Making Games That Make Stories

by James Wallis

In the ongoing debates about storytelling and narrative in games, the various commentators often overlook a key point: even in the most cutting-edge examples of the state of the art, it is not the players who tell the story, it is the game. Whether computer games with a narrative element, board games, card games, or face-to-face role-playing games, the essential plot and structure of the narrative is predetermined before the game begins, and cannot be altered.

This style of game is interactive storytelling only in the sense that the story is told. The plot is not interactive. It may contain variables, most notably multiple endings, but the basic shape of the story is fixed in advance and the only way the player can divert the storyline away from its pre-set path or prevent it from reaching one of its predetermined conclusions is to not finish the game. The player may have the illusion of being in control, but this is usually down to clever game design hiding the rails to give the impression of an open-ended narrative.

In many narrative games the plot is so untouchable that the players must watch it unfold in cutscenes or between non-player characters, while the gameplay consists of grunt-work that adds little to story or plot and would be elided in a novel or film - journeying from place to place, exploring buildings, locating objects, killing monsters, shopping, and deciding what to wear. Replay the game and the same story will be revealed in essentially the same way. Even in titles like Knights of the Old Republic, a Star Wars-themed console RPG hailed for letting players choose whether to embrace the light side or the dark side of the Force, one must still play through the same key encounters in the same order. In a slightly different vein, Fable allows players to craft their own story-path and customize their character, but while many of the game's encounters can be played in any order or even omitted, they are all pre-scripted around a central narrative.

Thankfully this is not the only way that games can handle stories. There is a small but growing number of titles that use the creation and telling of a new story (or in some cases, several stories) as an integral part of the gameplay. These are more than systems for group storytelling: they combine narrative elements with standard game features including competitive play, a clear ending point and a winner. For the purposes of this article I will refer to them as story-making games. Some significant examples in the field include Dark Cults (1983), Tales of the Arabian Nights (1985), Once Upon a Time (1993/1996), The Extraordinary Adventures of Baron Munchausen (1998), and Pantheon (2000).

Of those five I designed two and published a third, and this is not a coincidence. I think story-creation as a part of gameplay is important and under-explored. Books, films, and TV have conditioned us to be passive consumers of stories, not active participants in their creation, and games have gone the same path. Story-making games, by contrast, offer a way of creating a story within the structure of a game, which not only helps to structure the narrative but adds interaction and competition.

Some critics have complained that the act of storytelling is intrinsically cooperative and at odds with the competitive element of gameplay. But balancing the two sides of that dichotomy is one of the things that makes it such an interesting field. The challenge of simultaneously telling a story and playing a game - or if you prefer, telling a story while thinking not only of plot and character but also of tactics, strategy, and how to avoid the other players derailing you - is exhilarating.

Story-Making

Human beings like stories. Our brains have a natural affinity not only for enjoying narratives and learning from them, but also for creating them. In the same way that your mind sees an abstract pattern and resolves it into a face, your imagination sees a pattern of events and resolves it into a story.

Games have always had a close affinity with story-making. Adding a few lines of description to a video game or a background and artwork to an abstract board game gives dramatic context and an added sense of depth, allowing the player to create an internal narrative as the game progresses. To take an example, chess is primarily an abstract game but has pieces with titles that help in their anthropomorphization, and it is possible to create complete narratives based on chess games, as several authors have done. Examples include The Squares of the City (John Brunner 1965) and Alice Through the Looking Glass (Lewis Carroll, 1871).

The gameplay of most face-to-face games involving some element of strategy breaks down into a structure that looks a lot like that of a conventional story: setup, opening, middle section, endgame. That implies that mapping a story structure on top of standard gameplay, or at least fitting a game to a genre, is a fairly trivial job. It's not.

