The academic homestead of Annika Waern

Why I don’t believe in emergent narratives

(This post is about videogames only. There will be a follow-up post about larp.)

Every late autumn, I teach a course in game analysis at Stockholm University. We cover Salen and Zimmerman ’Rules of Play’ cover to cover with some sideboard readings, and the homework assignments include playing games and analyzing them from various perspectives. One of the homework assignments includes this question:

”To what extent does the game create any emergent narrative? Explain your answer.”

Students have huge problems answering this question for just about any videogame they analyse, and I  understand why. To summarize, emergent narratives are not created by games, but by players.

Morrowind aimed for a rich story world, and used a combination of scripted stories and procedural content (such as the reputation system) to create it.

The concept of ’emergent narrative’ is most commonly seen as part of a more general design ideal of a ’360 illusion’, a complete and consistent fictional universe that is there for the player to explore. The idea is that such worlds also could function as story spaces, in which numerous stories could emerge depending on what the player chooses to do and what to explore. (The 360 illusion is a very old design ideal that predates computer games, and for video games it has been advocated by e.g. Jane Murray in ‘Hamlet on the Holodeck’.) Games that have aimed to achieve this ideal include many single-player RPGs such as Morrowind. It is not an ideal that appeals to everybody: many players value challenges more highly than exploration, or favour a good story above a complete simulation.

The problem is that a rich and complex world is just a world. Worlds are not in themselves stories or challenges, but they can function as a canvas or a backdrop for both. (I believe it poses issues for both, but here I focus on stories.) As a storytelling canvas, the 360 illusion poses several limitations. The most obvious one is related to time: unless you tweak the representation, the player can only experience any plot in the order it is played out, as a story world ‘now’. For this reason, most 360 illusion games include a back story that is both represented in the world (through architecture and relics, clothes and cultures, posters, books lying around etcetera) and in dialogues with NPCs in the world. The back-story lends itself to free exploration but is static – in this sense little has changed since MYST. Already the separation of the back-story from the story in-game sets games apart from traditional narrative media, but more seriously, the stories that play out within the 360 illusion will always be told in the order that events happen in the world. This is true even when stories are triggered or generated as an effect of the player moving freely in the world and selecting her own interactions. More interesting and complex narrative structures can be created as players re-load earlier saves or replays the entire game to explore various options – but these are narratives generated by players and only occasionally foreseen by designers.

A second limitation is that it is the player who will play the protagonist in a 360 illusion game. In literature, it is possible to present stories from a first-person perspective or a third person perspective, and in film, the perspective is typically a third person perspective; all forms offer plenty of ways to present the thoughts and emotions of the protagonist to the viewer or reader. But when the player enters a 360 illusion world through an avatar, it is to play the role of the protagonist. If the game prescripts what the avatar can say or do, or how it should react, it will limit the expressivity and freedom for the player. But if it does not, it will be up to the player to create all those reactions and emotions – and there is no guarantee that he or she will. I believe that when Tom Cross’ argues for creating meaningful games by populating them with NPCs with their own goals and purposes can be seen as a way of overcoming this problem: using his model, the player becomes an in-game spectator and less of a protagonist.

The third limitation I want to bring up is both a limitation and a potential: the randomness of a world open for free exploration. There is a level of randomness already in letting players explore a world at will, and another layer in that the events they trigger can affect each other. But narratives are not perceived as random but as intentional structures: Tom Cross puts this as ”narratives speak of an intelligent, exterior design to readers”. When we are looking for narratives, we are looking for the intent to tell a story.  Random occurrences will still seem random unless they have been carefully designed (and yes, it is possible to design for coincidence, as discussed by Reid for location-based games). Hence, most computer RPG games include pre-scripted story content where the intent to tell a story is clearly in place, that serve as a context in which random occurrences become meaningful. Cross argues for procedural narrative: stories that have been designed into the game but that are generated rather than pre-scripted.

Hence,  in order to create coherent stories out of random coincidences (with or without linking to pre-scripted story content), the player must be active in making the connections. The player becomes not just the protagonist but part author of his or her own story. Steve Gaynor talks about this as ‘storymaking’:

“One should not ask a game designer to tell them a great story; rather, the game designed should be judged on the player’s ability to make his own stories within its mechanical framework.”

If we are given nothing but random coincidences from the 360 illusion, this is a very hard task. Such stories do not just ’happen’ or ’emerge’ out of a rich world to explore – they are actively and consciously created by players. It is not necessarily something that every player is good at or even wants to do, and furthermore, the 360 illusion is not necessarily the best way to support storymaking. Calling this form of narrative ’emergent’ just obscuring what is really going on.

Stay tuned, because there are some that got this right – the table-top and larp communities.


Cross, T. Analysis: Story and the trouble with ’emergent’ narratives. Gamasutra, July 10th 2009.

Elder Scrolls III: MORROWIND. Bethesda game studios 2002.

Gaynor, S. Storymaking. Blog post, Jan 31st 2009.

Murray, J. Hamlet on the Holodeck. MIT Press 1998

MYST Brøderbund, Robyn and Rand Miller, 1993

Reid, J. Design for coincidence: incorporating real world artifacts in location based games. Proceedings of the 3rd international conference on Digital Interactive Media in Entertainment and Arts, Pages 18-25. ACM.

Waern, A, Montola, M., and Stenros, J. The three-sixty illusion: designing for immersion in pervasive games. Proc. CHI’2009.

Single Post Navigation

6 thoughts on “Why I don’t believe in emergent narratives

  1. Pingback: New in Larp: July 22, 2012 >> Lizzie Stark

  2. Pingback: Storymaking in sports (a brief interlude) « Persona

  3. I had an incident in Minecraft that started my interest in emergent narratives and also illustrated the problem you point out.

    Something happened with the villagers, a result of my actions and their simple AI. I was trying to roleplay and almost instantly thought of a story to explain what was happening – but had no way to express that story in the world. Even if I could, it would have been by telling the villagers how they felt about me, the player.

    If there were some ‘magical’ way for the game to understand at least how I want the characters to feel and act, without conscious input, maybe a meaningful narrative could start to emerge?

    • Waern on said:

      Well, we all wish for magic programs now and then, but it’s kind of hard to build them …

      You could check out some of the work on character-based story generation systems, e.g. . But I am a bit sceptical, still, because most of the models that I have seen have still placed most of their emphasis on simulation, even if it is character simulation rather than world simulation. Realistic characters might not make the best stories! You should also check out Michael Mateas’ research game Facade, as it has a separate story engine.

  4. Pingback: Storymaking in Larp (an overdue post) | Persona

  5. Pingback: Larp Design for Storymaking | Persona

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: