I recently said that you should pay attention to questions designers are asking. But what I didn’t say was that often, these are questions the players have given them. I’m pretty sure this is what has given us such a lackluster roster of Pacific theater wargames over the years. “How do these carriers fight??” “Where are the critical hits??” Reasonable questions to ask about a campaign marked by carrier battles by a generation brought up reading Walter Lord’s Incredible Victory. So designers chased mechanics that allowed players to recreate specific encounters with ships and planes, while not quite knowing how to integrate this into a larger context. The logistical considerations, huge distances, intermittent pace of fighting, and lack of clear front lines all conspired to stymie designers from the beginning of the hobby. SPI’s infamous U.S.N. showed how back in 1971 there just wasn’t the development skill and mechanics vocabulary to address such a complex design problem. John Prados, designer of the revolutionary Rise and Decline of the Third Reich, failed utterly to adapt his economic system to the Pacific with 1977’s Pearl Harbor. That same year, a game that focused almost entirely on the ships fighting came as close as anyone could for a long time. And no one calls Victory in the Pacific a realistic wargame.
But wargame traditionalism is an unforgiving religion, and asking questions about the most dramatic elements of Pacific combat ensured that designers would continue to try to answer the questions that they thought their players wanted answered. Many focused on overlaying the hex-and-counter rubric onto millions of square miles of ocean across years and years just to focus on what a couple of ships would end up doing in fifteen minutes. Then, in 2005, a designer who knew how to ask strange questions asked his players to answer one out of left field: “How do headquarters work?”
Mark Herman’s Empire of the Sun came out fourteen years ago, the same year that Twilight Struggle started a revolution in gaming. Both games used cards to drive an unusual set of systems, but while Twilight Struggle leaped to the top of the Boardgamegeek ratings and became a household name, Empire of the Sun was greeted by some befuddlement. While there were players that realized that the game had done something extraordinary, the loudest complaints were by people who claimed that “the rules don’t work,” most likely because they didn’t like what the rules were saying. I admit that while I didn’t feel the rules were broken, the complaints about the game kept me from committing to it with an opponent in the way you need to in order to really evaluate how well a game plays. I no longer have my 2005-vintage copy, which means I likely sold it or donated it to someone, and I thought no more of it until 2014.
The years 2005 and 2014 are over three decades apart if you just count them in game-design years. In 2005 we got games like Caylus, a turgid worker-placement exercise that hoodwinked us with its (relative) interactivity into thinking there was something there. By the time 2014 rolled around, we were being treated to masterpieces like Dien Bien Phu: The Final Gamble and A Study in Emerald. I attribute this to the fact that game design is an iterative process, and that iteration not only takes a while for designers, but it takes a while for players to get on board. Empire of the Sun was a contemporary of Caylus, and much of the game audience then didn’t know what to make of a game where ships and planes and ground units could all fight in the same battle. Or where large fleets could meet and do little or no damage to one another by rolling a “0.” Something about the rules. They had to be broken. Surely, the designer didn’t mean to do this.
The years have wizened not only designers, but players. I attribute some of this to the natural process of experience, but hastened by the faster diffusion of information. In 2015, I prepared for a game of Churchill not by reading the rules, but by watching a video of someone else playing. While BGG was around in 2005, rules are now expected to be available online well before a game is published, and people will go through them in detail. Just like you expect your doctor to see you today instead of in two weeks, you also expect your opponents to have watched the same hype videos you have and to have read the latest errata. It’s a whole new game space.
So the second edition of Empire of the Sun came out to a dramatically different audience than its predecessor had, even if many of the faces in the audience were exactly the same. Transformed by a more robust hobby in general, gamers were eager to learn new things. Such as how headquarters worked.
Mark Herman had already answered a lot of questions about the Pacific war in his 1985 design, Pacific War. But that game was singularly focused on the operational level, with battles like Eastern Solomons taking hours. If the Eastern Solomons took hours, how long would the whole war last? Herman admitted that while Pacific War did the operational scale exactly as intended, the result was that the strategic level was too big and long. So twenty years later he devised a system in which turns represented three months and a single battle lasted as long as it took to roll a die. Which probably rubbed a lot of people looking for Midway recreation the wrong way.
Herman’s answer requires some abstraction, which drives a lot of people crazy but what are you gonna do. The card-driven system he uses gives each side its own deck, which is now out of fashion with the proliferation of COIN but works incredibly well here. You can play cards for events or operations, just like in other card-driven games. But many events are operations in themselves, allowing you to “activate” more units than you could with an ops card, and conduct more battles. The key is that in a given round, those are the only units you will use. In the whole Pacific theater, in that round, you might only move three units. Your opponent may only move a few, or none. Each time you do something, it represents a discrete thing. But that thing may change the next time you do it. Activating several air units and attacking Rabaul may represent intense strikes over several days, or it may represent a months-long campaign to wear down airpower. Or it may be the covering action for a naval battle that is fought in the same round but on a different “day” in the history being represented. The mechanics are the same, but the interactions dictate the context.
