Monthly Archives: August 2007

Narrative as Context

I noticed an interesting phenomenon when I played the game Planescape: Torment. Around 2005, I decided I had heard enough about what a classic this game was, so I got it and played it. It didn’t work for me. The pacing was slow, it failed to hook me, I was bored. I stopped playing in the first chapter.

At some point this year I happened across the Wikipedia article for Planescape. I read about the universe, the themes, the planes, the Blood War – this entire Planescape mythology. I soon found myself playing Torment again. And it was fun. It went from ‘confusing’ to ‘deep’.

The gameplay had not changed. The only change had been in the context. Every action had a lot more meaning for me. And thus the game became fun – even though it was exactly the same game.


We put narratives into games for two main reasons.

The first reason is that the enjoyment we get out of absorbing the story that is fed to us. This happens in all story-based media. What’s special about games, however, is their interactivity. Interactivity means that a storyline needs to fulfill an entirely different purpose, which doesn’t exist in passive media. In a game, a narrative also serves to to create context for, and thus give meaning to, the player’s actions.

This second reason is both more important in gaming, and also poorly understood. We should not be aspiring to beat classical media at pure narrative quality because in this arena we work at a significant disadvantage. We lack an established narrative “appreciation industry” for game stories, and the format inherently destroys the ability to precisely control pacing and tone.

Game narratives should focus on providing an effective, meaningful context for the actions of the player – not the NPCs. The best game narratives aren’t good because they are great stories in a classical sense, but because they enhance the emotional impact of the gameplay so effectively.


Take Quake II and the first Half-Life. Fighting a Strogg in Q2 isn’t that different from fighting a bullsquid in HL. Half-Life‘s battles, however, are more emotionally provocative, because Half-Life presents those battles within a much more effective narrative context. Why is Half-Life‘s context more effective?

I don’t personally know anyone who has killed hundreds of alien warriors, alone, on an alien planet, nor have I heard of anyone doing this. Were I asked to imagine myself in this role, I would have difficulty relating to it. It is so far out of normal human experience – it has so little in common to a normal life – that stepping into the role is impossible. Each dead monster, then, represents only a victory over a polygonal enemy in a computer game.

In Half-Life, however, the story is plausible. It is not realistic per se, but the characters are superficially humanlike, and the environments could exist. Gordon Freeman has a name, and interpersonal contact with other friendly humans. It is possible to step into his role and become him because his life makes sense and relates to familiar themes. Killing monsters in Half-Life isn’t just killing monsters; it is participating in mini-stories. I.e. “I quickly blasted that monster with my last few rounds so I could save the scientist in time who was calling for help.”


There is strong evidence to show that the main reason human beings have developed exceptional intelligence is not to gain advantage over nature. Many species of animals have been successful for millions of years with very low intelligence. Even tool-using animals like chimpanzees consistently remain at a relatively low intellectual level for aeons.

The reason humans developed intelligence was not to outsmary nature – it was to outsmart other humans. Mostly, this means social competition. The development of language allowed the creation of fantastically complex social strategies for achieving and maintaining social status. This is the root of our superior brain power. The implication is that human intelligence is much more geared towards interpreting and solving social and narrative problems than any other task.

For example, all ancient cultures understood nature by personifying it in spirits or gods. Why is this strange, roundabout way of explaining natural phenomena so universal? Because thinking socially and narratively is most natural for us. Even though social interactions are fantastically complex from an objective standpoint, we still find them easier to decode than abstract problems which are much simpler from an objective standpoint.

The implication for game makers is that we can always count on players to generate stories from the events in a game. Even given a hyper-simple representation of some gameplay, a player will tend to personify the active elements and subconsciously create some sort of narrative. Hence people view moving paddles in PONG as “wanting” to stop the ball, when in fact, they are just abstract data structures being displayed by photons on from a phosphor screen.

This spontaneous narrative generation is far more powerful than the ability of game makers to directly elucidate a narrative. Furthermore, it is mostly subconscious and happens so naturally and effortlessly that players do not even notice it. They simply feel that they are having an emotional experience.


Players will create a story from almost anything, but that doesn’t mean we should make their lives difficult. It helps to give the players a context in which the story is set, so that those spontaneously player-generated stories can use this context as a backdrop.

The most obvious way to do this is to create a backstory from scratch. Creating this sort of story context, however, tends to take a long, long time both for the creator and for the player. The amount of work and exposition requires to create and communicate a completely original universe is almost limitless.


A better solution to this problem is to use some degree of borrowing. Use a story context that your audience already understands.

The most obvious examples of appropriated story context are in licensed products like Star Wars or Star Trek. These universes have expansive and mature mythologies. These are limited, however, because a player will need to be familiar with them for them to have a positive effect on the game.

A second option is borrowing from various “stock contexts”. Whereas damatic writing sometimes uses established “stock characters”, sometimes games borrow from established “stock contexts” which have been established from many previous works. The most obvious stock context is classical high fantasy, in the vein of Lord of the Rings, Warcraft, and Everquest. These products all vary slightly from the high fantasy context, but the common elements are much greater than the differences.

A third option is to use a historical context, like the second world war.

Also very, very effective and underrated is borrowing from real life. Consider the success of The Sims – a game which many people (including myself) thought would die a quiet death, before we realized how important the power of context and player-created narratives are.


Take an example from GTA: San Andreas. GTA:SA borrows its entire context and feel from the early 90’s West Coast rap scene. Music is borrowed. Stock characters are borrowed. Themes are borrowed. Visuals are borrowed.

It worked fantastically well for these reasons:

-The context is socially acceptable and ‘cool’.
-The context is not already overused by other games.
-The context is immediately comprehensible by GTA:SA‘s target audience. This means much of the contextual groundwork is instantly laid as soon as the theme is advanced.
-The context has a rich mythology of stock characters, stories, and themes to draw from.
-The context is different enough from most people’s real life that it allows players to suspend disbelief even while improbable events are occurring.
-The context is close enough to our real lives that we can identify with the characters and our own role.
-The context make sense as a backdrop for all sorts of conflict, drama, and violence.

All these things come together to provide an excellent basis for subconsciously-player-created narratives.

This idea possibly to be continued in later articles…