The Illusion of Free Will: On "Bandersnatch" and Interactive Fiction
After seven years, four seasons, eighteen episodes and two specials, the conversation around Black Mirror seems to have settled itself into distinct camps. There are those who see it as a meaningful commentary on the growing role of technology in modern society and the pitfalls of our growing dependence on it. And there are those who decry it as a cynical, reflexively anti-tech exercise in nastiness. I tend to think of myself as falling between the two extremes—there are a lot of ideas in Black Mirror that I find interesting and unique, especially when it comes to the intersection of technology and capitalism; but I often feel as if many of them have happened largely by accident. The show's latest foray, however, the interactive movie "Bandersnatch", written by series creator Charlie Brooker and directed by David Slade, has shaken my indulgence. Not only does it revel in some of Black Mirror's worst excesses, it's also an extremely bad example of interactive fiction, at a moment when the form is enjoying a creative flowering.
"Bandersnatch" follows a young computer programmer, Stefan (Fionn Whitehead), who in 1984 pitches a game publisher on a game based on the cult novel Bandersnatch, whose author went mad and killed his wife. Bandersnatch, the novel, is a choose-your-own-adventure book, and Stefan's selling point for the game is the heretofore-unprecedented ability to give players a sense of control over the story they experience. So almost at once, the film sets up several layers of meta-fiction—not only the game-based-on-a-book-within-a-movie, but the fact that Stefan is trying to sell the type of narrative he himself is living within. It's also a reminder that, as with 3D projection, the entertainment industry has taken cracks at interactive fiction on more than one occasion, usually for pure business reasons, and rarely to any great effect.
The film wastes little time in confronting the viewer with choices, some as trivial as choosing which breakfast cereal Stefan wants for breakfast, and others as fraught as deciding whether to kill someone. The viewer can take Stefan on story paths that end up involving drug trips, government conspiracies, time travel, and emotional breakdowns.
Unfortunately, it very quickly becomes clear that "Bandersnatch"'s creators have placed far too many eggs in the novelty basket, neglecting to give the individual branches in their multi-branch story a reason to exist in their own right, or anything resembling a common theme. We get glib pronouncements on the creative process, as when Stefan's idol, the bad-boy game programmer Colin (Will Poulter), advises him to work on the game alone, because "teams are fine for things like action titles, but when it's a concept piece... bit of madness is what you need, and that works best when it's one mind". And we get the poisonous equation of mental illness and creativity, as in a story-path in which Stefan agrees to up the dosage on his psychiatric medication, and subsequently produces a game that is dismissed by a reviewer as having been written "on autopilot".
The closest the film comes to genuine wit is when Stefan starts to realize that some of the choices he's making are not of his own free will, and even fights them off. This can lead to darkly humorous scenes in which he asks the viewer's advice, and then acts disgusted at our choices, or a sequence in which we can try to explain to him what Netflix is. But it's a concept that, like many others in Black Mirror, outstays its welcome to very little effect. It's all very well to point out that a character in a work of fiction doesn't have free will (and this is trivially true whether it's the author or the viewer who is making the characters' decisions for them) but it's not what one might call a particularly interesting observation.
As many commentators have noted by now, it is more fruitful to discuss "Bandersnatch" as a game than as a film, and especially in the current moment, in which interactive fiction of its type has been enjoying a surge of popularity. When Netflix first announced its intention to develop interactive entertainments, there was no small amount of skepticism expressed, including by people who remember some of the previous iterations of this concept. An obvious counterpoint to these reactions would have been to point to the success and cultural penetration of recent walking- and conversation-simulators like Gone Home (Fullbright, 2013) or Firewatch (Campo Santo, 2016). Obviously, the audience for this kind of interactive narrative experience exists, so why shouldn't a work like "Bandersnatch" capitalize on it?
The semi-derogatory term walking simulator is an oversimplified way of describing a game whose purpose is to carry the player through a narrative. While there are usually simple tasks the player needs to carry out in order to advance the story—go to this location, find that item—these are often little more than tokens, and some walking simulators, like Gone Home, are merely a structureless exploration of a single location. A conversation simulator also gives the player the ability to choose different dialogue options (or not to speak at all) which affects the game's narrative. In the eerie ghost story Oxenfree (Night School Studio, 2016), the player character's dialogue choices affect her relationships with the other characters, and make subtle changes to the game's ending.