Take, for example, Cluedo (retitled Clue for the U.S. market), which describes itself as "the classic detective game" and is based on the classic English country house murder mystery story, as popularized by Agatha Christie. In most of these detective stories a crime is committed and subsequently analyzed, suspects are questioned, evidence discovered and analyzed, alibis examined and motives revealed, until in a thrilling climax the villain is unmasked. In Cluedo, you wander aimlessly around a board, not following clues but hazarding guesses at three unrelated elements of a murder. Not only can you win by proving that you committed the crime, you can accuse yourself and be wrong, and will lose the game as a result. As a game mechanic this works, but in story and genre terms it is a tale told by an idiot.

Cluedo is not, of course, a game that sets out to create or tell a story, though one might hope the best-selling murder mystery game in the world would make at least a token effort to be true to its genre. On the subject of Clue Dungeons & Dragons (2001) - in which the murder-mystery is combined with a wandering-monster mechanic, so players are trying to solve a killing while killing things - the less said the better. Instead, it attaches the tropes of its chosen genre to aspects of its gameplay, to create the illusion of a whodunit.

It is possible to create a story-making game about murder mysteries. In my forthcoming title, Youdunnit, currently doing the dance of the many publishers, players play characters who must work together to solve a murder that one of them committed. The members of the group are assigned characters, each of whom knows certain key facts about the others. They take turns to present pieces of evidence that can be canceled out or combined to form chains of means, motive, and opportunity, until one character has their guilt "proven" to the satisfaction of the group. Each Youdunnit case is about a specific murder - the specifics of the crime and the various potential murderers are all detailed - but can be played multiple times with different outcomes, using the same elements to create different stories. Youdunnit demonstrates many of the principles described in this article, and I will return to it.

In fact, designing a story-making game in which the story created and the gameplay used to create it are equal, which is both fun and creates a satisfying story, requires a synergy between the two forms that is not easy to achieve.

What Makes a Story?

To understand what makes a story-making game work or not, we must first consider what makes a story work. Like a good game, it needs to have a beginning, a middle, and an end. It needs characters who interact and do things, and things that happen to the characters. And it needs a story - a plot that runs from beginning to end and then stops. This is not as simple as it sounds. We all know how to tell stories, though as we grow older many of us forget that we know.

When small children begin to tell their own stories, these may consist only of characters, and this is enough for them. Chicago educationalist Vivian Gussin Paley describes an early story by Mollie, a three-year-old: "Once upon a time there lived a horse and a chicken and a dog. And the next morning there was a robber in the house. That's Frederick. He's the robber. That was scary." Frederick, one of Mollie's friends, has an even shorter story: "Frederick." Paley questions the fact that the story has only one word, but John, a five-year-old, corrects her: "It's not one word. It's one person" (Paley 1986).

Before long, as the teller grows in confidence and familiarity with the art of manipulating narrative elements, their stories acquire a setting and then sequences of events, and finally the coherent unity that identifies what we think of as a story. Within a few months of her first story, Mollie's storytelling skills have mushroomed:

My story is about a little girl who goes to the park to see her mother. And a little boy wild thing comes. Then a monster comes. The girl was bleeding because the monster hit her. So her mother told her she was Fire Star. Fire Star fixed the monster on fire and the monster couldn't get out of the fire. Then the little boy wild thing was friends with all the friends they ate with. They all went to bed and sleeped." (Paley 1986)

What makes these early stages of storytelling interesting is the degree that they resemble the results of some of the more rudimentary story-making games like Story Blocks (1990), the Goosebumps Storytelling Card Game (1996), and The Helpless Doorknob (1989), which all involve laying a sequence of cards or blocks to make a narrative.

Story Blocks are wooden blocks with pictures and brief texts. Lining them up produces a sequence like: "This is a story about Sam / near the magic monster / beside the park playground / and a crawling caterpillar / walking with a white wolf / grinning grandly / and a goofy gorilla / trying to touch a turtle / beneath the blue moon / The End." It's clever, but it's not a story: there is no sequence of structure, events, cause and effect, or conclusion. In its defense, it is designed as a toy rather than a game, and it is meant for young children, but one might suggest that young children deserve better.

