Emergent Meaning and Narrative in the Digital Space
Addressing Tensions in Games and Game-like Media
“Most people want to be told a story. Leaving it up to a random number generator is dicey.” – Ed Del Castillo, Producer, Command & Conquer
Storytelling is one of the oldest forms of entertainment in human culture. Throughout the centuries we as a species have invented many ways to tell ourselves stories, and continue to do so. In recent years, games have become one of the most popular formats for delivering narrative, but as a new form of storytelling, narrative games face several critical issues.[1] Although the problems facing this burgeoning medium are not insurmountable, they are uniquely twisted by the fact that games are, by their nature, a participatory endeavor – a transaction between the designers and players where the contact is far more direct than it is in other forms of popular storytelling. Although as an industry games are growing and are doing much better than older, more established mediums in the marketplace, critical thinking about games, and especially how narratives in games are constructed, is still in its infancy. There are at least two fundamental unresolved questions in games criticism –, “how do games mean?” and “how do games tell compelling stories?” Answering both of these questions is a much larger task than I am capable of achieving in this essay, but by the end I hope to provide a potential framework for how games mean and from there briefly propose a solution to the narrative question.
In many ways, the medium closest to video games is film; the games industry has been moving toward making games more ‘cinematic’ for many years at this point. Similarities between the two mediums abound: both are primarily visual, the productions are often large-scale, requiring the work of hundreds or thousands of individuals, both games and movies are entered in festivals, and both are created by small independent artists as well as large studios. The road splits when it comes to the way that games and films intersect with those consumers for whom most are ostensibly produced. While film-watching is a primarily passive experience where the viewer sits and watches the material presented on the screen without comment, games seek to engage their players on a fundamental level where the player has control over the game’s environment. As David Bordwell writes,
…the classical Hollywood film presents psychologically defined individuals who struggle to solve a clear-cut problem or to attain specific goals. In the course of this struggle, the characters enter into conflict with others or with external circumstances. The story ends with a decisive victory or defeat, a resolution of the problem and a clear achievement or nonachievement of the goals. The principal causal agency is thus the character, a distinctive individual endowed with an evident, consistent batch of traits, qualities, and behaviors.[2]
This is not so in games: what occurs in a game is the result of a player’s actions as much (or sometimes more than) the work of game’s programmers, directors, artists, or anybody else involved in the project. The character, or “avatar,”[3] often serves only as a vehicle for the player, not as an entity independent from the player’s influence. This influence can be as narrow as propelling the protagonist of the story forward from plot beat to plot beat in a generally on-rails experience or as wide as choosing which characters live or die, or what narrative threads are pursued from a massive tapestry of possible choices.[4] This is quite a scaling-up of choices from what audiences encountered in 1934 when they viewed Ayn Rand’s Night of January 16th, a play with a game-like twist at the end which allowed a jury selected from audience members to decide whether a woman on trial was guilty or innocent, with two different ending sequences depending on how the jury voted. What was once seen as a gimmick is now the primary distinguishing characteristic of a major form of media. The open-ended nature of many game narratives allows for a new sort of narrative, emergent narrative, to manifest.
Regardless of whether their medium is physical or digital, games may be divided into two varieties. The first is emergence games, which contain “the primordial game structure, where a game is specified as a small number of rules that combine and yield large numbers of game variations, which the players then design strategies for dealing with.”[5] Chess, poker, backgammon – all of these games fit into this category. The second, progression games, are games in which “the player has to perform a predefined set of actions in order to complete the game.”[6] This category is more recent, and many video or otherwise electronic games can be categorized as such. These categories are not particularly ground-breaking ways to think about game structure – it is only when narrative considerations are introduced that questions begin to arise. Due to their strong linearity and aesthetic ties to older mediums, progression video games are the easiest method for delivering narrative – there is a straightforward series of events that the player must carry out in a specified order as determined during the game’s creation and development process. These games are very developer-oriented (or author-oriented one might claim). The player exists within a predefined narrative in a predefined world with predefined characters, and the player’s ability to interact and influence these elements is strictly limited by the narrative that the developer wishes to present. Traditional methods of analysis can be easily tweaked and function fairly well when it comes to progression games. This is not the case, however, with emergence games.
Emergence games need not have a narrative at all; there is no narrative inherent in a game of go fish, save perhaps the narrative that is created and later related when a player describes his or her play experience – for example, “Sarah and I were playing Go Fish, and she guessed my cards correctly six times in a row! I suspect that she may have been cheating.” These sorts of narratives are interesting, but they are not inherent to the game itself, arising only as a result of the play experience. Looking specifically at video games, there are many emergence games that include pre-authored narratives. Emergence games differ from progression games in that in progression games, the player may choose to interact with the presented narrative to a large degree, only tangentially, or not at all. The depth of presented narrative can range wildly, from rudimentary storytelling in games like Minecraft or Spelunky to deeply involved plot lines, in titles like The Elder Scrolls V: Skyrim or Left 4 Dead. These four games will serve, following a brief overview of the development of procedural generation in games, as my primary examples through the rest of this essay.
