Dismal Global Equilibria

By Sean Carroll | June 14, 2012 12:00 pm

The Civilization series of games takes players through the course of history, allowing them to guide a society/nation from way back in prehistory up through the near future (say, 2100). You develop technologies, choose political systems, and raise armies. There are various ways to “win” the game: military conquest, achieving a just and happy society, or building a spaceship that will travel to Alpha Centauri. It’s a great pastime for any of us who harbor the suspicion that the world would be a better place if we were installed as a benevolent dictator.

Although the game is supposed to take you to the near future, apparently (I’ve never played) you can keep going if you choose to. Which is exactly what one commenter at Reddit did: he has been nursing a single game of Civilization II for ten years now, bringing his virtual global society up to the year 3991 AD. (Via It’s Okay to Be Smart, a wonderful blog.) At which point we may ask: what have we learned?

The news is not good. If you’ve ever read 1984, the outcome will be eerily familiar. I can do no better than quote:

  • The world is a hellish nightmare of suffering and devastation.
  • There are 3 remaining super nations in the year 3991 A.D, each competing for the scant resources left on the planet after dozens of nuclear wars have rendered vast swaths of the world uninhabitable wastelands.
  • The ice caps have melted over 20 times (somehow) due primarily to the many nuclear wars. As a result, every inch of land in the world that isn’t a mountain is inundated swamp land, useless to farming. Most of which is irradiated anyway.

It gets better from there.

What we actually learn about is the structure of the game. We have one player against the computer (who manages multiple civilizations), each with certain goals — a paradigmatic game theory problem. Such games can have “equilibrium strategies,” where no player can make a unilateral change that would improve their situation. Assuming that this player isn’t simply missing something, it’s likely that the game has reached one such equilibrium. That could be the only equilibrium, or there could be a happier one that might have been reached by making different decisions along the way.

What we would like to learn, but can’t, is whether this has any relevance to the real globe. It might! But maybe not. The Earth isn’t a closed system, so the “escape to another planet” option is on the table. But the Solar System is quite finite, and largely forbidding, and other stars are really far away. So limiting our attention to the Earth alone isn’t necessarily a bad approximation.

Right now the human population of the Earth is very far from equilibrium, either politically, or technologically, or socially, or simply in terms of sheer numbers. A real equilibrium wouldn’t be burning through finite resources like fossil fuels at such a prodigious rate, continually inventing new technologies, and constantly re-arranging its political map. But it’s possible (probably unlikely) that we could reach a quasi-equilibrium state in another couple of centuries. With a system as complicated as human civilization on Earth, naive extrapolation of past trends toward the future isn’t likely to tell us much. But “sustainable” isn’t a synonym for “desirable.” If there could be such a near-term equilibrium, would it be a happy one, or the game-prognosticated future of endless war and suffering?

Not clear. I have some measure of optimism, based on the idea that real people wouldn’t simply persist in the same cycles of conflict and misery for indefinite periods of time. It only seems that way sometimes.

CATEGORIZED UNDER: Entertainment, Technology, World
  • http://gnomonicablog.com Fernando Curiel

    Well, all this is well known I think but… Still shocked me to read it from you!
    I hope you can still do cosmology having this perspective in your head. Because the conclusion/hope is not that scientific…

  • Chris

    A strange game. The only winning move is not to play.

    Civilization II, not to be confused with a Type II civilization

  • GM

    But itโ€™s possible (probably unlikely) that we could reach a quasi-equilibrium state in another couple of centuries.

    I highly doubt we have that much time – if present trends continue, it is game over long before that…

  • David

    I’m sorry but that guy sucks at Civ II. I always won by conquering all the other civilizations using the “stack of doom” strategy. If he still has 3 “super nations” then he’s just playing a defensive game where he’s not trying to conquer anyone he just fends off any attacks. That’s the boring way to play.

  • AI

    It’s just a game.

