Tag Archive | player agency

Bioshock Infinite and player agency


[Note: This is an article primarily about the ending of Bioshock Infinite, and as such it contains numerous spoilers, as well as spoilers for the original Bioshock. Probably you shouldn’t read it if you haven’t finished these games. There are also some references to other games–the new Tomb Raider, Mass Effect 3, and Red Dead Redemption–but they’re less explicit. Nevertheless, fair warning.]

There have been a number of essays written recently about the strengths and weaknesses of Bioshock Infinite. It is a hugely significant game for a number of reasons, and it deserves most of the praise (and most of the criticism!) that it’s receiving from the gaming press. I can certainly go on record to say that I enjoyed my time with the game thoroughly, and despite a few criticisms, I’d easily call it one of the best games I’ve played in the last couple years. But the final half-hour of the game, after the final battle has been fought, in which all of the game’s secrets are laid bare before the player, left something of a sour taste in my mouth.

Bioshock Infinite has one of the most frustrating endings of any game that I can recall. It gains much of its emotional impact by being deliberately dissatisfying to the player, and it thwarts a player’s sense of agency in a way that few other games dare to–for better or worse.

Inevitability is at the heart of Bioshock Infinite’s narrative. It’s a theme that is brought up frequently throughout the game, most often by Robert and Rosalind Lutece, who show up periodically to remind Booker of how little choice he has in the events of the story–and, simultaneously, to remind the player of how little choice THEY have in influencing the narrative. None of the minor choices present throughout the game have any significant effect on the events of the plot, and indeed, some of them are even decided for the player.


An early scene in which the Luteces ask Booker to call a coin toss always comes up heads, but Booker can actually call it either heads or tails. Importantly, however, it is Booker, and not the player, who makes this decision. This scene accomplishes a couple different things: first, it reinforces the overarching theme of the game, that the events of the world will play out in a certain way no matter what the player does; second, it creates dissonance between the player and their avatar, Booker–”why,” the player must ask themselves, “does the game not offer me input on this binary choice, when I have already been asked to decide between two courses of action earlier in the game?” (The scene is also, as Kevin Wong points out at Gamasutra, a nice little hat tip to Tom Stoppard.)

Booker tells Elizabeth, on a number of different occasions throughout the game, that there’s nothing he can do to wipe away his past. Ultimately, of course, that turns out not to be true–not only is Booker able to atone for his sins, but he is able to literally prevent them from ever having occurred–and all it takes is allowing himself to be murdered.

This is an exceptionally powerful ending to the story, but at the same time it’s very frustrating to the player–neither Booker nor the player have a complete picture of the situation until the final moments of the game, and by that point it’s too late. Once all of the pieces are in place, there’s nothing left to do but watch yourself be killed and watch the credits roll. The inevitability of the narrative hits you hard, and you have to sit by yourself on the couch for a while and sort things out in your head.


The question that was left ringing in my head, for days afterwards, was: “If the whole point of the exercise was that there was nothing I could do, why did I play the game?”

Bioshock Infinite’s comment on player agency in games seems to be that, at least narratively, it’s an illusion. None of the choices that you make have any real consequences. There is nothing you can do to affect the outcome of the story. Just like Booker, you are brought into the world of the game in order to play your part in events over which you have no control. Even death can’t stop you from playing your part: die outside the company of Elizabeth (who would otherwise resurrect you), and you reappear inside Booker’s apartment, only one door away from picking up where you left off. (And is it the same Booker? Does it matter?)

Like everything else in the game’s story, Booker’s ultimate demise (or is it a willing sacrifice?) is thrust upon the player without ever allowing them to take ownership of it. It’s not clear that it’s a choice Booker is making, and it’s not at all something that the player can make the decision to undertake–we don’t even really know it’s coming until a moment or two before it happens.

In no other medium could you so powerfully confront the player with their own narrative impotence, and that makes Bioshock Infinite very potent! In some ways, however, it makes me wonder whether the plot of the game wouldn’t be more satisfying in a different medium. Someone’s edited every story sequence in the game into a 3.5 hour film. Would the same messages about inevitability be just as compelling if the story weren’t interactive? Is the impact of the game’s ending dependent on the player’s belief that they might have some agency, or is it weakened by the player’s expectation to take ownership of the actions of their avatar?

“Nobody tells me where to go,” Booker says, close to the end of the game, and it’s an ironic statement on a number of levels. On the one hand, we as the player are telling him where to go. And on the other hand, there’s really no choice in the matter. As Elizabeth points out, the game brings us to the same place no matter how we try to fight it. And in the end, what happens to us? We are, as the many Elizabeths so appropriately echo, “smothered.” Frustrating! There’s nothing we can do. There was never anything we could do.

In the original Bioshock, one of the greatest twists in gaming plays heavily off the fact that you can’t progress in the game without following its prescribed, linear set of objectives. “A man chooses. A slave obeys.” And you have, of course, been unwittingly obeying the entire time. One of the great elements of thematic dissonance in that game, however, is the fact that the second act involves ridding your avatar from another character’s mind control, and yet the linear progression of objectives remains–you’re just taking orders from a different character, and though you’re no longer under the influence of “mind control,” you cannot deviate from the narrative’s path without bringing the wheels of the game to a screeching halt. You are, essentially, still a slave, and not a man. Players ask themselves: what have I really freed myself from, in the end?