The Goosebumps Storytelling Card Game is a deck of fifty-four cards illustrated with characters, places, items and events. The rules say: "On your turn, repeat the story the others have created so far, and add to it with a card from your hand. For example, if the first set of cards creates 'We were in the GRAVEYARD' ... then 'We were in the GRAVEYARD and the PIZZA arrived.' And on your turn 'We were in the GRAVEYARD and the PIZZA arrived - but was covered in RATS!' If you goof, you're out. The last player left is the winner." That's the entirety of the rules, and a sample of the cards. The onus of creating an actual story is left utterly on the players, while the gameplay is nothing more than a memory test. As with Story Blocks, the resulting narrative has no intrinsic structure. It's almost unnecessary to say that it's not an enjoyable game and does not produce satisfactory stories.

What these games create is a basic narrative, not a story. The key element they lack is a mechanic or device to create the structure that would turn the former into the latter. In fact, the only simple sequence-of-cards game I know of that actually works is The Helpless Doorknob, "a shuffled story by Edward Gorey," a set of twenty illustrated cards that can be arranged in any order. The resulting story has no beginning, no ending, and a cast of characters whose motives and actions are inexplicable and unexplained: "Alfred returned from Novaya Zemlya / Adela became disoriented at Alaric's funeral / Alethea vanished from a picnic," and so on. While this narrative has the same structural problems as the "stories" produced by Goosebumps or Story Blocks, it is entirely in keeping with the style and atmosphere of a typical Edward Gorey story, with its themes of strange but unrelated actions hinting at dark, underlying plots, and therefore fits its genre and its purpose perfectly. More recently the card game Gloom (2004) combines Gorey-esque themes with a strategic card-game containing story-making elements. Although amusing, the story is not, however, an intrinsic element of the gameplay.

What does Gorey's shuffled story have that the others don't? Intentionally or not, it hits one of the four cornerstones that underlie any successful game or system that allows players to actively manipulate a story: a clear understanding, encapsulation, and communication of its genre. The other three cornerstones, which I will come to shortly, are structure, rules, and story/game balance.

Genre

The current generation of story-making games do not create a fully-fleshed story, which is one of the reasons they do not lend themselves to computer versions or computer-moderated play. Instead, they provide the pieces of the story's skeleton and the rules for assembling it. The players' interaction with the game builds these pieces into the framework of a story, while the players' imagination and improvisation simultaneously add the flesh of the narrative, bouncing off the prompts and inspirations provided by the game engine.

This has three implications relevant to this piece. First of all, many of the story-making games we are considering depend to an extent on the players' ability and confidence in constructing and narrating stories. This is undeniably a weakness: none of them are necessarily won by the best storyteller, but weaker and less assured players may feel they are at a disadvantage.

Second, the game's mechanics must take into consideration the rules of the genre that it is trying to create: not just the relevant icons and tropes, but the nature of a story from that genre. A fairy tale has a very different structure and set of requirements than a horror story or a soap opera, and a game must work to replicate that. Skilled players can do some of the work, but the nature of games, particularly commercial ones, means that a designer cannot make any assumptions about the people who will play their design.

Third, for a group of players to be able to create the same story, they must all understand the basic rules that underlie what's acceptable within it and what isn't - in other words, the basic rules, tropes, and narrative structure of its genre.

The best demonstration of the strength of genre within story-making games is Pantheon by Robin D. Laws. Pantheon itself is only one of five game scenarios in the book, all of them driven by a system of rules called the Narrative Cage-Match system, which is tweaked for each scenario. Using this system, players create characters and work together to tell a round-robin story about those characters within a preset genre. On her turn each player adds a sentence to the story, which must include her character and can include one other player's character. If a player objects to someone else's sentence, she can go head-to-head over changing it or letting it stand.

At its heart this is a very simple and familiar form of story-making, and Laws has the good sense to play to its proven strengths. What he adds is setup, characters, and a scoring system that doesn't kick in until the story-making part of the game is over.

