Game reviews are more or less informative, straightforward affairs. Pithy observation, narrative summary, 4/5 stars. Some people do this better than others. Some people say 4 is not the right number of stars. Some people don’t like stars.
Games criticism, a distinctly different topic meant to be concerned with how games work instead of just whether they do, is a tangled, complicated, scary catastrophe. No one agrees on who does it well. Its dark secrets are hidden either inside some of Amazon’s spookier publishing crevices or on blogs that don’t have URLs so much as ritualized handshakes. There are many good theories on how to make individual games, but judging finished games in critical terms seems paradoxically more difficult. How do you examine an art form that is 30 years old, has changed drastically in that time, yet whose first trembling steps remain culturally and critically relevant? How do you create a theory that describes the important elements of both Monkey Island and Amnesia? How about Passage? How do you consolidate huge differences in genre and style, then discuss where the art is heading?
Can games tell meaningful, interesting stories–and if they do, will we care? This particular question annoys some people, and that’s only one reason it’s worth asking.
A popular notion, particularly among developers, seems to be “more player control is better,” followed by “emergent gameplay/storytelling is better than pre-scripted narrative.” Terrence Lee says precisely this in a post that has gotten a lot of attention. He begins by saying games should stop using cutscenes because they distract the player and defuse emotional tension; he’s not really wrong. Cutscenes are often intrusive, and quick time events are almost always frustrating. In Lee’s opinion, stories that are driven by the player’s actions and forego distracting mid-game cinematics are always more enjoyable.
To his credit, if you play enough Minecraft you will eventually have a great story. The game features an open world with no stated goals, so the player is compelled to seek their own adventure. Even though the game is just made up of basic behaviors, simple interactions between the player and the world lead to events that feel unique. Likewise, Eve Online is a constant freaking source of real world subterfuge in imaginary space, often in ways its developers didn’t see coming, and almost everything within the game is run by players. This idea of players experiencing complex events from how game mechanics interact is what Lee ultimately means by “emergent narrative.” Emergent experiences can be empowering and fun, as it can feel like a whole digital realm is acknowledging your existence.
Removing cutscenes from games and focusing more on the player’s actions would probably help more often than it would hurt. If Lee’s argument stopped there, it might be fairly palatable.
But then he calls out Portal as being one exception to the rule…and there aren’t any cutscenes in Portal. Moments where you have to listen to GLaDOS and laugh until you hurt, sure, but no real cutscenes to speak of. As he continues, Lee serves up examples of what he thinks games should emulate: Journey, Brogue, Dwarf Fortress. While those are all great games, they all lack any overt, handmade story. Lee goes on to say that cutscenes are just a symptom, and any narrative that isn’t created by a player is the real problem:
What I mean is, instead of having any scripted elements at all, we let the explicit story describe the player story. We let the plot, climax, and characters all emerge from what the player experiences. In short, the story describes what the player did, instead of what the player needs to do.
Representing a post that long with a few sentences may seem blithe, but that quote really does sum up his idea. Games that leave the player alone with their environment are great, but what about games that focus on the player’s impact and have a scripted narrative? Even if every game strove for both, we’d lose some awesome stories that don’t allow for player agency. Requiring player agency also means ditching most art games, (some of) which are great.
In the closing paragraphs, Lee says developers who want to tell explicit stories should keep the player in mind. As he’s only posited a single example of good explicit storytelling, versus over a dozen examples of doing so poorly, this feels like some rather snide advice.
He also says that emergent narrative remains relatively unexplored, and this is just wrong. The idea that emergence is king, that it can only thrive in a very specific way, and that narrative storytelling is never optimal–these things warrant closer examination. I don’t think Lee is alone in these opinions, which I find troubling (in the case of that second link, Spector promotes Dishonored as a great example of emergence, but Lee condemns that game as prescriptive drivel–so there’s how wildly that word is used). I don’t think critics who hold similar opinions are appealing to the difficulties of indie game development. I don’t think creating more procedural games is an end-all solution to games narrative, but if you have an awesome idea for a procedurally generated game, make that. I do see many of Lee’s points as part of vitriolic tirade against AAA game developers and popular review sites. Those companies don’t need help defending themselves, but railing against them doesn’t help anyone, including indie studios.
At a minimum, Lee and other outspoken critics suggest that narrative in games is all just smoke and mirror garbage, and no one really cares about it, we’re just fooling ourselves. At their worst, they appeal to academic sensibilities and use criticism in a way that isn’t tolerated by any other art form, then disregard any dissenting opinions as anti-intellectual.
From here in, let’s examine the traditional usage of emergence, how it relates to storytelling, and maybe consider what direction games criticism is headed in. Keep in mind that I’m not an authority figure, I just play a lot of games, know other people who do that too, and I think most people don’t appreciate being told that practically everything they like is stupid. Of course, we should still strive for better games, but maybe we can have it all, eh?
Traditionally, “emergence” describes complex systems that arise from simple interactions. Put differently, it is the often surprising result people observe from a set of seemingly straightforward circumstances and behaviors.
Emergence is usually noted in games with behaviors that react to the player and with each other in meaningful, quantifiable ways, as opposed to scripted gameplay that acts upon the player more or less in predetermined order. In common usage, an emergent story isn’t progressed by an easily reproducible or rigidly scripted series of events. Making Nathan Drake climb a mountain to collect treasure is probably not an emergent experience.
