinese off ramps right like they seriously wanted to de escalate at this point. The player for China at this point says, no, they got one of mine. I'm gonna get one of theirs and launch his missiles at the combined US and Japanese fleet, and we end the game there. But we believe that there would have been human casualties aboard the US and Japanese fleet.

00:28:35
Speaker 1: Wait, why do you guys call the game just at the moment when China is launching missiles at the US and Japanese fleet. Isn't that It's like stopping the movie with ten minutes left. Don't you want to know what happens after that point?

00:28:51
Speaker 2: Yeah, you know, some of it was we ran out of time. We only had the conference room for a week, right, and we only had people in town for a week, So that is sometimes the unfortunate nature of these things.

00:29:04
Speaker 1: It was only a game, after all, not real life. People had to leave Narnia and go home, and the game of Dennet's work hadn't it raised a possibility that might not have occurred to anyone otherwise, a warning that might come in handy one day, or who knows, maybe already has Yuna. You would have loved the Leafreaking Party. Revision's history is produced by me LaBelle, Lee men Giestou and Jacob Smith, with Elouise Linton and on a naim Our. Editor is Julia Barton. Original scoring by Luis Guerra, mastering by Flonon Williams and engineering by Martin Gonzalez. Fact checking by Amy Gaines, and special thanks to the Pushkin Crew. Had a Fame Carl Migliori, Maya Knig, Danielle Lacan, Maggie Taylor, Eric Sandler, Nicol Morano, Jason Gambrel and of course Jacob Westberg. I'm Malcolm Gladmow. Don't forget my latest book, The Bomber Mafia, which is an expansion of several episodes, from the last season of Revisionist History. You can find it wherever books are sold, but by the audiobook at Bombermafia dot com and you'll get a bonus listener's guide, and you can listen in the podcast app you're using now