Gaming the Past

For this first post in a series, Adam Chapman and I begin to discuss, and hopefully unravel, the ideas of historical accuracy and authenticity in historical video games. What do we mean by these terms? Can games show accuracy and authenticity? Does it matter and, if so, why? We have authored this as a dialogue, each of us contributing a little text at a time and responding off each other.  We welcome participation and will respond to comments.

Jeremiah: It seems a straightforward sort of question: “how historically accurate is that video game?”,  whether it’s Assassin’s Creed: Origins, Call of Duty: World War II, Sid Meier’s Civilization or any of the myriad historical video games. Sometimes when we talk about historical video games, we use the term historically authentic to try to capture something different about the ways a historical game relates to the past it depicts. Either way, it’s not an easy question. But let’s see if we can unpack it.

What does it mean to be historically accurate in general? Does that mean that a medium (text, recording, image, video, game, etc.) represents or depicts events in the correct chronology and “as they happened”? If so, we’ve got a problem right there. It’s been quite awhile since mainstream historians have argued that historians can in any meaningful sense depict the past “as it was.” But let’s leave that aside for a moment. Let’s stipulate that historically accurate means presenting accurately in the medium the “historical facts”, the “generally accepted” view of events, the participants, the order they happened, causes and effects, that sort of thing,

If this is what we mean by historical accuracy, it is essentially impossible for any video game to be historically accurate. Here’s why. In a game like Assassin’s Creed, the player plays a completely fictional character in a completely fictional role; in a game like Call of Duty: World War II, the role is authentic–U.S. Soldier–but the character is fictional. In a game like Crusader Kings 2 the player can start as a historically documented figure in a historically documented role. At some point as they make in-game choices and the game AI players react, the in-game events will stray from the historical facts: the game, in this sense, will not be historically accurate. In none of these cases, is the game historically accurate in the sense noted above.

Some critics of historical video games, both in and outside of education, do make these sorts of claims. Playing a historical wargame that does not produce the historical outcome is simply unhistorical, and just not useful for understanding that past. Exploring a historical setting as a fictional character will not help one understand what happened and why.

I’m confident we should do more here; we can come to a more meaningful understanding of what historical accuracy and authenticity mean for historical video games.

Adam: I would agree that historical accuracy and authenticity are seemingly simple theoretical concepts that rapidly become complex when applied to real examples, particularly videogames. These concepts are also, at least from my perspective, different though related things.  But let’s continue to look at accuracy for the moment and deal with authenticity a little later.  I would agree that a general definition of accuracy is an alignment with the generally agreed upon facts of the past.  However, like all historical representations, videogames are made with a number of external pressures. Some of these apply to all kinds of histories (even academic books). For example, all representations have to leave something about the thing they represent out of their depiction. If they didn’t then they would cease to be representations at all and instead become the actual thing they try to represent!  To expect complete accuracy of videogames is to hold them to a standard that we don’t even expect of academic writing about the past and thus seems quite unfair.

Jeremiah: A fair point. The idea of the academic historian generating a complete and accurate account is misleading; it ignores the bias of any writer, the incompleteness of any picture or text, the lack of anything approaching perfect knowledge. We do not expect an academic historian to know and discuss everything about the past.

Adam: Absolutely! It is fair to say though that some other pressures are more specific to games. Of course as an entertainment media, games must appeal to players and also meet sales targets. This can naturally influence the type and nature of the history that they contain, just as it does with novels, films and documentaries.  These aren’t inherent limitations of these media forms though.  That is to say it doesn’t specify some kind of inherent inability of games or films as historical representations in comparison to academic books. This is just a result of the surrounding pressures of production that these media tend to be subjected to.  There are also pressures introduced by the conventions of the media form (be it film, game or book).  These are many and complex but an obvious example is found in length. For instance, a Hollywood film will probably last around 2-3hrs on average and a typical FPS campaign will last maybe 6-10hrs on average.  This means that often historical events or the experiences of historical characters will be compressed into single characters or events.  This is generally not an attempt to deliberately mislead audiences but to tell them more about the past within the necessary confines of the conventions of the chosen means of communication.  This is one of the reasons why we find videogames (and other popular historical media) tend to use characters that are fictional, as you note.  Perhaps a more useful question to ask of such characters, rather than “are they accurate?”, is to ask “are they historically typical?”, i.e. are these the kind of things that people in the past genuinely probably experienced?  A character in a game may be depicted as experiencing many things that it is unlikely a single person in the past experienced. But if the game shows players real experiences from the past, even if these experiences in the past actually happened to many different people, then these games can still have a lot of value in terms of learning.  

