Flash of Steel header image 2

Bunk, Progress and Process – Games take on history

August 20th, 2005 by Troy Goodfellow · 5 Comments · Uncategorized

History is complicated, but, if we are going to have historical strategy games, it has to be grappled with in one way or another. Game designers always say that when history and fun collide, that fun will win. This is not an unreasonable statement, but it does leave open the question of which sacrifices are made and for what reasons. The game design approach to history can take three forms – dismissal, simplification and embrace.

History is bunk

This quote, attributed to Henry Ford, implies that, ultimately, history doesn’t matter all that much. It is just everything that has happened before and therefore is of little relevance to the here and now. What we know about the distant past can’t be trusted anyway, so game designers and game players can pillage the days of yore for stuff that looks fun and avoid stuff that isn’t.

My personal bête-noire is the elephant unit in strategy games. They are almost always early tank substitutes – strong, durable, scary – even though they have historically proven to be unreliable units except in the jungles of South and Southeast Asia. In the West, they were expensive prestige weapons that were almost never successful in battle.

But elephants look cool, so you have to have them in your game. And if they are going to be in the game, they can’t be useless as history dictates. No one fights with elephants now so a little revisionism won’t hurt anyone. It’s a minor piece of history that no one really cares about so the cool factor can trump reality.

Is any real harm done by the history is bunk school? If it is limited to the invention of new weapons of war (head hurlers and wardogs in Rome: Total War) or the “culture is destiny” approach to distinguishing between civilizations, there is no reason to get too exercised about it. If a game makes pretensions to historicity, it should be judged on these merits, but few do. Sparta: Ancient Wars has made such claims, so it should expect a firm drubbing for its errors.

But what about when a game elides over important historical events in the interests of good taste or to avoid depressing a player? Paradox Interactive has a policy of not including references to the Holocaust or other genocides in its games, though slavery is treated as just another historical fact. Games that deal with the discovery of America tend to stay away from both slavery and the eradication of the native peoples. Designers don’t want to make light of these issues, but aren’t trained in how to deal with them in a sensitive, historical way that makes a game playable.

History is one damn thing after another

This quote is also attributed to Ford, but has been attributed to Voltaire and other historical figures. In this worldview, history is linear – facts and events stack up neatly and there is a preconception that people and civilization is inherently progress oriented.

Nothing exemplifies this outlook as well as the tech tree. Sid Meier’s Civilization is the archetypal game. To achieve higher levels of technology you must go through lower levels. This gives you access to higher forms of government and efficiency. In Civ 2, eras and epochs were introduced, keeping you in specific levels of development until you learned enough to leave it. RTS games since Age of Empires have had the concept of “ages” (naturally). There are higher and lower levels of civilization. Moving as quickly as possible to the higher levels is the best road to success.

This is done to keep the player playing. If you give him/her all the goodies at the beginning of the game, he or she has little reason to go through all the hunting and gathering. “Aging up” or technological progress is also linear to keep the process comprehensible. It makes intuitive sense to think of history as one story of progress and those who progressed faster “won”.

In fact, historical discovery is rarely linear and is never planned out. No one knew that Invention would lead to Gunpowder. Writing was originally intended for accounting, not Map Making. Jared Diamond’s Pulitzer Prize winning Guns, Germs and Steel demonstrated how geographically contingent success was.

In a move that almost conceded some points to Diamond, Civ III placed strategic resources randomly so you could find yourself without iron and therefore limited in what you could produce. The new Civ will make these resources more common and level the playing field, but loosen the constraints on research by not limiting you to an era.

Gamers often complain about how linear certain games are, but they rarely have strategy games in mind. For the amateur, history is a straight line from the past to the present. We, however, see history in reverse. No one in the Reformation knew that the Industrial Age was only a few generations away.

History is an unnatural act

Sam Wineburg’s Historical Thinking and other Unnatural Acts emphasizes how difficult it is to learn real history. Real historians are often engaged with primary materials and reconciling conflicting viewpoints. History is an interpretation of the past more than it is a settled account. We can agree that there was a First World War, but disagree on why it happened or why it turned out the way it did. Was American Independence inevitable, or the consequence of specific policies at specific times?

Historical uncertainty is, of course, anathema to good game design. But there are very few games that actually take a position on history, that try to consciously present it as an interpretation. Most games, it is true, stick to what is known. There were men with sharp sticks who got beaten by men with metal sticks who lost out to men with booming sticks who were eventually run over by metal cars or hit from above by flying fortresses.

Sid Meier’s Colonization will always, to me, be the perfect historical strategy game. Not as a game. It had many flaws and gaming idiosyncrasies that kept it in the “good, not great” category. But it wore its history on it sleeve.

It was the story of America, and by America I mean the United States. The goal was not an empire, but independence. Independence would be rooted in a tax dispute with the Crown. So, if you played the French or the Dutch you would still be as American as the British settlers that liberated the Thirteen Colonies.

This design choice undoubtedly turned off some gamers. The goal was predetermined and the goal was a highly historically contingent one. It was probably partly driven by the fact that the AI was never very good at improving his colonies, so the deus ex machina of a super army teleporting from across the ocean made a good enemy.

But there are all kinds of touches in here that separate it from other historical strategy games. The genocide of American Indians isn’t ignored, but is made a game mechanic dependent on player choice. Just like colonists in the past, tensions will grow with the natives if you encroach on their lands. You can avoid their lands if you like, but the desire to win at all costs will convince some players to grab that silver mine in Cherokee land. Conversion to Christianity is possible, but it’s an awkward tool that produces some resentment and inefficient workers. But it can give you an early productivity boost if you don’t mind convert labor.

Colonization is not history in the professional sense. You are never asked to make an interpretation; the interpretation is all done for you. Slavery is skipped, even though prison labor and indentured servitude remain. (Somehow, wholesale destruction of enemies has been made acceptable but forced labor has not. Games are strange that way.) Colonization does present an interesting interpretation of the colonial experience that required making historical judgments, and not simply doing things the way they were done in Civ or other related titles.


5 Comments so far ↓

  • Anonymous

    Fuck it, rewrite number four.

    You can’t have “historical accuracy” and still have a game.

    Things happen because of other things happening in other places, at other times.

    Give the player choice over what can happen and you immediately lose your accuracy because you’re changing “what happened”.

    The colonial period and settings (America, Africa, India) are the only times and places with even the chance of real historical accuracy because they are the only situations where you have a large enough “army” in the field for long enough to set a long-term strategy game in while AT THE SAME TIME having the technoligical and social development of the player’s culture following historical lines because those things were largely controlled by events outside of the setting of the game.

    This comment is nothing at all like what I wanted to say in the first place but it works.

  • Michael A.

    By its definition, a simulation is a modelling of the essential features of a situation, and then extrapolating from these. Simplification is an essential part of any simulation.

    I’ve never met anyone interested in “Historical Accuracy” as described by Anonymous above – what people seem most often interested in when they ask for historical accuracy is a historical simulation detailed enough to go in historical directions when historical choices are made.

    This is easily achievable in lots of games (not just colonial games): most wargames, for instance. It is mostly in the grand-strategic field where the simulations tend to fail miserably.

    This is not so surprising – an accurate simulation is very hard to achieve when the simulated “system” starts encompassing the behavior of many dozen nation states over many decades. But this is unfortunately not the only cause of poor simulations.

    More devastating, IMO, is that the games industry seems unable to think beyond the – by now -traditional mechanisms when it comes to historical simulation.

    An example, as mentioned in Troy’s post, is the tech tree. The direction of development has only rarely been possible to direct in such a manner – and even less often with success. Equally often, it is a redundant (i.e., boring) game mechanism, as it only provides interesting choices on the first play of the game (due to most tech trees being poorly designed, there is rarely more than one sensible path through them). Yet this is a standard mechanism in 95% of all historical games!

    Among the other mechanisms which are standard, but which rarely model anything is the “city development” (building hiararchies) and “unit progression” (rapid military development has only really been a factor of the past century).

    The cover-all excuse for this sorry state of affairs is quite similar to the one given above: A historical simulation can’t be a fun game.


    Fun is all about having “interesting” choices. Most historical situations provide plenty of those.

    The problem is that you can’t create a historical simulation, using the same old flawed concepts that game developers for some reason feel their games can’t live without.

  • Troy Goodfellow

    Historical accuracy is not at odds with having a game. The question of whether history gets in the way of a good game is up to the designer.

    For example, a Medieval city building sim in which your citizens were constantly dying of the plague would not be a lot of fun. Disease is not fun to simulate, in can get frustrating to lose a game because of something you have little control over, and once you give the player options to control the plague (sanitation, scientific method, etc.) you are putting him/her in an ahistorical situation.

    And many good games are ahistorical – I didn’t suggest anywhere in the essay that historical accuracy should be the sole currency of gaming. It shouldn’t for obvious reasons.

    At the same time, many gamers expect a certain level of historic accuracy. They will accept priests making conversions in a glorious cartoon like Age of Empires; try it Rome: Total War and see what happens. Context is everything.

    The essay above merely examines approaches to games and accepts that designers will make different choices.

    Increasingly, game designers are being more explicit about the choices they are making regarding history. Read any interview with Ensemble regarding AoE3 and it is clear what choices they are making and why.

  • Michael A.

    How history is used depends a lot on the designer, as you state.

    In a medieval city building sim (to stay with your example), the plague could be simply used as a random event (much like earthquakes, etc) in Sim City that the player would need to respond to – or exploit (plague often led to social mobility with resulting societal changes). It all depends on what kind of game – and what kind of simulation one is aiming for (simulations – by their very nature – abstract). And of course context, as you say – i.e., whether simulation is in fact being attempted at all.

    My own pet peeve is when one sees a game that claims to be historical, but where it is obvious that the designers haven’t gone to any effort to get it right. And especially when the excuse is that old Sid Meier classic: History isn’t fun!.

  • Ben Fritz Thinks Colonization is Bad

    […] written about this sort of thing before. A few times, in fact. The tough issues in America’s history are something that game designers have tended to […]