What follows is a post in our JAAC x AFB collaborative series, where we highlight articles from the Journal of Aesthetics and Art Criticism. This post features Marissa D. Willis’ recent paper, “Choose Your Own Adventure: Examining the Fictional Content of Video Games as Interactive Fictions“.
“Video games don’t tell stories,” he told me. “They’re just games.”
So said a friend of mine when I told him I was writing about video games as works of fiction. And despite his mansplaining my own topic to me, my friend was giving voice to the very problem which I hope to address. Despite the fact that more people are playing video games these days than ever before, and game makers continue to create more inventive and engaging narrative works every day, my friend is not alone in his opinion.
Although there have been some recent exceptions, such as the works of Jon Robson, Aaron Meskin, and Grant Tavinor, video games are often overlooked in the philosophy of art. Many people don’t consider them to be a form of art at all, rather “just games”. And when games do turn up in aesthetics textbooks, they tend to be lumped together with film and television and afforded no special interest. This is not only a shame but a loss for philosophy. Not only does a great deal of artistic effort go into making a video game, but the result is more than the sum of its parts. Video games are unique. They present several new and interesting philosophical challenges that may, if we let them, reshape our understanding of art as a whole. My work aims to shed some light on these challenges—specifically that of establishing fictional truth in video games.
Finding Canon in Video Games
“Fictional truth” may sound oxymoronic, but I bet you know quite a few. Sherlock Holmes lives at 221b Baker Street. Bruce Wayne is Batman. Hamlet says, “To be or not to be?” All of these are fictional truths. They are not actually true—no detective named Sherlock Holmes ever actually lived on Baker Street—but they are true within their respective fictions.
Interest in fictional truth has become more commonplace in recent times under the name of canon. Something is considered part of the canon of a certain fiction if it is an established aspect of that fiction. This term can be useful when discussing more complex fictions, such as the Marvel Cinematic Universe, or the even more complex world of the comic books on which these films are based. Most of the time we can determine what is canon by checking the text, as long as it remains unaltered. Even if every time I watch Black Panther I notice something new, the film itself stays the same, and so does the canon.
But what do we do when the text itself is continually changing? This is the problem that video games pose. Take for example my favorite game, Mass Effect 3. The game tells the story of Commander Shepard, a space marine who fights to save the galaxy from the Reapers – massive aliens bent on destroying all organic life. But every time I play Mass Effect 3, that story can change. Sometimes Commander Shepard is a woman, and sometimes a man. Sometimes Shepard saves everyone, but other times causes the death of an entire race. The story changes because Mass Effect 3 is interactive. It takes the player’s choices into account. All video games do this to one degree or another. Some, like Tomb Raider, are very linear, while others like Skyrim vary wildly every time they are played. So how are we to determine what counts as canon in a video game?
This question is more important than it might seem. Having a canon enables us to treat video games as a unified whole, like we might the Marvel Cinematic Universe. Without that unification, all we have is a collection of different stories. When I play Mass Effect 3 it tells the story of a woman in love with an alien, but when my sister plays Mass Effect 3 it tells the story of a man who is not in love with anyone. How can we say that we are interacting not only with the same computer program but with the same work of fiction? To answer this, we will need a more nuanced view of fictional truth in video games.
Video games present two important challenges when it comes to establishing canon: variable content and variable scope. The first of these I have already explained. While some things are constant, like Commander Shepard’s rank, other things are variable, like Shepard’s gender. Fortunately, video games are not the only type of art to face this challenge. Performance art, like stage plays, also deal with variable content. For example, every time that Hamlet is performed, certain things stay the same: Hamlet always dies. But other things vary between performances, like Hamlet’s hair color or height. While all of these things are part of the performance, some of them are also true in Hamlet itself. They are included in the play’s script, and in every performance of Hamlet. So while we might contrast different performances, we can still talk about Hamlet as a whole.
We can use a similar approach to games by differentiating between the game itself and an individual playthrough of the game. These playthroughs are analogous to performances. Video games are mediated to us via playthroughs, which contain both constant fictional truths (Shepard is human) and variables (Shepard is female). The constants are part of canon, part of the game itself, while the variables are only true in that specific playthrough.
The only problem now is determining which fictional truths we encounter while playing the game belong in which category. Plays have an easy go of this, as they tend to have scripts which we can read apart from a performance. But games do not. We can only encounter a game via a playthrough. One way we might use playthroughs to generate a canon would be to play through every possible variation of a game and compare them. Even if not feasible in practice, this is at least a theoretical possibility. We could then define canon as anything which is true in every variation, since it would be constant (Shepard is human), and exclude from canon anything which is true in some but false in others, since it would be variable (Shepard is female). However, there is a tricky snag in this plan, one that brings us to the second complexity of video games: their variable scope.
By saying that video games have variable scope, I mean that most games have a good deal of optional content which the player may investigate or may skip at will. For instance, in Dragon Age: Origins there is a character, Alistair, who has a sister. The player only learns this if they speak to Alistair about it, and could go the entire game without learning it. Because not every playthrough of the game features Alistair’s sister, we might be tempted to say that “Alistair has a sister” isn’t part of the Dragon Age: Origins canon. But this move is not as simple as it may first appear.
If how we determine what is canon and what is not depends on comparing all possible playthroughs, then what “counts” as a playthrough is important. What if I played a game for less than one minute and then turned the game off and never played again? What if I encountered <1% of the potential fictional content? If we allowed such a playthrough to “count”, that would drastically narrow what we can consider to be canon down to almost nothing. To avoid this, I suggest that we instead define canon in video games like this: Something is canon in a game if it is true in at least one possible playthrough, and never contradicted in any other playthrough of the game. This means it would be fictionally true in a playthrough even if it is never presented to the player, because it is part of the canon of the game. Alistair always has a sister whether he tells us or not.