Each of the five scenarios in Pantheon starts with a detailed genre-specific setup that most games-players will be familiar with: the crew of an undersea base are attacked by a monster; relatives battle for a deceased gazillionaire's estate; post-Tarantino gangsters feud over a safe of stolen cash; giant monsters destroy an Asian city; and the gods create the universe and mankind. These are the jumping-off points for each story, and beyond that the players are on their own, using only their knowledge of the genre's conventions and tropes and the actions of the other players to construct the story. Usually the narrative created is convoluted but complete and entertaining.

Once the story is done, players are scored depending on how many genre tropes they hit. These can initially be counterintuitive: points are usually awarded for surviving the story, but in some cases you can score more by dying in a generically appropriate way - for example, in "Grave and Watery," the player of the lunatic mercenary scores big if his character is killed not by the monster but by another player. Players don't know in advance the specific tropes they're supposed to use, but it's astonishing how many will crop up in any given game, right down to specific phrases. Familiarity with genre is a powerful tool in story-making.

That doesn't mean that a story-making game has to come from a genre that's well known. The Extraordinary Adventures of Baron Munchausen, which I designed, challenges players to tell stories in the style of the late Baron, the eighteenth-century nobleman whose after-dinner descriptions of his singular exploits are still in print today. Not everyone is familiar with these tall tales or the distinctive overblown style in which they are told, and one of my primary concerns when playtesting the game was whether new players would pick it up quickly enough, or at all. My fears were groundless: it seems there's something in the Baron's style, or in the style of larger-than-life stories generally, that is easily communicated or that people understand instinctively. During the game's first playtest, players were joined by a nine-year-old who had missed the explanation of the rules, but who listened intently to our boasting and tall tales. When his turn came, he spun a completely original story, perfectly in-genre and perfectly in-game.

Communicating genre across a group is one thing, but it's another to pass it on to people who have never seen the game in action - typically people who buy a game in a store. Written down, the rules of Baron Munchausen are about five hundred words long and boil down to: "One player is challenged to tell a story on a subject they've never heard before, and must start immediately, while fending off interruptions from the rest of the group." At first sight this appears terrifying, if not nigh-impossible. In fact almost everyone can do it, as long as they're shown in advance that it's possible. But how can a designer show a prospective player not only that the game is playable, but give them the right idiom for playing it?

The solution I devised for Baron Munchausen had a useful side effect: it made it publishable. Nobody was going to pay for a few hundred words of rules, but the Baron Munchausen rulebook is written not by me but by the Baron himself - by charming coincidence, I discovered the long-lost manuscript of the game that my ancestor John Wallis, a games publisher in the late 1700s (true) had commissioned from Baron Munchausen (not quite as true). The Baron takes 14,000 words to explain the rules, with frequent humorous digressions, diversions, excursions to refill his glass, and anecdotes about his adventures. By the time the reader reaches the concise version of the rules on the inside back cover, Or in the current case, the final pages of this volume. they understand the genre, idiom, and how to play it.

The same idea can also benefit conventional face-to-face role-playing games. The notorious complexity of the mix of ideas in R. Sean Borgstrom's "R. Sean Borgstrom" is the same person as Rebecca Borgstrom, whose "Structure and Meaning in Role-Playing Game Design" appears earlier in this section. Nobilis is eased enormously by its 20-page example of play that demonstrates not only a number of the game's core rule mechanics in use, but also several types of character and how they should be played, the style of the game, and the structure and unfolding of a Nobilis adventure.

By contrast, the card game Once Upon a Time has a clear and immediately graspable genre: it is based explicitly on classic western European fairy tales. In the game players tell a story using elements from cards in their hands (for example: a Princess, a Forest, Two People Meet, a Sword, This Animal Can Talk), each Storyteller trying to steer the narrative so that she can play out her hand and win by finishing the story with her "Happy Ever After" card. However, other players can interrupt her and take over the story in midflow. These interruptions are the game's key mechanic: players can either interrupt using an Interrupt card, or if the Storyteller mentions a typical fairy tale trope for which another player holds the card.