I think this is the biggest problem many people have with the game. If you activate a unit for an amphibious invasion, and end up fighting an air/naval battle in the hex you are invading, and you lose, your invasion force is simply sent back to port. This represents the compression of events into a single round that actually took place over separate days. Your invasion force may have sailed with your ships, but it would not have tried to land until you had defeated the enemy navy blocking its way. Just because the ships “moved together” doesn’t mean they “moved every single nautical mile next to each other.” Which is hard to reconcile with the rule that a ground unit invading a hex defended by enemy naval units must be accompanied by escorting naval units not just for the invading units entire path to the target, but for the naval units’ entire move as well. Someone once insisted to me that the discordance of these two rules meant the game made no sense. “You can’t just say this means this and then that it means that,” my literalist buddy concluded. “It’s dishonest.”
Or worse, abstract! I suspect that was his real objection. I, on the other hand, understood the restriction on naval unit movement during invasions as a requirement that you start the round with your naval units in the same port as your invading ground units if you expect naval opposition. Which is really a planning requirement, not a movement requirement. Could units have physically just sailed across the map in one round, picked up some invading flotilla, and accompanied it to the target? Well, sure, in theory. But what is being represented is the planning necessary to coordinate an operation where a bunch of disparate elements concentrate on a single target in a battle that might last several weeks or more if you take into account the multiple points of interaction. You don’t just “show up and go,” no matter where the cardboard is sitting on the mapboard and how many movement points it has.
“That’s great,” you say. “Abstraction. Victory in the Pacific is kind of abstract, even though it has silhouettes of ships that shoot dice at each other. It sounds like you’re just hand-waving some explanations of why this game is inconsistent in its scale.” No, I’m explaining the reason the game works once you understand what the system does. This is because discrete activations allow you to treat individual operations as coherent entities, as long as you solve the problem of battle resolution. Which for many years was a design problem that defeated all comers. Herman comes up with the idea of the “air/naval battle” and lumps all the air and naval units in combat into an additive total. Your Junyo, my Marine air wing and B-25s, your Hiei and Kongo, my Enterprise. After some simple modifiers, we each roll a die. The number of factors you end up with is how many “points of damage” you do to the other side. But that die roll determines how many factors you actually have. Roll high enough, and you will do as much damage as you had combat factors. Roll lower, and you may only do half as much. Roll low enough, and it’s a quarter. Ouch!
This highly variable combat system is the heart of the game, and will make some people crazy. I agree that it can be disheartening. But I played a 33-hour campaign game with a friend of mine over VASSAL last summer, and while it was withering to watch my low rolls cascade down the screen, it never soured me on the game itself. Because the interaction of this variability is how you get Midway. The Japanese roll low, even though they have a huge fleet, and the Americans roll high (with some modifiers for Ambush). High enough to get “critical hits” that allow them to sink the Japanese carriers despite a casualty system that gives you strength in numbers because you normally can’t sink a ship until all other ships (and planes) in that combat are damaged. This, as far as I’m concerned, is the only way you can account for operational intelligence on this time scale. Hidden movement would be incredibly cumbersome—you’d have to switch from hidden to non-hidden movement all the time—and ahistorical—since the location of most ships or force concentrations over a three-month time period was generally known. I’m not sure why wargamers are willing to accept a die roll of 6 on a 2:1 against Tobruk but not a couple of 0’s in the Marshall Islands.
Further, this card-activation system has a built-in intelligence mechanism: each card has an intel level that is associated with its play. Cards that allow you to activate more units have a higher intel number, which gives the opposition a better chance to detect it. This means that large operations will be hard to keep secret. Also, because your movement points depend on the ops value of the card you play, operations conducted over great distances will be harder to plan (lower frequency of high-value cards) and will give the enemy greater flexibility in response.
Why greater flexibility in response? Because the number of movement points a unit has is variable, and as I said above, depends on the ops card you play. More importantly, the number of movement points your opponent has is determined by your ops card. So a large operation takes more time, and the enemy is more likely to detect it, and thus marshal more forces to contest it.
It’s the flip side of this, though, that shows how brilliant this ops mechanism is. In most card-driven games, low-ops cards are just low-ops cards. There is nothing about them that is better, except that in single-deck games they may have opponent events that are weaker because they are low-ops, and thus less dangerous to play. But the card is still worse than any high-ops card you could theoretically have. However, in Empire of the Sun, low ops cards are valuable precisely because on your turn, they restrict your opponent’s movement points as well as yours. So if your main force is more proximal to a target than your opponent’s is, you may be able to achieve local superiority simply because your opponent, while having detected your operation, does not have adequate forces in range of your target, while you do. If you had played a higher-ops card, your opponent would get those movement points to use. By using the low-ops card, you can prevent him or her from reacting. It’s a way of making low-ops cards valuable while being completely consistent with the game’s theme and conceits.