Video Games and the Theory of Make-Believe
This is where the rubber meets the road when it comes to the theory of fiction. So far, I have done a lot of problem solving trying to find a way to determine what is canon in a video game: how we address a video game as one unified work of fiction, even if it tells a different story every time. But I haven’t actually addressed what it means for something to be “fictionally true” in the first place, or what is happening when we interact with fiction.
One of the most influential theories of fiction is Kendall Walton’s theory of make-believe. Walton says that the way a work of fiction generates fictional truth is by inviting us to play a game of make-believe in which the words on a page or pictures on a screen are true. So we make-believe that there is a detective living on Baker Street, or that in a hole in the ground there lived a hobbit. We make believe that once upon a time, etc. We don’t actually believe it, of course. We aren’t fooled or lied to. We know it’s a fiction. But we play a game in which we let the work tell us what to make-believe. To Walton, that’s just what a fiction is: Something we interact with during a game of make-believe that tells us what to imagine.
By rights, Walton’s theory should apply to video games. Not only because video games are works of fiction, but also because his entire theory is based around comparing fiction to playing a game. If any theory applied to games, we would expect it to be his, and if his theory applies to anything, we should expect it to be games. But it doesn’t.
There are two major conflicts between Walton’s theory and video games as I have described them. The first is that according to Walton, if a work does not mandate that we make-believe something, then it’s not part of the fiction. This means that, contrary to what I maintain, “Alistair has a sister” cannot be canon in Dragon Age: Origins, as not every playthrough mandates that we imagine it.
However, video games are not entirely alone in this. Lee Walters has pointed out a similar issue in serialized fiction by asking, “Is Darth Vader Luke Skywalker’s father in Star Wars Episode IV: A New Hope?” Vader is clearly Luke’s father in Star Wars Episode V: The Empire Strikes Back, and we know this means that during Episode IV he was Luke’s father all along. But if someone had only ever seen Episode IV and never Episode V, they would have no way of knowing that Vader is Luke’s father, and would never be asked to make-believe that he is. And thus, according to Walton, the answer would have to be that Darth Vader is not Luke Skywalker’s father in Episode IV, as ridiculous as that sounds.
To fix this problem, Walters offers what he calls a “Waltonian” solution in which he appeals to the existence of a larger overarching fiction – the Star Wars series as a whole – which mandates that we make-believe that Vader is Luke’s Father, even if Episode IV doesn’t. This is exactly what I have been referring to when we say we can determine what is “canon” in the Marvel Cinematic Universe or in Mass Effect 3 itself. However, while this is a good solution, it isn’t truly Waltonian.
If we maintain that when someone plays a video game what they interact with is a specific playthrough of that game and not the game itself (just as we watch a performance of Hamlet and not the play itself), then according to Walton’s theory the game (and play) is not, itself, a fiction. Because the game does not mandate that we imagine something, the playthrough does. The same can be said of Walters’ example. We do not watch Star Wars in the abstract, we watch Episode IV. According to Walton, Episode IV is a work of fiction, but Star Wars is not. And if it’s not a fiction, then we cannot appeal to it as Walters suggests and still call it Waltonian.
It seems that no matter where we turn, there will be at least one aspect of fictional truth in video games which Walton’s theory cannot explain. And this is concerning. As I said, if Walton’s theory is to apply to any art, we should expect it to apply most easily to games. But it does not. And if Walton’s theory of make-believe cannot explain games, we must ask whether it can truly explain anything else.
Walton’s theory is of course not the only philosophical approach to fiction. But the fact that his theory, which is built around children’s games of make-believe, cannot account for video games is telling, not only about Walton’s theory but about the philosophy of art in general. If we want to discuss the philosophy of art we must not be bound to the traditional, but also must not turn a blind eye to the popular or rule out art targeted at a different demographic than our own. And as the philosophy of art expands we must work hard to examine new forms of art and shape our philosophy around them, rather expect them to fit into old molds. And above all, we must be committed to exploring fiction in all of its mediums, and never dismiss anything as “just” a game.
Notes on the Contributor
Marissa Willis is a graduate of Oxford University and John Brown University currently working as a substitute high school teacher while she tries to apply for PhD funding. Her research interests center around fiction: it’s structure, ethics, and epistemological importance, and especially on interactive fictions and on mythology.
August 2, 2019 at 5:57 am
I think you think too much buddy putting way too much thought into this.
August 8, 2019 at 12:09 am
Shepard is male just look at both covers for ME 1 and 2. Also can’t help but feel a little concerned for the author of this post.
February 5, 2023 at 12:18 pm
No he is not ever creators of the game have said femshep is ment to be the canon shep
February 4, 2021 at 10:33 pm
No matter which way you played its a emotional rollercoaster if you get attatched. Played trilogy first time in 2011 not even researching the game and by the end of me3 i cried big giant tears of sadness and bitter happiness.
Pingback: How To Romance Kaidan Without The Whiskey – Whiskey Dry Restaurant
December 22, 2022 at 7:15 am
Interesting article. Something you may not be aware of is the ability of even the gameplay itself to be a method of storytelling. There need not be any specific fiction for it to be canon that mario can run fast and jump high. The reactions of AI to the player character within a certain game environment can reveal a huge amount about the canon. Just my 2p