When my co-designers and I were playtesting OUaT, we found that though people were creating great stories while playing the game, it was impossible to use some cards to interrupt because no player ever spontaneously introduced the fairy tale tropes on them into their stories. What was strange is that these cards related to what seemed to us to be standard elements from the best-known fairy tales: dwarves, spinning-wheels, apples, frogs, woodsmen, genies, ghosts, and a sausage. Actually I'm not sure how the sausage got in there. But the players seemed to be in a general unconscious agreement that these elements were not part of the genre of fairy tales they wanted to tell. People don't always tell the sort of stories they like to hear, or the type of story they're familiar with.

Structure

Structure is something I stress a lot when talking about game design. It's a catch-all term to describe the underlying mechanics, the combination and interplay of the game's setting, rules, components, and players' thought processes, assumptions, and behavior that cause the game to work the way it does, and sometimes to work at all.

In most games, the structure is simply the way the game is played. In story-making games, it is also the principal way that the narrative shape of the story is formed, whether that means the overall arc of the story to be created, the nature and motivation of the character or characters at its center, the addition of difficulties for the protagonist to overcome, or the way it moves towards an endgame and satisfactory conclusion.

Structure is not the same thing as rules. The Once Upon a Time rulebook is a handful of small pages, none of which describe the game's structure or explain how the story comes together in play. The rules of The Extraordinary Adventures of Baron Munchausen, as noted, boil down to "Each player tells a story while fending off interruptions from the others. Then everyone votes for their favorite story." That's how the game plays. It's not how the game works.

Improvising a story, telling it as fast as it occurs to you, is very not easy. Describing Baron Munchausen makes it sound like being constantly interrupted by the other players would make the job even harder, but in fact the opposite is true. Each interruption becomes a narrative rung that the storyteller can choose to reject or build into their story. The other players may think they're hindering the storyteller, but in fact they're doing a lot of the work of progressing the narrative. They are also introducing an important facet of stories: a succession of new difficulties and dangers that the protagonist - who in Baron Munchausen is also the storyteller, all stories in the game being told in the first person - must overcome to reach his goal and succeed. The game wouldn't work at all without the interruptions: they function on both a gameplay level and a structural narrative level, and they inject humorous and social elements into the game as well.

In many face-to-face games, RPGs in particular, structure comes through density of rules, I treasure the review of Once Upon a Time from Dragon magazine (the house magazine of then-publisher of Advanced Dungeons & Dragons, TSR Inc.), which complained that the game didn't have enough rules (issue 204: April 1994). defining the in-game characters' possible and impossible actions with charts and tables. RPGs are an interesting case, because in theory a game-character's actions are limitless: they can go anywhere and do anything within the bounds of the game world, but that would often let them wander outside the scope of the story.

By defining the character's role within the world and giving them implicit goals - defeat things, gain rewards, advance in experience, power, and status - an RPG's rules make sure that characters and therefore players have a focus that allows the referee to create scenarios for them to play through. In a few RPGs that don't do this, usually the slightly more avant-garde ones in which players can create characters from many backgrounds without any shared goal (I'm thinking particularly of the otherwise superb Over the Edge by Jonathan Tweet), that focus is lost and scenario creation and play is noticeably less simple.

Story structure can also come from the game's setting and the assumptions that people take from a game's components and packaging. If you buy a game called "Kill the Dragon," you assume that there will be a dragon and to win you must kill it, and that is the direction your play will take. In fact there may be better ways to win the game.

Mertwig's Maze (1988) - not a story-making game, though each player's actions do create a narrative of sorts - presents itself as a humorous quest-style board game: players must undertake adventures to the far corners of the kingdom, beat monsters, recruit allies, and find treasures in order to inherit the throne. However, players can win Mertwig's Maze by spending most of the game in the central town, buying equipment and recruiting followers, while never being attacked by monsters, before attacking returning players who have already collected victory tokens, and stealing them. Not very heroic, though it explains a lot about the quality of rulership in fantasy kingdoms.