“But Bruce,” you say, “I thought this was all about headquarters, and you’re just going on about cardplay and dice. I thought there was something revolutionary here.” If you don’t see the revolution I’ve described, please sit down and play the game, because you will. Once you learn about how headquarters work.
Although Air Zone of Influence is pretty great, too. It’s an elegant abstraction (that word again) that clearly and cleanly demonstrates the central importance of airbases in the Pacific campaign. In fact, this mechanic underpins the giant history lesson you get when you start playing the game and realize, oh, I see why the U.S. didn’t try and capture every island. Or why Rabaul was so important. Or Guadalcanal. Or, really, everything.
Because within a two hex radius of every air unit (and aircraft carrier!) is an “Air ZOI” that prevents the tracing of supply and communications unless it is “neutralized” by friendly Air ZOI. Since we’re such good friends we can just call it AZOI. But the point is that as air forces establish themselves, the freedom of action for enemy forces decreases.
That this folds directly into the intelligence system shows how well integrated Herman’s mechanics are and, as a consequence, how easily good strategy flows from basic elements. As mentioned before, because this is really a CDG and each turn is driven by card play, every card you use will have an intel value that dictates how likely it is for your opponent to detect your move and react to it. And if you conduct an operation within your opponent’s AZOI, he or she will get a bonus to the intel roll. Although even if your intel roll succeeds, it can be done in by your headquarters.
What? Something about headquarters? Oh yeah, I was saying. Everything I was talking about above needs to happen through a headquarters. These headquarters have a range and a logistics value. If a headquarters is in supply, it can activate units. If it isn’t it can’t. If a unit is within range of a headquarters, it can be activated. Some headquarters have a relatively short activation range and low logistics value. Some have a much longer range and higher logistics value. The location of your headquarters will dictate your options in different parts of the map, as well as the enemy’s distribution of AZOI. When you react to enemy offensives, you’ll be limited by the logistical ability of your headquarters. The units themselves play the game, but the headquarters are pulling the strings.
This focus on headquarters drives play in many ways. In the early game, the Japanese are working to render the disparate Allied commands (Malaya, Dutch East Indies and the Philippines) ineffective by putting their headquarters out of supply. The Japanese are working to maximize their effectiveness though three distant HQs with varying logistics values. The Japanese Combined Fleet HQ is extremely valuable because it has the longest range and the highest logistic value. Furthermore, the map is so well laid out and finely tuned that the event card that renders Admiral Yamamoto killed has a huge effect on the game simply by reducing the range of Combined Fleet HQ by one hex. And the logistics by one also, although I’d argue the former is actually worse. It’s hard to make a game where the distance between every single hex on the map seems to be purposeful and not just a geographical accident, but somehow Empire of the Sun does it. It’s a textbook example of the confluence of design and playtesting.
If you just look at the way the gameplay fits together, and understand it, you have a pretty hard-to-beat solution to the Pacific war problem. But there were some other issues that people brought up back in 2005, and still bring up to this day. One of these is “The Bomb.”
In Empire of the Sun, it is possible for the atomic bomb not to be deployed. There is a certain sequence of cards which require Tojo to resign and the Soviets to invade Manchuria. If this happens, then if the Japanese mainland is in range of a US B-29, has been successfully strategically bombed for four game turns in a row, and has one or zero resource hexes, then The Bomb is dropped and the Allies win.
If this doesn’t happen, the US may be able to impose a “blockade victory” in which no resource hexes are able to trace a supply line to Japan for three consecutive game turns. If this doesn’t happen, either, the US needs to invade the Japanese home islands to win. This means controlling every hex on Honshu. Which has an inherent defense strength in addition to whatever is left of the Japanese army. It can get bloody.
What this does, as Mark points out clearly in his designer note on the victory conditions, is to “force the historical mindset on the Allied player that an invasion of Japan had to be contemplated and planned for, especially since most wartime personages were unaware of the Manhattan Project until the A-bomb was ready for operational use.” But having cards dictate that the atomic bomb doesn’t get dropped Drove Some People Nuts.
I attribute this to the difference in wargamers between what I call “historical recreationists” and “historical strategists.” Historical recreationists are very uncomfortable leaving out events that occurred because there is a certain determinism to their interpretation, or some imaginative verisimilitude that is violated when major events don’t happen. Historical strategists are fine with historical divergence as long as the decision space is the same as it was historically.