  • Rationalist

    The guy sucks at Civ II. When he posted the savegame file to Reddit someone called Stumpster managed to beat the other two Civs in just 58 years.

    The entire internet is reading a hell of a lot into this, but when you go to the Reddit thread and read about it, you find:


    “Lycerius: And the howitzers; I had no idea. I would build legions of tanks with a howitzer every now and then. Smash into the Vikings, and be repelled. I especially like how you flanked the Vikings just south of those choke points. That’s where my doom was always sealed. Glory to Comrade Stumpster’s 58 year plan!”

    The guy was attacking with tanks instead of howitzers for two thousand in-game turns without noticing that an offence strength of 12 (howitzer) is better than 10 (tank). Duh. Plus I think the Howitzer ignores city walls.

    Basically, the AI in Civ II finally found a human who was exactly as bad as it. And in any case the game artificially stops all technology after the present era.

    There are no grand implications for the future of earth. Just an example of how gullible and lacking in critical thinking people are when you feed the a “cute” story.

  • Entropy

    @David I believe he won the game in the 20th century and then continued playing, and he can’t use stack of doom because each civ has multiple nuclear weapons. As soon as your units leave a city and aren’t protected by SDI, they can be targeted by a nuke.

  • Rationalist

    @Entropy: “canโ€™t use stack of doom because each civ has multiple nuclear weapons”

    Wrong. Read the post I linked to.


    stumpster beat the American Civ with a massed attack,

    “I got the jump on them with a majority of my Howitzer army and knocked them from ~430 units on map to under 250 in a turn”

  • M. Chen

    I fear that Sean’s post is based on some misapprehnsions. The state of Lycerius’ game in 3991 AD is a reflection skill, not an inevitable outcome of extended play. It is perfectly possible, as noted elsewhere in the Reddit comments, to establish sufficiently overwhelming dominance over competing AI cultures that they dare not attack, permitting an infinite peace. (There is no modeling in Civilization of cultural decline caused by decadance, to the best of my knowledge.)

    As Al says, it’s just a game.

  • Gizelle Janine


  • http://theaunicornist.com Mike D

    That’s pretty hilarious to read this now, because last night I fired up my first game of Civ V in several months…. and played for six hours. Oy. Digital crack.

  • Entropy

    @Rationalist Seems to me that he managed it by breaking a peace treaty (“getting the jump on them”) and attacking before nukes could be used. Just marching an army all on one tile across the territory of a nuclear power is a bad idea (except in Civ Rev, where only one nuke can be built anyway). Stack of doom is OP (and game-breaking) until this point.

  • http://lablemming.blogspot.com/ Lab Lemming

    I’ve had CIV2 global meltdown nuclear stalemates on maps where continents were separated by more distance than a transport can move in a turn.

  • Christian Takacs

    I don’t think Civilization has anything to do with modeling human history, but it is like being in a live action version of James Burke’s “Connections” tv series… and is also mind numbingly addictive to play… Just curious, I noticed as the game goes past the alpha centauri mission it goes slower and slower with each turn… I was speculating it was using some kind of matrix to keep every tile change in accordance to the game mechanics and then the AI’s moves with the various civilizations would ripple outwards, then the human players, etc… It almost seemed like watching competing bacteria colonies grow. Is the game based on some kind of digital cellular automata model? I just remember the game coming out around the same time those digital life programs started becoming available on the internet to play with.

  • Meh

    Anyone familiar with Game Theory, the branch of mathematics that can be used to predict the future with 87% accuracy? That’s what we’re talking about here. And it seems like one of the rules for predicting the outcome of our civilization is that people will always be selfish and jealous. Not all people of course, but a surprisingly large number of the population. Until that isn’t the case, this is a possible future for us.

  • IW

    Civilization is just the opposite: it’s not about being civil and creating an idyllic society from which everyone benefits; it’s entirely about promoting endless conflict! No matter how you try to play it you will end up fighting tooth and nail – with societies with which you may well have had previous years of peaceful coesxistence and cooperation – and for no good reason at all!