I was always fundamentally disappointed with this element of Bioshock, and I think that I’m not alone in that assessment–the game’s ending is generally considered to be less compelling than its mid-game climax. I’m re-evaluating that assessment in the wake of Bioshock Infinite, however. My dissatisfaction with Bioshock’s second act was only retrospective–in examining the themes of the game once I’d finished it, I felt like there was a bit of hypocrisy there. While I had the controller in my hand, I didn’t feel any of that dissonance, because the narrative was aligned with my interests as a player: Tenenbaum wanted me to take down Fontaine, and she gave me instructions as to how I might do so. That was fine. After the betrayal I’d just experienced, I wanted to take down Fontaine, and I was willing to follow her “suggestions” as to how I might proceed, “would you kindly” or no.

Instead of asserting the player’s narrative impotence halfway through the narrative and then falsely “freeing” them from that impotence, Bioshock Infinite uses the last thirty minutes of the game to show us just how much of a pawn we really are, and then drive it home by murdering us and letting us watch the credits with slack jaws and a vague sense of betrayal.

As gamers, we’re willing to suspend our disbelief about how much agency we have in a linear narrative in order to see that narrative through to the end–as long as the game gives us a compelling reason to do so. Playing through the Tomb Raider reboot, I didn’t care that I was helping to turn Lara into a mass-murderer, or that so much of the narrative (and indeed, the gameplay!) was “on rails.” The premise of the game is that Lara’s in a terrible situation and that she and her friends will all die if she doesn’t resort to some extreme measures. (Kirk Hamilton calling out the game’s homages to The Descent over at Kotaku underscores this pretty well.) Tomb Raider isn’t a game that’s particularly concerned with issues of player agency, and so players don’t feel too compelled to consider moral culpability as they play it. I think that’s just fine.


There’s a reason, however, that there have been some good arguments raised about whether or not Bioshock Infinite’s ultraviolence undermines its narrative (and, of course, its accessibility to non-gamers or casual gamers). When a game strives to remind us, throughout its story, both that our character cannot wash the blood off his hands and that we may or may not have any choice in the actions that we’re taking, we start to look a little more critically at the slaughter that we’re perpetrating. This isn’t Yamatai–we’re not forced to choose between helping Lara slaughter thousands of men or watching her die. This isn’t even Rapture, where we’re forced to put down scores of Splicers that want to harvest us for our ADAM.

In Columbia, we kill hundreds of men because… Booker’s a bad guy already, and that makes it okay? Because we’ve been brought here by fate (or the Luteces) and because we have no choice? Because if we let them kill us, we’ll wake up back in our apartment and step out the door and… keep on killing? That’s a sort of WarGames scenario, right there. The only way to win is not to play. In Tomb Raider or the original Bioshock, there’s a light at the end of the tunnel–kill all the bad men, and eventually the killing will stop, and you’ll be safe. In Bioshock Infinite, if you kill enough bad men, then… none of it will ever have happened? Well, that’s cool. I’ll tell you how else that could happen, though: don’t play the game in the first place.

Obviously this isn’t a course of action I suggest anybody take. The game is awesome. If you don’t play it, you’ll be missing out on all sorts of wonders. I only mean to highlight the ways in which the game’s ending makes you reflect on your experience throughout the narrative and ask yourself, “why?” This isn’t Mass Effect 3 or Red Dead Redemption, games which also involve inevitable final sacrifices but allow the player to internalize, come to terms with, and follow through with those sacrifices as though they were their own decisions. Nor is it Shadow of the Colossus, which telegraphs its protagonist’s fate within the first hour of the game and dares the player to continue anyway.

Bioshock Infinite, by keeping the player’s blinders on until the final sequence, never allows players to understand the full import of their situation until they’re waist deep in the river and the only thing left to do is drown.

It’s a bold, powerful choice. Is it a good one? I’m not sure I’ve figured that out yet. Maybe the fact that I’m still sorting through it in my head means it was a good decision. Not all art needs to be cathartic, and sometimes brilliant pieces are deliberately dissatisfying. Nevertheless, I’m not sure I’ll be returning to Columbia anytime soon. The game aims to frustrate, and it succeeds! I will go and play some other things, and perhaps when my frustration has subsided, I will return. I’ll let you know how long that takes.

The Player as Player: Games as Theater

I’ve read very often, in descriptions of a game’s premise, a phrase which I’ve begun to consider: In a given game, “players take the role of…” A Google search of this phrase naturally turns up almost exclusively hits relating to video games, but it wasn’t until recently that I sat down and really considered the terminology here.

The only other possible medium besides gaming to which this might apply is theater, and yet I don’t see a lot of criticism in games journalism drawing parallels between game players and stage players. It strikes me that picking up a controller and picking up a script are, in many respects, quite similar.

Most folks who’ve done some theater are familiar with “the actor’s nightmare”: You are in the middle of a performance for which you have forgotten all of your lines and most of your cues, and you are constantly being urged by your castmates to continue despite the fact that you have no script, you have no costume, and indeed you have no pants.