In theory, "Bandersnatch" offers the player more freedom and control than your average walking simulator game. The reality is quite different, however, which can tell us a great deal about why "Bandersnatch" fails, and what it is that makes these games appealing. In Firewatch, for example, you play a lonely park ranger traipsing back and forth across a patch of Wyoming wilderness, with only the voice of a friendly but potentially untrustworthy colleague for company. The game offers the player no meaningful choices, locking them into a single story path and only allowing them to choose dialogue options. And yet it is undeniable that the experience of watching—or rather, playing—"Bandernatch" is much less engaging than an average playthrough of Firewatch. Is it because Firewatch has a better story? Or is it because the feeling of exploration you get from directing the player character is more satisfying than the actual ability to control the narrative in "Bandersnatch", which aside from its decision points gives you no control over its storytelling, down to the camera angles?
Some walking simulators, like Gone Home or Fullbright's follow-up Tacoma (2017), resemble immersive theater far more any game. And as in an experience like Sleep No More, the participant choices are expressed not in how they affect the story, but in how they choose to consume it—where they choose to look, what scene they choose to follow. In Tacoma, you visit a space station that has been evacuated and watch recordings of the crew in the hours following a disaster. You can choose to rewind the recordings, following different characters each time, or focusing on crowd scenes. There's nothing you can do to affect the story, which ended before your character arrived on the station. But your active exploration of it creates a sense of participation that is far more engaging than anything "Bandersnatch" has to offer.
The more you look at games of this type, in fact, the more obvious it becomes that no one actually wants to be able to make up the story they're consuming. Even if you move away from pure walking simulators like Gone Home, choice is something that designers have to carefully corral and shepherd, finally reducing it to an illusion. The mega-hit Life is Strange (Dontnod Entertainment, 2015) based its entire storyline on the issue of choice, even giving its player character the power to rewind time and change her choices after she's made them. A major throughline in the game involves the heroine's choices affecting the state of mind of one of her friends, which ultimately determines whether the player can stop her from committing suicide. But in the end, the only choice in the game that actually matters is the final one, and whichever option you choose, it will negate all of your previous choices, either by winding back time to the beginning of the game's events, or by allowing everyone you've interacted with and affected to die.
Even a game like Night in the Woods (Infinite Fall, 2017), which distances itself from other walking simulators by taking the form of a side-scrolling platformer and including some rudimentary running and jumping challenges, uses choice sparingly, to the extent of not even having a final choice. The entire final encounter with the game's chief menace is essentially a cut-scene without even any diverging dialogue options. The only effect your choices earlier in the game—to explore side-quests and get to know minor characters—have is to give this final confrontation deeper layers of meaning and significance. Oxenfree, meanwhile, leans into the expectation that players will repeat the game multiple times in order to get slightly different configurations of its ending by revealing that its heroine is stuck in a time loop, so that ultimately no choice in any individual playthrough can be said to have a lasting effect.
All of these games are deeply satisfying as both narratives and playing experiences—far more so than "Bandersnatch", which on the surface level gives the player more control over its story. The conclusion is inescapable: the impression of control and agency is much more important, when creating interactive fiction, than the actual ability to determine the story's direction and conclusion. On the contrary, giving the player, viewer, or reader complete control over a story is recipe for dissatisfaction. If you think back to the Choose Your Own Adventure books you read as a kid, were any of them particularly fun as stories? Or was it simply the novelty of skipping back and forth and trying to find the good ending (which is a lot faster and easier in a book than in a movie where you have to sit through the same scenes again and again) that kept you entertained?
The reason that letting the audience choose its own story keeps failing when the entertainment industry tries it is that it's a bad idea. It's the author's job to write the story. They can then choose a way to convey that story that gives the reader freedom in how they experience it. But if the story itself is merely a loose collection of different options, each in a different genre and with a completely different tone, then what they've created isn't a coherent work, but a self-indulgent mess—like "Bandersnatch", in fact.
In fairness, the film itself eventually reaches this very conclusion. In its most satisfying and thematically coherent ending (and the one the film signposts as the "correct" ending, since it's the only one in which Stefan produces a game that gets a five-star rating) it has Stefan explain that "I've been trying to give the player too much choice. So I just went back, stripped loads out. Now they've only got the illusion of free will. But really, I decide the ending". The problem is that the film presents this as a revelation—a slightly sinister one, in fact, in keeping with the unconvincing parallel it keeps trying to draw between "Bandersnatch"'s storytelling method and the philosophical question of free will. When in fact, it's doubtful that any game publisher would hand their biggest release over to a developer who did not already know this extremely basic fact of narrative design.
Or maybe this is simply the latest expression of Black Mirror's tendency to prioritize gimmicks over any real engagement with challenging concepts. It's actually a bit difficult to get to the five-star ending. Or at least, it is if you're trying to make the right choices for Stefan and give him a happy life. The only way for Stefan to have enough time to make the game of his dreams, it turns out, is to kill his father. And the more times you resist "Bandernatch"'s urging to do this, the more likely it is that Stefan won't be able to finish his game even if you do give into it. So if you play the entire film without killing Stefan's father until you're forced to, you still won't be able to get to the "good" ending without starting over and playing again.
In other words, the game corrals you into making a morally reprehensible choice, and then—as in this interview with Slade, in which he talks about the viewers' complicity with Stefan's act—chastises you for it. The gaming industry has for years been rife with games that scold their players for playing them, and it's not surprising that a show as prone to glib cynicism as Black Mirror would dive headlong into that impulse. But it's not a conclusion that any self-respecting viewer should be satisfied with.
If anything good comes of "Bandersnatch", it'll be that more people discover the rich, fascinating world of interactive fiction as it's being expressed in the gaming industry. It might also spur a conversation about why those games work (and why "Bandersnatch" doesn't) that could finally lay to rest the notion that there's any value in an author abdicating their responsibility to tell a story. As for "Bandersnatch" itself, perhaps it's best to take its protagonist's exhortations seriously, and just choose not to play.
"Bandersnatch" follows a young computer programmer, Stefan (Fionn Whitehead), who in 1984 pitches a game publisher on a game based on the cult novel Bandersnatch, whose author went mad and killed his wife. Bandersnatch, the novel, is a choose-your-own-adventure book, and Stefan's selling point for the game is the heretofore-unprecedented ability to give players a sense of control over the story they experience. So almost at once, the film sets up several layers of meta-fiction—not only the game-based-on-a-book-within-a-movie, but the fact that Stefan is trying to sell the type of narrative he himself is living within. It's also a reminder that, as with 3D projection, the entertainment industry has taken cracks at interactive fiction on more than one occasion, usually for pure business reasons, and rarely to any great effect.
The film wastes little time in confronting the viewer with choices, some as trivial as choosing which breakfast cereal Stefan wants for breakfast, and others as fraught as deciding whether to kill someone. The viewer can take Stefan on story paths that end up involving drug trips, government conspiracies, time travel, and emotional breakdowns.
Unfortunately, it very quickly becomes clear that "Bandersnatch"'s creators have placed far too many eggs in the novelty basket, neglecting to give the individual branches in their multi-branch story a reason to exist in their own right, or anything resembling a common theme. We get glib pronouncements on the creative process, as when Stefan's idol, the bad-boy game programmer Colin (Will Poulter), advises him to work on the game alone, because "teams are fine for things like action titles, but when it's a concept piece... bit of madness is what you need, and that works best when it's one mind". And we get the poisonous equation of mental illness and creativity, as in a story-path in which Stefan agrees to up the dosage on his psychiatric medication, and subsequently produces a game that is dismissed by a reviewer as having been written "on autopilot".
The closest the film comes to genuine wit is when Stefan starts to realize that some of the choices he's making are not of his own free will, and even fights them off. This can lead to darkly humorous scenes in which he asks the viewer's advice, and then acts disgusted at our choices, or a sequence in which we can try to explain to him what Netflix is. But it's a concept that, like many others in Black Mirror, outstays its welcome to very little effect. It's all very well to point out that a character in a work of fiction doesn't have free will (and this is trivially true whether it's the author or the viewer who is making the characters' decisions for them) but it's not what one might call a particularly interesting observation.