But very few people play the game like that, partly because the implicit structure angles them away from it, and partly because it's boring. In Mertwig's Maze, as in many games, entering into the "spirit of the game" is more enjoyable than playing to win, and in a story-making game that means the creation of a satisfactory story.

Rules

Early role-playing games derived their structure almost entirely from their rules. The first editions of Dungeons & Dragons explain almost everything about the game's play by describing the rules by which it works - and by implication how the world the game is set in works. Additional background and setting came mostly from the game's artwork.

Focusing on rules rather than structure or genre to provide narrative shape is a perfectly reasonable approach. All stories must follow the rules of their genre and of storytelling in general if they are to satisfy an audience. However, in fiction and verbal storytelling those rules are concealed from that audience, and visible only to the writer or anyone who analyses the story. In a story-making game - as in games in general - the rules are front and center, and every player has to know them.

A story-making game that uses rules to impose structure is the granddaddy of the field, Dark Cults (1983). Based on the stories of horror author H. P. Lovecraft, or more accurately on the atmosphere of stories by H. P. Lovecraft, Dark Cults is a card game in which Life and Death - represented by two teams of players - battle for control of the fate of a single protagonist as he takes a sinister late-night excursion. Each player plays a card in turn, either pushing the protagonist into greater danger or offering him a way to safety, and tells a section of the story as she does so. There are several categories of cards, and each one says what other categories can be played after it. This structures the story, creating situations of impending and deepening peril that end either with the character surviving or meeting a grisly fate.

Dark Cults illustrates one of the perils of using rules to give narrative structure. While it works reasonably well as a game qua game - its strategy is not sophisticated and its tactics depend on the luck of the card-draw, but it is easy to learn, fun to play, and at the end one side has clearly won - it has two problems.

In the game every card played scores a different number of points depending on which team played it. This has to be tracked turn-by-turn, so attention is constantly drawn away from the story to check the card's point value and write it down. This might not be a problem with many genres of story-making, but horror stories depend greatly on the creation and maintenance of an atmosphere of dread. Dark Cults does this with suitably dark card art and a game mechanic that replicates the style of its genre, but the point-scoring constantly breaks its atmosphere. And in a game that finishes when one side has won control of the character's destiny, why then decide the victor on who did better overall?

I've tried to avoid this pitfall in the design of Youdunnit by keeping the rules as stripped-down as possible. The genre of murder mysteries is well understood, so players do not need additional rules to prompt them toward discovering the murderer, and the winner is obviously going to be the person who uncovers the murderer's identity first. There is a bidding mechanism similar to the one in Baron Munchausen (q.v.) involving tokens to determine whether each piece of evidence revealed is true or false, but this is backed up by player rhetoric.

The only counterintuitive piece of Youdunnit's design, where the rules subsume the genre conventions, is that the identity of the murderer is not fixed at the start: the game's skill involves pinning the crime on another player before anyone can pin it on you. This awareness of the game-nature of the activity forces a detachment between the player and the character they are playing in the game (Youdunnit is at heart a role-playing game, though far removed from the Gygax-Arneson design paradigm that has defined the field since the mid-1970s), but it doesn't seem to impede anyone's enjoyment of the game, or their ability to play it to win.

Story and Game Balance

I mentioned before that Dark Cults had two significant problems. The first is with its rules, the second lies in the narrative it creates. Although its mechanics build the atmosphere of a classic horror story, it focuses entirely on what in fiction terms would be the midsection of the tale. The establishment of setting and character are perfunctory, as the character leaves his or her boardinghouse for a late-night walk, and the narrative ends either with the protagonist returning to the same location or meeting a suitably unpleasant doom. There's nothing in the cards, the rules or the structure of the gameplay that encourages the creation of an actual plot, and without plot a horror story is just a succession of scares.