Any history of procedural generation and emergence in modern video games must begin with the 1980 game Rogue. In this game, the player navigates a dungeon that is procedurally generated as the player progresses, thus ensuring that no two plays of the game are the same. Rogue’s influence cannot be overstated; whereas many games are considered formative in their respective genres, only one modern genre is named specifically for its progenitor: the “roguelike” genre. While there are some issues inherent in the use of the word roguelike,[7] and the exact nature of the genre is currently still being debated, in this essay I will be using the word to describe a game that meets the following criteria:
Permanent player death: The player starts each new game session with a single life, and death results in all progress being lost.
Procedural generation: The area through which the player navigates is a randomly generated space that is not constructed by the developer (“bespoke”) ahead of time. The effects of items of the same name may vary across plays, or in other words, a red potion may heal the player in one play and damage the player in another.
Step-based gameplay: Movement takes place on a grid in the four cardinal directions. While you are not moving, nothing else is moving or acting. Each movement or action taken by the player allows all enemies in the level to take a movement or action.
This is by no means an authoritative definition, but it is illustrative enough to serve my purposes.
Roguelikes have traditionally not been a popular genre. The unfriendly nature of the gameplay put off widespread adoption of roguelike games by most players, who found games like Moria (1983), Nethack (1987) and Dungeon Crawl (1997) too punishing to be enjoyable. The genre found some success with Japanese gamers. The popular Mystery Dungeon series spawned a franchise that continues today with spinoffs featuring characters from popular series such as Dragon Quest (1993), Pokémon (2005) and Final Fantasy (2008), but games in this series have always been considered niche titles – they sell, but not often in blockbuster numbers, and when success comes it is attributable not to the genre but from the licensing of popular characters. It was not until 2011, when independent developer Edmund McMillen released The Binding of Isaac that the roguelike genre exploded into the gaming world’s collective mindspace, selling over two million copies[8]. While not a true roguelike under my above terms (it did not feature step-based gameplay), Isaac’s unexpected popularity and massive sales led to a heightened awareness of the power of procedural generation and emergent gameplay systems in both the game-playing public and the community of game designers. While games in other genres had used emergent systems in the past to enhance their gameplay, Isaac served as a turning point. Never before had there been such a strong emphasis on the emergent nature of the gameplay; that which had once been utilized either sparingly or under layers of narrative intended to shield the game systems from players was now placed front and center. Emergence had become king.
The success story of Minecraft, a game by Sweedish company Mojang, is the story of the triumph of emergence in perhaps its most distilled essence. More so than nearly any other game, Minecraft allows its players the complete freedom to build, quite literally, their own story. Presenting players with an infinitely large world map comprised of randomly generated lakes, mountains, deserts, oceans and other biomes, players collect hundreds of types of blocks with which they can build, block by block, cities, robots, computers, or any other number of things that their minds can imagine. It is not dissimilar from a never-ending bucket of LEGO bricks – if you can visualize it, there is a good chance that it can be built. The game cycles between night and day – while it’s light out, players are free to build, but at night enemies will appear to attack the player and his or her construction projects. Players are encouraged to build bigger, discover new realms, accumulate weapons and armor and eventually slay a mighty dragon. None of these tasks, however, is required for play. A player may even choose to turn off enemies, and play in a purely “Creative” mode, eschewing the “Adventure” mode entirely. The preexisting narrative in Minecraft is almost non-existent; there is no text in the game outside of menus, save for a short poem that displays before the credits after the player kills the dragon in the climax of the game’s Adventure mode. There are non-player characters that live in villages, but they do not speak. The world, being entirely procedurally generated each time a new game is begun, has no preexisting history. The story is left entirely to the players. This lack of direct narrative has caused many to attempt to fill in their own stories. Notably, Jason Rohrer’s Chain World imposes additional constraints on the player in order to create a narrative: the player is allowed only a single life, and when the player dies, he or she must save the game and pass the save file on to another person.[9] The new player inherits the world in the state that the prior player left it, and through a cycle of players passing the game from one to another, the world itself obtains a history, with prior citizens, “ancient” ruins and hidden treasures that players have left behind.
Dereck Yu’s Spelunky is a game with slightly more narrative than Minecraft, but still one deeply rooted in the emergent tradition. In an homage to Indiana Jones, the player steps into the shoes of the Spelunker, a bullwhip-wielding fedora-clad archeologist and explorer who has set out to plunder an ancient tomb for artifacts and search for the legendary golden city. The Spelunker will loot idols from boulder-trapped altars, confront the spirits of the dead and come face-to-face with the god Anubis before his quest is complete. The game is divided into five distinct worlds comprised of four levels apiece. The worlds and theming are predetermined by the game ahead of time, but the individual levels are procedurally generated. The computer builds the cave, checks to make sure that an unobstructed path to the exit exists so that the level can be completed, and then populates the level with enemies, treasures and traps.[10] The levels in Spelunky are designed such that every object has the potential for interaction with every other object, allowing for unexpected sequences of events to occur. For example, the Spelunker might trigger an arrow t