As many commentators have noted by now, it is more fruitful to discuss "Bandersnatch" as a game than as a film, and especially in the current moment, in which interactive fiction of its type has been enjoying a surge of popularity. When Netflix first announced its intention to develop interactive entertainments, there was no small amount of skepticism expressed, including by people who remember some of the previous iterations of this concept. An obvious counterpoint to these reactions would have been to point to the success and cultural penetration of recent walking- and conversation-simulators like Gone Home (Fullbright, 2013) or Firewatch (Campo Santo, 2016). Obviously, the audience for this kind of interactive narrative experience exists, so why shouldn't a work like "Bandersnatch" capitalize on it?
The semi-derogatory term walking simulator is an oversimplified way of describing a game whose purpose is to carry the player through a narrative. While there are usually simple tasks the player needs to carry out in order to advance the story—go to this location, find that item—these are often little more than tokens, and some walking simulators, like Gone Home, are merely a structureless exploration of a single location. A conversation simulator also gives the player the ability to choose different dialogue options (or not to speak at all) which affects the game's narrative. In the eerie ghost story Oxenfree (Night School Studio, 2016), the player character's dialogue choices affect her relationships with the other characters, and make subtle changes to the game's ending.
In theory, "Bandersnatch" offers the player more freedom and control than your average walking simulator game. The reality is quite different, however, which can tell us a great deal about why "Bandersnatch" fails, and what it is that makes these games appealing. In Firewatch, for example, you play a lonely park ranger traipsing back and forth across a patch of Wyoming wilderness, with only the voice of a friendly but potentially untrustworthy colleague for company. The game offers the player no meaningful choices, locking them into a single story path and only allowing them to choose dialogue options. And yet it is undeniable that the experience of watching—or rather, playing—"Bandernatch" is much less engaging than an average playthrough of Firewatch. Is it because Firewatch has a better story? Or is it because the feeling of exploration you get from directing the player character is more satisfying than the actual ability to control the narrative in "Bandersnatch", which aside from its decision points gives you no control over its storytelling, down to the camera angles?
Some walking simulators, like Gone Home or Fullbright's follow-up Tacoma (2017), resemble immersive theater far more any game. And as in an experience like Sleep No More, the participant choices are expressed not in how they affect the story, but in how they choose to consume it—where they choose to look, what scene they choose to follow. In Tacoma, you visit a space station that has been evacuated and watch recordings of the crew in the hours following a disaster. You can choose to rewind the recordings, following different characters each time, or focusing on crowd scenes. There's nothing you can do to affect the story, which ended before your character arrived on the station. But your active exploration of it creates a sense of participation that is far more engaging than anything "Bandersnatch" has to offer.
The more you look at games of this type, in fact, the more obvious it becomes that no one actually wants to be able to make up the story they're consuming. Even if you move away from pure walking simulators like Gone Home, choice is something that designers have to carefully corral and shepherd, finally reducing it to an illusion. The mega-hit Life is Strange (Dontnod Entertainment, 2015) based its entire storyline on the issue of choice, even giving its player character the power to rewind time and change her choices after she's made them. A major throughline in the game involves the heroine's choices affecting the state of mind of one of her friends, which ultimately determines whether the player can stop her from committing suicide. But in the end, the only choice in the game that actually matters is the final one, and whichever option you choose, it will negate all of your previous choices, either by winding back time to the beginning of the game's events, or by allowing everyone you've interacted with and affected to die.
Even a game like Night in the Woods (Infinite Fall, 2017), which distances itself from other walking simulators by taking the form of a side-scrolling platformer and including some rudimentary running and jumping challenges, uses choice sparingly, to the extent of not even having a final choice. The entire final encounter with the game's chief menace is essentially a cut-scene without even any diverging dialogue options. The only effect your choices earlier in the game—to explore side-quests and get to know minor characters—have is to give this final confrontation deeper layers of meaning and significance. Oxenfree, meanwhile, leans into the expectation that players will repeat the game multiple times in order to get slightly different configurations of its ending by revealing that its heroine is stuck in a time loop, so that ultimately no choice in any individual playthrough can be said to have a lasting effect.