I think that playing a video game, or at least any game with a narrative, is a bit like the actor’s nightmare with all of the bite removed: you have assumed a role, you set foot onto a grand stage knowing none of the lines, and yet miraculously, every time you reach the next bit of the story, the words come tumbling from your mouth unbidden. When we say that the player “takes the role of” John Marston, or Commander Shepard, or even Sly Cooper, the implication is that we’re not just using these characters as agents by which to accomplish gameplay objectives–we are actively taking on a role, as an actor takes a role in bringing a play to life.

I wish I’d had a health bar when I did theater in high school.

The Assassin’s Creed franchise has probably the most direct implementation of this of any game that I can think of. Your health bar effectively measures how in character you are when you’re playing. Because of the narrative conceit that you are Desmond Miles taking on the personae of his ancestors, the game will cause you to fail if you act “out of sync” with these historical assassins. Kill a civilian? Nope, Altair didn’t do that! Don’t do it again or you’ll be “desynchronized!” Essentially, you get a game over if you are out of character. This is an excellent excuse for giving you a game over if you’re detected on certain missions (“Oops! Ezio wouldn’t have gotten caught there!”) as well as constructing the limits of the game world (“Altair never swam a foot in his life!”).

Just because the Assassin’s Creed games are some of the only ones to take this concept and construct their world around it doesn’t mean the principle isn’t at work in other games. How many of us felt an uncomfortable dissonance playing Cole Phelps in L.A. Noire? Cole’s kind of a jerk!

Look at him! He knows it.

It’s not easy being a jerk onstage. Cole Phelps, however, isn’t the player–he’s an agent rather than an avatar, meaning that he doesn’t directly represent the player’s persona. We’re not jerks, we’re just playing one in the game. Nevertheless, this sort of thing isn’t as black and white as it can be in other media: the player has some agency over how Cole behaves in conversations, but not a lot (certainly not as much as they do in the conversations of Commander Shepard–and yet even there, there’s some dissonance!). Sometimes, in the midst of an interrogation, I would back off in alarm. “Whoa, Detective! They told me the X button meant ‘doubt,’ not ‘tear the witness a new one!'”

Nevertheless, I played an awful lot of L.A. Noire “in character.” I was careful with my driving unless we were in the middle of a chase scene, I walked when I could have dashed, I didn’t run people down in the street. Many players like to push the boundaries of the world and test the limits of the play space they’re given. There are plenty of others who don’t–players who don’t like to break the game because it breaks the illusion.

I’ve read a lot of arguments considering the matter of cutscenes and their place in gaming, but I think that this one, by Alois Wittwer over at Nightmare Mode, might be among the few I’ve read in defense of them. In discussing player agency, Wittwer uses the excellent example of Final Fantasy IV’s opening to show us how a scene over which we have no control can be used to invest us in a world: we’re taking on the role of Cecil Harvey, who does something pretty reprehensible in which we’re immediately complicit, and so his redemption over the course of the story is in some way shared by the player.

“Boy, it’s a shame you didn’t do something about it, huh?”

Cecil is a member of a military organization which ransacks a tiny, helpless village to obtain a precious resource (one of the ubiquitous Crystals in the Final Fantasy franchise). The player has very little agency here: a long story sequence, including a handful of battles, plays out before we’re given control of our hero. The argument against cutscenes, the argument for greater player agency, suggests that we can’t feel guilt over Cecil’s actions because we had no choice in them. Like Cole Phelps, Cecil is an agentnot an avatar: he’s not representative of the player–he’s just a role we’re taking on.

Cecil’s plight, however, is that he’s not in control of his situation any more than the player is–he’s in service to a king he no longer trusts, committing acts for which he feels remorse. A sympathetic player naturally wants to enable his escape from such a situation (and will feel satisfaction for him when we finally help him to do so).

Anyone who has played a part in a theater production at any level will tell you that there is great joy in putting on a play. Taking on the role of character and working with comrades to bring a story to life for an audience is an enormously satisfying endeavor. What a good narrative in a game allows us to do is to participate in this experience on our own, without the aid of others. We take on one role, just as we would onstage, and trust to the game’s designers to carry the rest.

Much has been said in games criticism about how the player is simultaneously creator and audience, something which is true in no other medium–but I would add that if any other medium comes close, it’s the theater. Onstage, one can participate in a story which unfolds in real time and which has no “post-production.” The only line drawn between the creators of a stage production and the audience is one which is invisible, hypothetical, mutually agreed upon, and respected by all parties involved (or not!).

Playing a game is like having that fourth wall removed, being invited up from the audience to participate, and immediately knowing all of the lines. What a wonder! Is it so surprising, then, that some of us gamers are perfectly willing to abandon our agency if it’s in service of the production as a whole?

I think that those of us who consider games critically ought to be more cognizant of the parallel between games and the stage: it strikes me as a more apt comparison than games and cinema. It might behoove game designers, as well, to consider this comparison when thinking about matters of player agency in their design. If we think about the roles that we’re taking on as gamers, we might be inclined to play differently–and the experience might even be improved!