For my money, Dark Cults depends too heavily on rules and not enough on implicit structure: its stories come secondary to its gameplay, and as a result are rarely satisfying in their own right. While it's fun enough to play, the end of the game usually leaves a somewhat hollow feeling, similar to a Hollywood movie that takes an intriguing premise and realizes somewhere in the third act that it doesn't know the answer to the question it posed at the beginning, or a whodunit in which the murderer is never found.

In defense of Dark Cults, I have spent years trying to create a story-making card game based on classic Victorian ghost stories, specifically the work of M. R. James. When this game is done it will be called It Was a Dark and Stormy Night, but as yet it doesn't exist. The structure of a ghost story is much more intricate and delicate than that of, say, a fairy tale or a murder mystery, and to capture it satisfactorily in a set of rules and game components has so far proved impossible. More than once I have come up with a new idea for making the game work, only to realize I have, in fact, replicated Dark Cults' core mechanic in different clothes.

By contrast, Once Upon a Time works well because the fairy tales and fairy tale archetypes on which it is based are universally known and very robust: in the course of one game elements can be introduced, switched around, transformed into a frog, and written out, and you've still produced a satisfactory story with a conclusion and a winner in 10-15 minutes. It works because people understand, not just the tropes of the genre, but the shape of a typical fairy tale. The game mechanics only structure the story in two principal ways: each player's "Happy Ever After" card gives them the ending they must reach; it, and the need to play all the cards in their hand, drives the content of the story they are trying to tell. The rest comes from the knowledge, learned at an early age, of how to assemble discrete elements into a story. I'd be the first to admit that the stories told in a session of Once Upon a Time lack depth or emotional resonance, but for the most part they're coherent and complete.

Any game that deals with the creation or telling of a story has to find a balance between its narrative and its gameplay, and the designer has to choose where the emphasis lies. Story-making games have an added difficulty in that, for the game to work properly, the players have to be prepared to both play the game and tell the story, and give equal weight to both. Once Upon a Time has a bad reputation among games purists, because it can be won by gabbling out a nonsense story in order to play all your cards as quickly as possible. We were aware of this as a flaw in the design but couldn't find a way to prevent players like that from playing like that. It may be a victory, but as with the Mertwig's Maze "shopping" tactic or the first-turn victory in Mornington Crescent, it's not in the spirit of the game.

Once Upon a Time's mechanics are flexible and appear as if they ought to be portable to many other types of story. However, a ghost story told using the same game-engine would be a slapstick affair with nothing resembling the steady build of tension and horror that I'm trying to create with A Dark and Stormy Night. There have been fan-created games using OUaT's game design, but with subjects ranging from comic book superheroes to fetish erotica. Most of them work passably well as games, but not well as engines for creating stories in their chosen genre.

No game can exist without rules, at least implicit ones. Many games have nothing else. But in story-making games, too great a density of rules or a single rule in the wrong place can destroy the cohesion or the effect of the story.

Conclusion

The key to a successful story-making game, at least in the ones that have been released so far, is simplicity of design. That doesn't necessarily mean a lack of complexity in its gameplay or narrative structure, but it does mean that rules have to be integrated with structure and genre to form a coherent package. I am a self-confessed proponent of "elegance through simplicity" in game design, and I realize that this doesn't fit every taste, or every style of game. However, if you're presenting a game in which players have to think simultaneously about gameplay, tactics, and creating a coherent story, then there aren't many mental processor-cycles left. In Once Upon a Time, a player who is interrupted must draw a new card. It's not a complex rule in a game that doesn't have many, but I've found that around one player in four must be constantly prompted to do it. They don't forget, they're just thinking about other things.

There are more complicated story-making games. Tales of the Arabian Nights has a different approach that derives in part from the solo-play adventure game books of the 1980s. Players take the part of Middle Eastern adventurers, exploring a mapboard and encountering all manner of characters, monsters, and hazards. These are determined by a combination of cards, die rolls, and a complex table of matrices, all of which direct the player to the "Book of Tales," which has over 1,300 numbered paragraphs, each one a short episode based on incidents from Middle Eastern folk tales: players are presented with a situation, choose an option, and are rewarded, punished, or thrown down a well as a result.