This dilemma is also apparent when it comes to the victory conditions. In Empire of the Sun, the U.S. can lose the game by running out of political will, which represents a collapse of national morale. Mark Herman addressed this in his designer notes to the first edition back in August 2004 (available in full on his webpage):
The key challenge in the game was how to deal with the victory conditions. The historical reality is the Japanese never had any chance of winning the war. The U.S. never devoted more than 20% of its overall resources to the Pacific War, so once Germany was defeated it was only a matter of time until Japan would be defeated. The solution was in how to define Japanese victory. The Japanese intellectually, if not emotionally, understood that they could not defeat the United States in a long war. They felt that if they could make the U.S. pay a prohibitive cost for its inevitable counter-offensive they could coerce a negotiated settlement that would allow Japan to ‘legalize’ some of its key conquests.
As unrealistic as this notion appears in historical hindsight, it was the ultimate solution to this design issue.
As Mark says, the concept of the U.S. losing on political will failure seems improbable in hindsight. However, as the years have passed and some of the mythology surrounding the war has dissipated, this turns out to mirror the thinking at the highest levels of the U.S. government and military. I recently picked up Waldo Heinrichs and Marco Gallicchio’s Implacable Foes: War in the Pacific 1944-1945 (Oxford University Press: 2017) and found the authors establishing this in the first chapter:
A second principle was the belief that popular support for the war would not be without limits, in terms of either the amount of sacrifice or the length of time. Roosevelt worried that the American people did not understand the level of sacrifice needed to defeat the Axis in a worldwide struggle. He frequently spoke publicly about the danger of public indifference and what he referred to as the “peaks and valleys” of overconfidence and despair. … Lagging progress, dispiriting casualties, and unfulfilled objectives would dissolve national unity and produce pressure for a compromise peace, one that would leave Japan essentially unchanged. In the estimation of JCS leaders a five-year war would be problematic, a six-year war–one that went in to 1947–unacceptable.
I’m fine with the political will victory condition because it validates one aspect of Japanese planning that shaped the context for their strategy, while not denying the basic fact that gauged by the standards of a traditional war, the Japanese had no chance. Empire of the Sun gives each player historical headaches that mirror those of their real-life counterparts without clear answers, which is more than most historical games achieve.
I eventually try to teach Empire of the Sun to everyone who comes to my house to play games. I am firmly convinced that you can teach the game’s general outline and most important rules in 30 minutes. After that, it’s just a matter of layering on more specific cases.
So why am I reviewing a game fourteen years after it was released? Well, I happen to believe that any game deserves a review, no matter hold old or new it is, because part of my enjoyment of boardgaming is comparative design analysis and historical evaluation. Empire of the Sun doesn’t just hold up remarkably well after almost a decade and a half–it actually looks better compared to what has followed it. But for those who insist on the new and shiny, this is a particularly good time to revisit Empire of the Sun, because just last month, GMT Games shipped the third reprint (technically the second reprint of the second edition). And unlike most reprints, this one has new components and additional content. It includes the “South Pacific: Breaking the Bismarck Barrier” scenario published in c3i Magazine #30, including a copy of the map. It has setup cards for both sides, just like games had in the olden days. And it includes an expanded version of the Erasmus solo bot, for those who want to play without an opponent. All wrapped up in a deeper 3” box to accommodate all the goodies. In fact, this game has something to offer everyone, including a tremendous VASSAL module by Francisco Colmenares (massively updated within the past few days, in fact) for those who are separated by distance but not desire.
As a design, Empire of the Sun has slowly crept up my top ten all-time wargames list since I started devoting the attention to it that it deserves. Given the game’s current level of polish, I don’t think much is going to change rules-wise from here on, and the inclusion of the much shorter but quite engaging South Pacific scenario gives players quite a few options when it comes to learning and playing the game. It’s an incredible achievement to have put the Pacific war onto paper and cardboard in this way. Just keep in mind developer Stephen Newberg’s comment on consimworld back when the game was first published in 2005:
Guys, this is a strategic game. Virtually grand strategic. The resolution systems have some nitty and gritty in them to give feel and to get things to come out right at the far end, but at times that gets it a bit too easy to forget that we are dealing with many weeks of time and huge amounts of men and materials for each resolution. We may call it a “battle” in a particular hex, but in fact it could easily represent more than one engagement of some or all of the forces involved, directly or in support functions, over that long stretch of time. Do not get fooled by the levels of detail in the various systems and resolution, it will lead you astray. These are just mechanics to get the iterations of the forces involved to work out correctly. Concentrate on the strategic, and you will come out ahead every time, both in understanding the rules and in playing the game.
That’s what Empire of the Sun does better than any Pacific theater game available. Playing it is a sublime experience.