All of these games are deeply satisfying as both narratives and playing experiences—far more so than "Bandersnatch", which on the surface level gives the player more control over its story. The conclusion is inescapable: the impression of control and agency is much more important, when creating interactive fiction, than the actual ability to determine the story's direction and conclusion. On the contrary, giving the player, viewer, or reader complete control over a story is recipe for dissatisfaction. If you think back to the Choose Your Own Adventure books you read as a kid, were any of them particularly fun as stories? Or was it simply the novelty of skipping back and forth and trying to find the good ending (which is a lot faster and easier in a book than in a movie where you have to sit through the same scenes again and again) that kept you entertained?
The reason that letting the audience choose its own story keeps failing when the entertainment industry tries it is that it's a bad idea. It's the author's job to write the story. They can then choose a way to convey that story that gives the reader freedom in how they experience it. But if the story itself is merely a loose collection of different options, each in a different genre and with a completely different tone, then what they've created isn't a coherent work, but a self-indulgent mess—like "Bandersnatch", in fact.
In fairness, the film itself eventually reaches this very conclusion. In its most satisfying and thematically coherent ending (and the one the film signposts as the "correct" ending, since it's the only one in which Stefan produces a game that gets a five-star rating) it has Stefan explain that "I've been trying to give the player too much choice. So I just went back, stripped loads out. Now they've only got the illusion of free will. But really, I decide the ending". The problem is that the film presents this as a revelation—a slightly sinister one, in fact, in keeping with the unconvincing parallel it keeps trying to draw between "Bandersnatch"'s storytelling method and the philosophical question of free will. When in fact, it's doubtful that any game publisher would hand their biggest release over to a developer who did not already know this extremely basic fact of narrative design.
Or maybe this is simply the latest expression of Black Mirror's tendency to prioritize gimmicks over any real engagement with challenging concepts. It's actually a bit difficult to get to the five-star ending. Or at least, it is if you're trying to make the right choices for Stefan and give him a happy life. The only way for Stefan to have enough time to make the game of his dreams, it turns out, is to kill his father. And the more times you resist "Bandernatch"'s urging to do this, the more likely it is that Stefan won't be able to finish his game even if you do give into it. So if you play the entire film without killing Stefan's father until you're forced to, you still won't be able to get to the "good" ending without starting over and playing again.
In other words, the game corrals you into making a morally reprehensible choice, and then—as in this interview with Slade, in which he talks about the viewers' complicity with Stefan's act—chastises you for it. The gaming industry has for years been rife with games that scold their players for playing them, and it's not surprising that a show as prone to glib cynicism as Black Mirror would dive headlong into that impulse. But it's not a conclusion that any self-respecting viewer should be satisfied with.
If anything good comes of "Bandersnatch", it'll be that more people discover the rich, fascinating world of interactive fiction as it's being expressed in the gaming industry. It might also spur a conversation about why those games work (and why "Bandersnatch" doesn't) that could finally lay to rest the notion that there's any value in an author abdicating their responsibility to tell a story. As for "Bandersnatch" itself, perhaps it's best to take its protagonist's exhortations seriously, and just choose not to play.
Comments
Likewise, technically challenging games put you more in the role of a performer. This is most clear in Tony Hawk or Bayonetta where scoring looks at both your technical execution and style.
I'd point to role playing games as one of the story oriented that sometimes have an expansive player role, both more of an actor/improviser, in choosing how to portray your character, and a director in choosing what scenes and stories to pursue and which get left out.
I also hated that the game forces you to kill Stefan's father, when there's already such a strong cultural perception of people who experience psychosis as violent or dangerous. In reality the only person Stefan would be likely to pose any danger to would be himself. It's frustrating to me that portrayals of psychotic symptoms are extremely common in villains, but almost nonexistent in heroes.
I've played a number of role-playing games but am new to the term "walking simulator". A recent example is Detroit: Become Human by Quantic Dream, which does have different endings depending on whether you support a peaceful resolution to AI independence or not. The story is a guided tour and choices cannot alter the main path of the story, but can alter the context of different events.
A role-playing game that could alter direction like our real life choices would be a virtual simulator within an unfolding virtual world: taking far too many resources. One series I enjoyed for its decision points was the Mass Effect series by Bioware. Again, the story path was largely fixed, but the context and relationships with different characters could vary greatly depending on your choices (paragon/virtuous or renegade/ends justify the means).
If it's not too much to ask, as you have seen and read a great deal of Sci-Fi, are there any books or films covering the topic of Artificial Intelligence and the near future that seem prescient to you?
Post a Comment