As the game continues, players mentally build these episodes into a continuing narrative that describes their progress through the game world. That's not the game's primary intention, but while in rules terms it is no different from any board game in which one's turn involves a move, an action, and an effect, the specifically narrative nature of each action, the sheer diversity of them, and their cumulative effect all give the game a unique flavor and add color, depth, and replayability. Tales of the Arabian Nights stands in a subtly different classification: the passive or unintentional story-making game. Other related games include Star Trek: The Adventure Game (1985), City of Chaos (1996), and the lamentably bad Lone Wolf and Cub (1989).

The bulk of its gameplay and its rules, however, are not concerned with the story, which gives it a curious split personality. On one level there is the strategic play concerned with gaining the money and status needed to win, and on another the amusing and often arbitrary story-fragments (brilliantly written and perfectly in-genre) that are clearly at the heart of the game and provide its major enjoyment.

All the games I've described above are competitive rather than collaborative: while the players may be collaborating on the creating the story, they are competing to win the game, either by manipulating their character into the optimum position (Pantheon, Youdunnit) or by controlling the narrative (Once Upon a Time). In the current generation of games the skill needed to create a convincing and enjoyable narrative has little overlap with the skill needed to win the game. Only in Baron Munchausen does the teller of the best story win. It's possible that in the future greater complexity - or more refined elegance - in the design of story-making games will produce games or at least game-engines that combine gameplay and story-making more closely, and which can create stories that carry greater emotional resonance or meaningful content.

Story-making games are not a major genre at the moment but they offer ideas and techniques relevant to in-game narrative that can, with some intelligent application, make beneficial contributions to any game, whether face-to-face or based on computer. Interactive entertainment does not have to mean the viewer should be a passive participant in the act of storytelling. Creating stories is an ability we all have, and constructing an exciting tale can be as exhilarating as the thrill of winning a tightly-fought game. Like their traditional cousins, story-making games have only one winner, but if the gameplay has created a story that is complete, enjoyable, and satisfying, then everybody wins.

References: Literature

Paley, Vivian Gussin (1986). Mollie is Three: Growing Up in School. Chicago: University of Chicago Press.

References: Games

City of Chaos. Colin Thornton and Martyn Oliver; Monocle Games. 1996.

Cluedo. Anthony Pratt; Waddington Games. 1948.

Clue Dungeons & Dragons. Hasbro. 2001.

Dark Cults. Kenneth Rahman; Dark House. 1983.

The Extraordinary Adventures of Baron Munchausen. James Wallis; Hogshead Publishing. 1998.

Fable. Peter Molyneux; Lionhead Studios; Microsoft Game Studios. 2004.

Gloom. Keith Baker; Atlas Games. 2004.

Goosebumps Storytelling Card Game. Waddingtons. 1996.

The Helpless Doorknob. Edward Gorey. 1989.

Knights of the Old Republic. Bioware; Lucasarts. 2003.

Lone Wolf and Cub. Matthew J. Costello; Mayfair Games. 1989.

Mertwig's Maze. Tom Wham; TSR Inc. 1988.

Nobilis. R. Sean Borgstrom; Hogshead Publishing. 2002.

Once Upon a Time. Richard Lambert, Andrew Rilstone, and James Wallis; Atlas Games. 1993, revised edition 1996.

Over the Edge. Jonathan Tweet; Atlas Games. 1992.

Pantheon. Robin D. Laws; Hogshead Publishing. 2000.

Star Trek: The Adventure Game. Greg Costikyan; West End Games. 1985.

Story Blocks. Mary Sinker and Robert Venditto; Rhyme & Reason Toys. 1990.

Tales of the Arabian Nights. Eric Goldberg; West End Games. 1985.

Cite this Essay: