The Orchestrated Story and the Emergent Narrative

The issue of games being "about gameplay alone" versus games being a vehicle for moving and developing a narrative is nothing new. Since the earliest days of gaming, there have been attempts to use backstories for context and emotional attachment beyond simply completing mechanical objectives. The preexisting mediums of film and literature provided the foundation for game stories in the form of largely non-interactive narratives attached to a "game" concept: backstories, cutscenes, dialogue, etc. Games can do more than that. I've talked a lot about the concept of an Emergent Narrative, or a narrative that results naturally from the events that transpire during gameplay. Yet when discussing game stories, the tendency of gamers seems to always be to focus on uninteractive stories, as though only they have the qualities necessary to be taken seriously. So what's the difference? How does one, which relies on developing scenarios through gameplay, differ from the other, which ignores gameplay in favor of a pre-built story arc?

First off, let's define the term "narrative". Narratives are sequences of events in which characters interact with each other and with their surroundings. Characters are shaped by events, and set events into motion. The  qualities and depth of a given character is what allows us to identify with them, to take them seriously within the narrative, and to suspend our disbelief about the fictitious nature of the enterprise. Similarly, the events must be logical and coherently arranged in order to preserve our understanding of the faux-reality of the situation. In short, narratives are "events" acted on by, and acting on, "characters".

Obviously, however, a narrative is not usually just a retelling of events. There are always genre-related tricks to enliven those events and to make them more real, more memorable, and more artistic. In literature, this is called prose. In film, this is called cinematography. In games...well, there isn't really an equivalent, because traditionally we'd just use the things from literature and film. The fact is that "games" as a whole have not really embraced the idea that you can use the interactive, naturally developing events of "gameplay" as a way to tell a story, rather than simply a distraction from it. Our idea of "game story" is a movie or book attached to gameplay, with maybe some minor thematic connection between the two. The actual "interactive" games - the grand strategies, the city builders, the old-school RPGs - are disregarded because they're too "low-content" in terms of...prose and cinematography.

As it stands now, there are three major types of game story.

Linear
The linear narrative does not pretend to offer choice. It does not make "the story" a part of the game so much as it makes it a setting for the game. Characters from the "story" certainly intersect with characters from "the game", but they are not one and the same. A character who dies in the "game" is not dead in the story, because the story cannot allow for that sort of malleability. The story exists in a separate dimension from the game. You play the game to have fun and enjoy the gameplay; you watch the story to get invested in the events, and never shall the two meet. In some severe cases they may even operate on radically different rules: a normal wound in gameplay becomes a severe wound in a cutscene, etc. The story's effect on the game, at most, is to provide some context for the actions and create emotional attachment. The game's effect on the story, at most, is to serve as an obstacle to create some sense of achievement or "earning your ending".

Linear narratives, quite honestly, should not even really be considered "game stories". They are certainly stories, but they are irrelevant to their games. A linear story is like watching a movie, but being periodically interrupted by gameplay. In fact, that's exactly 100% what it is. Yes, there's some sense of achievement from earning "the movie" instead of just getting right to it, but that's still what it is. You play the game, and then when you're done messing around with whatever it is you're doing, you go back to watching the movie. It doesn't matter at all what you DID in the game, because you're not watching a game, you're watching a movie.

In simple terms, there's nothing wrong with a linear narrative, but it's getting absolutely no mileage out of the fact that it's a game. There's no interactivity between the player and the story, it's a movie. Obviously there are some very good linear storylines, but they must be analyzed as movies, based on the cinematography present in their cutscenes, the development of characters, etc. You can't even think of it as being part of a "good game", because you cannot look at the "game" parts of the product to judge the whole. There is no "whole". There are two separate, distinct parts. The movie part of a game is a movie, and thus must be judged by its cinematography and its storytelling capabilities (PLEASE WATCH THIS) (AND THIS TOO WHILE YOU'RE AT IT). Things like "fun" and "engaging" stop being important in a critical sense, and things like "writing quality", "character development", and "use of camera" start being important. Meanwhile, the "gameplay" parts of the game must also be judged on their own merits: it doesn't make sense to excuse bad gameplay in a game with a good story, because that's a movie with badly-made interruptions.