Jeremiah: That’s another helpful distinction. Different media have different characteristics, and it is unreasonable to suppose that a historical text, game, and film would all refer to the past in the same way, else there would be no point to having different media at all. On top of it all, these histories are developed and presented in a context with constraints and opportunities that shape their creation.

Adam: Another one of the other reasons we often find fictional characters in historical games brings us to an even more specific issue of videogames: agency.  How can a game detail the exact experience of a single person from the past if they also allow players to make decisions for that character that may not have been made in reality? Indeed, how can games hope to show any sequence of events and simultaneously give players agency to influence the course of events in the game?  Of course one of the ways out of this bind is to give players influence over some (generally smaller) aspects of the narrative (e.g. movement, tactics etc.) and enshrine some other narrative elements in parts of the game that aren’t dependent on player action (e.g. using cutscenes or having historically accurate objectives that the player must achieve to progress the story).  However, at some point the player has to have some kind of choice. They must always be able to influence the narrative in some way, however small, or the game, arguably, ceases to be a game at all.  Thus the potential for the game to stray from the historical record is always present.  

As you note this is often a critique levelled at history in videogames.  From this perspective, games can never be accurate.  This is particularly the case in games that give players a lot of agency over the game narrative (e.g. games such as Crusader Kings and Civilization). But this can be a very limiting perspective.  There are different kinds of accuracy, or at least different aspects of the past that we are interested in trying to understand and discuss. Assassin’s Creed, for instance, has a pseudo-historical Dan Brown-esque storyline that might be very confusing to those members of an audience unable to unpick the fact from the fantastical.  This might lead us to dismiss such a game (though I wouldn’t endorse dismissing even the most fantastical of histories as unimportant). Still, the environments of the game have an enormous amount of visual data and detail about past architecture, geography, clothing, tools etc.  

Jeremiah: I think you’re absolutely right to call it player agency. I think of it too in terms of counterfactual history, which is just another way of approaching what you are saying. By definition, games are interactive and in the sense of historical games that means players get some choices about what their player character can do. Remove that agency, and the experience is a video, or text, or recording, or graphic novel, but not really a game. By definition then, a historical video game has to engage in some form of counterfactual history; its gameplay creates a narrative that counters what our best estimations and evidence suggest happened in the past. There are two basic kinds of counterfactual narrative created in a history game and, ultimately, they point to an important division between the  Crusader Kings 2/ Civilizations type games and the  Assassins Creed/Call of Duty type. As far as I’m concerned, you really captured that when you posited  a difference between conceptual simulations (the CK2/Civ varieties) and realist simulations (the AC/CoD varieties).

I think another way one can helpfully make that distinction is between

1. Games whose counterfactual history takes the form of a fictitious historical agent (the assassin or the infantry soldier) who makes lots of personally important decisions that do not change the overarching narrative of the game. This is what I sometimes call “the historical fiction” approach because historical fiction often adopts the same approach.
2. Games that give the player a historical setting, point in time, and, to some extent, “character” (if “Babylonia” or the “Aztecs” in Civ can be called a character) but then allow their games systems to operate, producing a counterfactual narrative as soon as the player or the AI acts in a way that counters accepted historical narratives.

Ultimately, this leaves us in the same spot that you noted. Games will simply not be historically accurate in the straightforward narrative sense. But again, as you said, the idea that academic historians generate a single factual, agreed-upon narrative for anything is problematic.

Adam: Similarly, I think one of the common misunderstandings of the work historians do is that historians spend their time only describing what happened in the past. But a lot of the work that historians actually do is asking why things happened in the past.  Strategy games can have value in precisely this manner.  Games like Civilization, Making History, or Crusader Kings may not be very good at offering retellings of what happened in the past, but they are very good at offering explanations for why things happened.  If we build a colosseum in Civilization for instance, the game offers an argument about what kinds of resources and technological developments allowed the building of such a thing and what kind of political and societal benefits this might have offered.  In the wider game, we find out some of the processes and resources involved in the establishment and maintenance of empires.  This is a simplistic example, but these arguments can also often be very complex (for instance in games such as Crusader Kings or Making History) and perhaps most importantly, these arguments relate to particular perspectives on the past. The kind of real issues that historians talk about.  Such games may often deal in counterfactual history, straying from the events of the past because of player choices. But such histories are inherently concerned with causality, i.e. why things happened in the past.  And the player’s choices in the games are often constrained by rules based on real historical factors and real theories about the past.  Thus such games can refer to actual events, without actually recounting these events themselves.

So, are these games accurate? Well, somewhat frustratingly, the answer is both yes and no. They are often not accurate in their depiction of the events (or at least order, frequency, or significance of the events) of the past. But they are often accurate in their forwarding of the kind of theories about why things happened in the past that historians discuss. We cannot say if these theories are accurate in the same way we can say that a representation of a sword in a game or film looks like the historical artefact it intends to depict.  These are theories after all. But we can say that these kind of arguments are an important part of thinking about the past that games are readily engaging. So whilst these games might not be accurate depictions of the past, they may well be accurate depictions of serious theories and discourses about that past. As such, perhaps asking about accuracy has limitations. This might be useful when dealing with material culture (and with games that have complex 3D visual representations). But both the past and history (as the study of the past) are made up of much more than this.  As such, I think a more important question than accuracy when looking at games and other popular media is: do they say something meaningful about the past?

Jeremiah: The idea that a historical game, particularly one that offers a historical world whose major events can be influenced and changed by the player’s choices, essentially offers an argument about how and why events happened in the past is crucial to finding historical meaning from a game, especially, I submit, from the educator’s perspective, hoping to engage a game in order to enrich the study of history. I wrote an essay quite a while ago called, “The Unexamined Game is Not Worth Playing?” essentially arguing that the critical issue when trying to use a historical game to learn about the past is, can and does one have a meaningful discussion about the game’s portayal of the past, a discussion that attempts to separate the authentic from the spurious, considers evidence, and, above all, engages in discourse about how the game portrays the past. In short, the act of examining a historical game,  thoughtfully discussing the historical accuracy of a game makes the game a legitimate means of  historical thinking regardless of whether the game’s historical models are reasonable or silly. I wrote this in the context of formal history education at the time, but I think that history education, or more properly, learning history in a formal context, is simply part of a spectrum that includes learning and thinking about history from a novel, a political reference, a movie, a book, or a game. So I would say again that the thoughtful doing of history occurs anytime a person considers or discusses, based on reasonable evidence, the ways in which a game can explain and illuminate, or fail to illuminate, cause and effect in the past. And, in fact, you can see this all the time in civil conversations on the forums about this or that game’s historical accuracy: people engaging in the thoughtful doing of history, the meaningful interpretation and constructions of cause, effect, and narratives about those.
Teasing this conversation out further, though I fundamentally agree with you that the real question is whether a game says something meaningful about the past, this does lead to a sense of relativism. I stand fully accused of contributing to that relativism when I say that the important thing about the historical value of a game is not the game’s portrayal alone, but critical discussion about the game’s portrayal. Is there a problem here in this relativist idea: as long as we discuss it (i.e. a game’s portrayal of the past), there is historical value? Taking that to the extreme, a rubber ball can be a useful historical model so long as discussing how it does and does not authentically model the past occurs. Should we expect more from historical games? Can we? I’m reminded that when I wrote Gaming the Past, I was particularly attached to the term “historical simulation game,” as a way to separate how an Assassin’s Creed relates to the past, an Age of Empires, and a Civilization. I suggested that the core criterion for determining whether a game was a valid historical simulation game was this: “its core gameplay must offer defensible explanations of historical causes and systems”. Since then, I’ve read and thought and talked about that and I now think that it’s more helpful to talk about historical games in general and then subdivide into conceptual and realist simulation approaches. But I still wonder if there is something to what I tried to capture in that criterion, that beyond a certain point, a game really isn’t a simulation game any more. I’m also reminded of a very uncharitable internet forum poster: a person who did not know me saw an article I had written about the educational potential of Rome: Total War, and suggested I must not know anything about Roman military history (ironic) because I suggested the game had defensible models about ancient warfare. In their mind, there was nothing defensible about the game. Does your separation into two types of simulation approaches help us determine when a game is an interesting and at least somewhat defensible model of the past or just a rubber ball?

To be continued …  (read part 2 here)