    So it’s no wonder that the “decade player” has such a situation, but evidently he never thought of cleaning up the pollution, either.

    But the game isn’t about rela life. Not even close. Pretty much every encounter, including initial encounters, with other societies begins with mindless, meaningless threats. Cooperation, pacification, gift-giving, whatever, gains nothing in return, so war becomes increasingly the only option, and once you find yourself forced into a war with one other society, several of the other societies pile on, including societies you’ve hitherto considered friendly.

    So-called “Civilization” is more like a first person shooter game than about anything the name might conjure up. The very game rules, down to the movement of pieces is *designed* to create conflict.

    Having said all that, it’s fun to let loose and go psycho once in a Blue Moon! Rampaging across the globe is a great way to unwind after a rough day at work. Other than that, though, “Civilization” is just a waste of time and brain.

  • Simon

    Maybe Marx will be right and after the productive forces of society have been fully developed we will abolish private property and live in harmony!

  • Low Math, Meekly Interacting

    OK, so clearly this is a crap simulation, albeit an entertaining diversion for those times when you gotta nuke something. Any games worth playing in this genre that aren’t crap simulations?

  • http://lablemming.blogspot.com/ Lab Lemming

    As in any other artistic endeavor, a video game is an expression of how its creator sees the world. In the case of pollution, it was added to the game because the guy who made the game (Mr. Meier, in this case) thought it make the world more interesting, or more fun, or more resounding with his potential audience.

    For those unfamiliar with how Civilization works, pollution can come from three places:
    1. Cities with lots of industrial production.
    2. Cities with large populations (esp. after the invention of cars)
    3. Nuclear attacks and meltdowns.
    If you let pollution accumulate, then you get an episode of climate change (about one climate change event per 10 polluted lands), which turns coastal cropland to swamp and inland plains to desert.

    In the scenario linked, the player hasn’t the resources to clean up pollution as fast as it is produced by nuclear war, so there have been many rounds of climate change.

    The interesting statistic shown there is the population: 15 million people. The carrying capacity of a CIV2 planet is generally in the hundreds of millions range. So the out-of-control nuclear stalemate has reduced the carrying capacity by more than 90%.

    Whether or not this is ‘real’ is anyone’s guess; it was just a feature that the game designer put in to represent the side-effects of large scale nuclear war.

    How about a nice game of chess?

  • Gizelle Janine

    The only thing I heard was “Blah-blah-blah Diablo 2” No offense…

    Did anyone else get that memo? ๐Ÿ˜€

  • Fjord

    Glad everyone agrees the real problem here is that this guy isn’t very good at Civ II

  • amphiox

    Actually I think the human player’s deficiencies at the game actually make the results more interesting. Any game equilibrium can be broken by a discrepancy in skill between the players, which is why in game theory one always presumes perfect play, and in Civilization good human player will always beat the AI.

    Since I doubt anyone knows what constitutes “perfect” play in Civ II, from a game theory point of view, the next best thing is to have human player lower his quality of play to more closely match the AIs, and see what happens.

    Back when I played Civ II, I often pulled my punches in the early game to let the computer controlled opponents grow strong, in order to get a more interesting mid to late game to play (when I wasn’t just interested in going the warmonger route and crushing all beneath my boots as quickly as possible….)


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Cosmic Variance

Random samplings from a universe of ideas.

About Sean Carroll

Sean Carroll is a Senior Research Associate in the Department of Physics at the California Institute of Technology. His research interests include theoretical aspects of cosmology, field theory, and gravitation. His most recent book is The Particle at the End of the Universe, about the Large Hadron Collider and the search for the Higgs boson. Here are some of his favorite blog posts, home page, and email: carroll [at] cosmicvariance.com .


See More

Collapse bottom bar