If either part of a linear game is bad, or at least if a player doesn't like part of it, it's going to be noticeable because you often cannot have one without the other. Bioware's Jennifer Hepler received a lot of negative attention - a LOT of negative attention - for saying that gameplay was her least favorite part of playing games, and that there should be an option to skip combat as per the ability to skip cutscenes. The ostensible reason she received this negative attention was that gameplay is considered an intrinsic part of the game experience, or at the very least it's something they're not going to ship the game without. Some games, like Deadly Premonition, weren't even supposed to have combat, and yet it was added due to genre expectations. Despite this, there are games that have no combat, and focus entirely on dialogue and decision-making; they're called interactive fiction, or visual novels, and they tend to have better choice-making than most computer-RPGs because that's their primary selling point.

Here's why I don't like the idea that "combat should be skippable". It's not because I feel that combat as it currently exists should be preserved or treasured. It's not because stories in games are a bad idea. It's not because I'm trying to say that people who like x element of a game shouldn't be allowed to. It's because it draws so much attention to how meaningless the combat is without trying to fix that issue. The concept is that "gameplay is irrelevant to a game's story", and relatedly that "combat in a story-centric game is often terrible". And you know what? It's true, at least for Bioware games and their ilk. If you skipped the gameplay in Dragon Age or Mass Effect or even KOTOR you would lose nothing because nothing important happens during gameplay. The people you kill don't matter, the potential for death doesn't exist, the objectives you accomplish are handwaved in cutscenes anyways. The reason it would be okay to skip gameplay is because you did nothing of value. The story is never affected, because that would provide too many variables. The only reason to play the game is if you're having fun with it, and if you're not it's just a big old roadblock between you and the next part of the movie.

One game that sort of messes with that concept of "cutscenes as movies" is Half Life 2. HL2 was well-known for "not having cutscenes", or rather having cutscenes where the player-character is the camera. These were cutscenes, okay? Let's get that out of the way. That's what they were. Characters went through their motions and said their lines and nothing the player did matter, you could be standing stock-still or jumping around like an idiot and it wouldn't ever matter. At the same time it robbed the game of the advantages of an actual controllable camera, such as interesting angles and perspectives, by strapping said camera to the boring, flat angle of "a guy looking briefly at the characters talking before resuming knocking plants off shelves" or, in some cases, "a guy looking around because he's been stuck in one spot". Bioshock had much the same problem; it's worth noting that in the most memorable scene of the game, the game itself took control away from you for dramatic and cinematographic effect.

With a linear story, there's still obviously the issue that gameplay doesn't matter; for purposes of the story, the gameplay doesn't even have to exist, and vice versa. Yet you can still find enjoyable experiences in terms of game story - you just have to judge them as a movie and not as "a game cutscene". Unfortunately, by that comparison, most game stories fall short of what would be considered a "good movie". I've heard many people suggest that a game story is different than a movie's story because movies don't offer interactivity, and thus writing for a movie is not the same as writing for a game or writing a novel. So let's talk about that, shall we?

Branching
A branching game story is a movie, but sometimes you get to choose which movie you're watching. That's it! I don't even feel like I should get more in-depth than that. Actual movies have had branching paths or multiple endings. Books certainly have as well; there's an entire GENRE of the concept. The fact of the matter is that, while the player is no longer wholly excluded from the story, the gameplay is still totally irrelevant, and nothing that happens during gameplay matters. While many of these games are praised for offering relatively good choices (such as the Witcher 2), the fact of the matter is that the "game" part still doesn't really matter. The story exists over here, and the gameplay exists over here.

Branching exists because dialogue and cutscenes are pre-made. Game developers make triggers, and then they attach dialogue and events to those triggers. If you reach x point, you are given choices y and z, which manifest themselves as cutscenes. And the thing is, that's sort of how they have to be. Writing can't be systematically created, it has to be written (for the most part - you can do a LOT with procedural content, but you can't totally replicate human speech patterns). If you want movie-level design, as most game developers do, then you have to make a movie and staple it to your game. The amount of crafting and care that people expect out of a game's story is based on movies. Movies are the standard by which cutscenes are judged, because cutscenes are movies, so it's basically expected that if you're going to have a dedicated story and not a bare-bones skeletal story you need to have movies to accompany it.

And yet, such games advertise themselves based on the ability to make choices and play roles - I mean, that's what an RPG is, right? It's more than just "a numbers-centric system", right? The whole appeal of the genre is the freedom to make choices, to have an influence on the story, to do things. What separates a game from a movie is interactivity, and that's why games like that tout it as a feature. The ability to change your environment, to tell a different story depending on what you do and how you play? That's amazing, isn't it? Plus, the "things you do" can be pretty fun when they include fighting monsters or shooting robots or whatever.

The problem is that games using this sort of hybrid mentality are never as good at what they do as a dedicated platform. When you get down to it, whether it's Dragon Age or Mass Effect or The Witcher or Skyrim or whatever, a "branching-path RPG" is a movie attached to a game that happens to have minor choices with regards to what movie you're watching. You could put the whole thing up on YouTube and then have little annotations for which link you follow (something that people already do) and that would be enough. Nothing about those types of games really relies on the medium of "a game", in the same way that nothing about a linear cutscene relies on the medium of "a game". The gameplay itself is rarely as good as a dedicated gameplay game, either; it's tolerated as an acceptable compromise because "well, I'm also playing it for the story, after all". We change our standards to support the model because that is what we are trained to do.

So if branching doesn't really provide enough real options, what's the alternative? How do we get actual choice - actual important gameplay - out of a game?

Emergent
The qualities that differentiate emergent stories from linear/branching stories is that the gameplay itself becomes relevant. The things that you do in the game change the story, and in some cases may even DEFINE the story. The game IS the story; if you take out the game, there is no story. The setting provides characters, and the gameplay provides events. Games like Way of the Samurai change dramatically if a character dies, or if an opportunity is missed, or if the player does something other than "proceed at leisure, pick whatever choice you want". The scenario is crafted in such a matter that the player's input is incorporated into the scenario, and the gameplay becomes important because it can change the story. What you do becomes part of the story, not just an irrelevant side note.

However, emergent games are weak in a few fields. Dialogue, for example, is something that cannot be replicated systematically. There are too many minor changes and inconsistencies in people's speech patterns to believably recreate speech through procedural generation. However, this is one of the few things that games cannot do, which allows me to focus on all the things they can do. They can create events by following logical patterns. They can negotiate potential consequences and responses based on a combination of theoretical answers and random chance. They can continue operation after a character has died or otherwise "left the system". They can make things happen, they can assemble a narrative, they can tell a story.

I'll give an example; in fact, this example is why I'm writing this article. Crusader Kings 2, recently released by Paradox Interactive, is a Grand Strategy nobility simulator. The player takes the role of a noble in medieval Europe (anyone from a lowly count to a king) and attempts to expand their realm and continue their dynasty. Crusader Kings 2 is most notable because of the incredible detail to the feudal system and dynastic successions that defined much of the medieval era, and it would not be an exaggeration to say that those two things make up the near-absolute majority of the gameplay.

The events that make up a narrative in Crusader Kings 2 are things like succession crises, civil wars, demesne claims, and holy wars. These things occur entirely naturally; when a ruler dies, their land spreads to their children, who may quibble and fight over their inheritance if they do not have a good relation. The complicated system of marriage-based alliances and inheritance policies leads to the kind of legally-justified wars that defined the medieval era. They're the kind of thing that Shakespeare wrote about all the time. Henry V is about the king of England enforcing his ancestral claim to France, inherited from Edward III, the grandson of Philip IV of France. Richard III is about the titular king using and abusing the laws of succession by essentially murdering and manipulating his way into a legally recognized position of inheritance and making enough allies to back his claims up with force. Macbeth is about the titular character assassinating his liege in such a way that his claim as a kinsman allows him to take the throne, and then becoming brutally obsessed with the idea of wiping out any potential claimants who might seek to take it from him.

These plots, and others like them, are based on the existing legal precedent of the period. They use the laws to form a basis for the characters' actions, in much the same way that we in the present have police dramas, stories about the wrongfully convicted, and so on. In Crusader Kings 2, these events are a normal part of gameplay that must be deal with as part of the interaction between characters. Every vassal and courtier has their own personality traits, their own motivations, their own ambitions, and their own ways of behaving. Every single one is a potential catalyst for an event, either in terms of causing them or being caused by them. A single vassal, if treated poorly, may gather enough support to try to take your throne, or betray you to another country, or cause a civil war by taking up arms against one of your other vassals. That same vassal, treated well, may give you crucial aid during your own war, or they may serve as a worthy tutor to your child and ensure the continuation of your dynasty, or simply support you in diplomatic and social matters. The player, as a ruler, is also called upon to make decisions during the course of the game: which vassal to favor, how to raise their child, whether or not to accept a call-to-arms by an ally, etc. The actions taken in the game influence the game's narrative, and in fact wholly define it. The setting and characters are generated by the game, and the events are a combined result of the game's system and the player's input. That's where the narrative comes from.

In one game, playing as the Duke of Toulouse, I found myself conspiring to acquire the Duchy of Sardinia (through inheritance) by marrying my son to its duchess. To successfully pull this off, however, I would have had to ensure that my heir (i.e. my next player-character) would be the uncontested heir to both my holdings and her holdings. To do this, I essentially had to kill off everyone who was in the way, including some of my own children. Even this proved to not be enough, as my son joined the church after marrying the duchess (thus removing him from the line of succession). Desperate, I assassinated him and had my second son marry the duchess, hoping that he would be able to have a child. However, the duchess chose to rebel against the throne of France, making Sardinia an independent entity - and making me, as a vassal of France, unable to acquire her holdings. The situation spiraled out of control until eventually I was barely able to hold onto my own duchy of Toulouse, having ultimately failed to accomplish anything despite the road of corpses I had made to achieve my ambition.

What this story illustrates is an interweaving of gameplay and narrative. Everything I did in that story - the marriages, the assassinations, the plotting, the scheming - was "gameplay". They were all part of the basic mechanical concept of the game. However, the fact that they were events given context by the characters and setting is what transforms it into a narrative. Each of those characters, my avatar included, had their own personality traits, their own desires, and their own ambitions. My choices were a reflection of my character's traits - he was ambitious, he was cruel, he was a schemer. Obviously I didn't HAVE to play him that way, but I did, because that's what made sense (and if I hadn't played him that way, the game would have eventually changed his traits to reflect my play-style). The events that happened would not have happened in the same way to any other player because they were based on the incredibly webbed-out sequence of events that had gotten me to that point in the first place.

That's my brother, and he's trying to kill my daughter. That's my BROTHER, and he's trying to kill my DAUGHTER.

Again, though, the weakness of emergent gameplay comes up: there was almost no dialogue in this whole affair. There were actions and reactions, negotiations and messages, but no true dialogue. There were events representing dialogue - a marriage proposal, a peace treaty, a goodwill gesture - but the dialogue is not really provided apart from a few key lines ("I gracefully accept your treaty", etc.). There is no prose to the whole affair. Yet at the same time I think the way CK2 handled it made that a strength, rather than a weakness. Your brain is an incredibly adaptive machine, moreso than a computer program. CK2, as a system, handles hard data like probability, logic, and concrete character traits. Your brain can fill in the story, because that's the part that is too "human" for a computer to recreate. When I look at my character's son and see that he's brave, loyal, and ambitious, I can characterize him based on that. When his actions go one way or another, I can rationalize his decisions and his internal struggles. The game gives enough information naturally that I can come to conclusions about who these characters are, how they think, how they feel, and how they're interacting with each other.

A few weeks ago I wrote about Yahtzee's article concerning both tabletop games and his character's backstory in The Old Republic. In that article I criticized Mr. Croshaw for being (a) wrong about tabletop game stories being "entirely made up" and (b) hypocritical about his video game backstory. He accused tabletop games of "making games up out of nothing", which is untrue as the whole point of the rules is to serve as moderation for the story. He claimed that his backstory for TOR was better because it had a grounding influence in the form of the game itself, which turned out to be irrelevant to every detail of his backstory. Nothing about the game supported his backstory; it existed only in his head, and meant nothing "in game".

Crusader Kings 2, and games like it, are examples of stories where the system provides almost everything in terms of events and characters. It's not just limited to developed narratives like the one I offered; any game where a player can say what happened during gameplay is like this. The nature of gameplay as being a largely random state of affairs, rather than a scripted story sequence, allows players to experience things that other players haven't, which allows them to exchange information and have it still be new. I've played Hitman: Blood Money for days and I still hear stories of people finding new and innovative ways to take down their targets that I haven't even thought of, because it's not just "do this, target dies", it's about arranging things based on the tools and resources available to you. The fact that each potential option connects to many other options means that there are an exponentially large number of available choices.

You see, Yahtzee was right about one thing. The structure and enforced reality of a game does add a certain value to its proceedings: you can't just make shit up, things happen in a logical fashion. The story is given merit because it "actually happened" to some extent. However, the example he gave was that of a backstory unconnected to the game's events, because that's all that The Old Republic will let you do. The game lets you "make choices", but they're so incredibly limited - and so much characterization is essentially automated - that it's not a question of you making a character, it's a question of the developers making a character and you get to pick little things for him or her.

Now maybe some of you will say that, within the context of the game's rules, you can still use your imagination. Yahtzee imagined his character's backstory, he imagined his character's motivations, and he made the choices that he thought suited his character. The problem with this is that, unlike Crusader Kings 2, the dialogue is all there in the game. CK2 gives you choices and lets you mentally expand upon them because it doesn't give you much concrete dialogue. TOR gives you choices, yes, but also shows you exactly what your character is saying and how people are responding. The fact that they are giving you so much explicit information about what's happening means that you cannot use your imagination without overriding a huge amount of the data given to you - at which point the game's story becomes a total hindrance, not a guiding set of rules. It's hard to pretend your character is a stoic mercenary when the only "give me money" choice is accompanied by snarking and quipping. You're not "roleplaying", you're selecting a pre-generated character.

It's a lot like the Star Wars prequels and the effect they had on original films: it's better to not know anything, and thus be able to use your imagination, than to know everything about something and have it turn out to be poorly-written and poorly-executed. The stories you made up in your head about things like the Clone Wars, Obi-Wan and Anakin being friends, Anakin distinguishing himself as a great pilot, etc., were probably better than what the actual prequels were, and they didn't cost a dime to imagine. You didn't have to hire any CGI artists or special effects guys or anything, you just filled in the gaps on your own using your brain, and that was good enough for purposes of getting immersed in the story. They gave you the characters and the events, and you (as a viewer) assembled them into a coherent story with your imagination.  Emergent stories work exactly the same way.

Conclusion
The issue is that each type of game needs to play to its strength: if a game is going to be a movie, it should be a REALLY GOOD movie. If a game is going to be an exercise in decision-making, it should cover all the potential details and really let you get into the universe instead of throwing 2 or 3 pathetic options at you that don't mean anything. An orchestrated story is like a concert, whereas an emergent narrative is like an instrument. It's okay if you spend a bunch of money on an instrument and can't play it perfectly the first time, because the process of discovery and putting songs together is part of the experience. It's NOT okay if you spend money on a concert and the musicians are doing the same thing, because the fact that they're professionals and you're paying for a completed product totally changes the product. The quality of the work needs to drastically make up for the lack of interaction.

What I'd like you to take from this article, meandering as it is, is that if you're going to make a game, make it good at what it does. If a game is a movie & gameplay, both of those need to be as good as they can be. If a game is trying to get the player involved in the story, their choices need to be as important as they can be made. This idea of accepting sub-par products because "it's all we can do, that's just what games are!" needs to stop. If you're going to make a story game, make it really good at telling a story. If you're going to make a game for the gameplay, then make it really good. Different philosophies are great. Different types of games are great. Lowering standards because "it's just a game" is why nobody takes the medium seriously.