Creating your own little pen of piglets in Minecraft only to have them get exploded by a sneaky creeper is both an emergent story (the game didn’t have that event hard coded into itself) and emergent gameplay (you are suddenly faced with rounding up surviving piggies, which is not generally the game’s focus). In a different way, the feeling of dread you get as you explore the world, terrified of losing your pack full of diamonds, is also emergent. The game isn’t expecting you to be scared, that’s just how you feel against the horrific possibility of losing your hard work so far from home. Spelunky may seem like a simple platformer to first time players, but becomes more like a puzzle solving game when experienced users debate how to traverse the ever-changing dungeon. Emergence in this sense is ultimately a label players apply to interesting experiences and challenges they meet in a detailed game environment.
Players exploiting game mechanics in ways designers don’t explain or anticipate, often in the form of subversive play, can also be considered a form of emergence. This can range from tilting a pinball machine to murdering your Sims spouse by taking out the pool ladder; the idea is that players find ways to get what they want by using the game’s systems (without using external hacks–though I could see an argument for this counting–and excluding cheat codes and the like as those are an intentional game feature). Speedruns may reasonably fall within a designer’s gameplay expectations. Doing backflips the entire game because it’s a slightly faster way to get around or finding bizarre glitches that allow the player to slip through seams in the world so they can beat a stage in seconds probably don’t. Subversive emergent play is seen in Eve Online so often (usually enacted on other humans and occasionally in the form of a long con) that it’s basically a primary feature.
When a player twists game systems, it’s usually to their own benefit. This may just mean messing with the game for the sake of exploration. It also may mean that sprinting around with a knife or dual wielding shotguns becomes the most efficient (and therefore popular) way to kill people in a tactical shooter. This sort of chicanery is usually balanced out posthaste. World of Warcraft patch notes expose a tremendous history of players having too much fun, infinitely stacking deer and the blood plague being some of the more striking examples. While some of these quirks have seriously disrupted their game, it’s still a little depressing when weird, player-discovered madness is culled. You can feel the game’s benevolent operators saying, “we don’t have the capacity to deal with this, sorry.” It’s the kind of spirit and circumstance that leads to unkillable NPCs, binary morality meters, and having to press “e” to help Elizabeth. You can feel a designer’s desire to keep you involved and make you feel as if your decisions matter in a quantifiable way, but how they insist on doing so is sometimes off-putting; for better or worse, these things expose a game’s underlying framework.
Similarly, anyone who has bumped into a Skyrim guard or played a Roguelike knows when an open-ended simulation is working and when it’s not. It feels great when you become the leader of a guild and the characters you’ve been working with congratulate you, or when you’re on you’re on your last hit point and stumble upon a healing potion. On the other hand, when a guard compliments your level one character on their destruction magic or when you have to swim across a shallow lake full of eels that are sure to kill you for the third run in a row or you spawn with arrow traps on all sides with nothing to spring them except your own squishy body, the seams become visible. When you bounce a piece of furniture off of someone’s face and they just continue staring at you, you know you’re playing a video game.
So, narrow, arbitrary mechanics and strange restrictions can frustrate the player. Procedural generation or random events (staples of open world games) can create annoying, flow-breaking moments–even if the player is leading the story about. That doesn’t mean there isn’t a case for these games, but clearly the quality and depth of a game space is important. If the player’s options fall short of their expectations, they will be dissatisfied. On top of this, we have quicktime events and awkward cutscenes. We can jumble these things together and make a really wretched game.
This all roughly equates to two kinds of emergence: when the world surprises the player by reacting to what they do, and when players contort the world into doing what they want.
Taking just these two categories of emergence and the problems that can undermine them in hand, the clearest path to emergent gameplay does seems to be creating an open world with simple behaviors and little narrative so the player isn’t restricted. Emergent gameplay and stories can be very powerful, so many developers would like to encourage them if possible. A few objections to this mindset quickly crop up.
The first is that Planescape: Torment —essentially an okey-dokey combat system on top of a very thorough choose your own adventure novel–is heralded as one of the best games ever made. Its basic plot setup uses one of the oldest, most frequent cliches in writing, and it still adored. Add to this games like Gone Home, The Walking Dead, and basically every adventure game ever made, and you’ll come up with a list of games that would be essentially nothing without their scripted narratives. And that’s before considering games like Portal and Gunpoint, which would still be fun without its story, but their writing is fantastic. The refutation for this seems to range from “players have been taught to want the wrong thing” to “most players are stupid for thinking story matters.” If you think that, okay, but if that’s true a lot of people have the wrong idea, myself included.
The second issue is that every design choice affects the player–both what’s included and what’s omitted. Saying a scripted narrative restricts the player is one thing, but everything in a game is meant to encourage the player to action–from what colors walls are to the pleasant noises you hear when leveling up. Some decisions will just be about what’s generally fun, but all players have biases. If you’re trying to soothe the player by ensuring their desires are met, you have a steep hill to climb. A designer can make the most interactive fantasy kingdom imaginable, but if a player absolutely loathes dungeons and collecting treasure as motivations, they probably won’t be grabbed. Parallel to this, some players want an implicit story in their games. For all the people who love Minecraft, not everyone is keen on its goal-free environment.