Tools for building a story in classic adventure games.

I am presenting a set of tools for using a classic adventure game model to tell stories. While my focus is the narratively driven genre of adventure games, these tools can help tell a story in any type of game. In traditional storytelling, the the teller of the story controls the pacing and the order how their stories are presented to their audiences. With books and video and other linear media, that control remains solely in the hands of the creator. But in video games, the presentation of narrative can range from strictly linear as in the Dragons Lair games to a series scenes where the player has some degree of freedom to explore and interact with the world and the story being told that are connected to each other in a linear or branching manner, to truly free form games like the Sims series where any narrative that exists is imposed by the minds of the player.

I am going to look at four tools of particular interest. The first, puzzles, provide the obstacles and difficulty that the player overcomes in the process of revealing the story. The second, beat analysis, helps map the emotional flow of the story being told telling. It can also help the author determine if there are any structurally important pieces of the story that are being left out. The third tool is the establishing scene. It is where you introduce both the protagonist and the rules of the world that the protagonist inhabits. The last tool I want to use is Easter eggs. These hidden pieces of the game can provide additional difficulty, help further define the setting, and reward players who enjoy exploring the world in depth, but since they are completely optional content, they do not break the flow of the narrative involuntarily.

Any sort of game can be used to tell a story, but the classic adventure game format is particularly well suited to the endeavor. Adams (2010) defines an adventure game as:

An adventure game is an interactive story about a protagonist character who is played by the player. Storytelling and exploration are essential elements of the game. Puzzle solving and conceptual challenges make up the majority of the gameplay. Combat, economic management, and action challenges are reduced or nonexistent.

In their original incarnation, adventure games were pure text exercises that rapidly spanned a wide array of stories from the classic dungeon delve of Colossal Caverns to the murder mystery of Deadline,to the exploration of multiple viewpoints of an art gallery in Exhilaration, to the literary fictional morality play of Trinity (Montfort 2003). They eventually gained graphics and sounds and a wide array of interfaces and run the gamut from archeological mysteries, to madcap detective stories to musical adventures (Kalata 2011) Any sort of story you wish to tell can be told in the adventure game format, and modern entries to the field such as Heavy Rain (Anderson 2010) can leverage graphics and computing power to tell deeply engaging emotional journeys with adaptive storylines

The closest fictional genre to the standard adventure game model of “find clues and items, figure out how to use them to solve puzzles that the game has presented and progress the narrative” is the mystery or police procedural. A well written mystery gives the reader all of the hints they need to solve the core mystery before it is revealed at the end of the story. The trick for the reader and for the protagonist is to use all of those clues to solve the main puzzle of the story. In Esoterrorists, his first implementation of the tabletop roleplaying system Gumshoe, Robin Laws (2009) says that to tell an interactive story, it is important that the things you need to progress the story should be easy to find and that the challenge should come from figuring out what they mean and how to use them. This insight into tabletop role playing is no less true when we are talking about the sort of storytelling that we do in adventure games.

Tool One: Puzzles

A puzzle, for my purposes, is anything in the game that you must interact with in a particular, non-obvious manner to change its state. A puzzle can take many forms; it can be a riddle, it can be a set of mechanical manipulations needed to change the state of some object in the game, it can be finding a piece of information that is needed to further progress the story. A well designed puzzle is challenging without being so hard that it stalls the progress of the story. The solution of a puzzle should usually reveal most information about the setting or bring the plot of the story a step closer to completion.

Adventure games are full of puzzles and obstacles. They are things that can get in the way of the audience’s ability to follow the narrative if they are not carefully placed and designed, but they provide the challenge that is needed to fully engage a player with the story. It turns out that humans get a great deal of enjoyment out of hard work that is optional, has discernible rules, provides rich feedback on our progress, and seems surmountable (McGonigal 2011). Difficulty is good, so long as we perceive it to be fair.

When we overcome a difficult task, it generates a feeling of pride, a chemical and emotional rush generated by victory that game designers use an Italian loan word “Fiero” for (McGonigal 2011). But if that task lacks feedback to let us know that we are approaching it in non-futile manner, or if the solution has no bearing on the way we understand the world to work, instead of feeling rising tension followed by the emotional release of victory, we are likely to just feel frustration. Instead of victory, we break our engagement with the story, and unlike a living storyteller, there a game can not read our audience and recapture their interest.

Early adventure games were loaded with puzzles with dream logic solutions (Montfort 2004), with situations that were unnavigable without repeatedly trying a comprehensive set of actions, attempting to use each item in your inventory in each situation, at every possible time, often with fatal or game derailing results. This sort of difficulty doesn’t generate Fiero in most people. Instead it convinces them to go do something else.

It wasn’t until Brian Moriarty’s Loom that a major adventure game company adopted a design philosophy that was intended to circumvent that sort of frustration. Lucas Arts started producing adventure games where your character could never hit an unrecoverable dead end, and character death was either impossible or uncommon (Kalata 2011). Their games still provided challenge, but the challenge resided not in nonsensical puzzles and unexpected deaths, but in figuring out how the parts of the story fit together and how to use the clues the player had collected to solve them. This model of difficulty never tore them out of the story, and it minimized frustration episodes without robbing players of their sense of achievement.

Every unavoidable puzzle in a game should serve a narrative purpose. The act of searching for the solution to a puzzles should move the story forward to its eventual outcome, and solving a puzzle should usually provide moment of positive emotion in the flow of the story. Difficult puzzles should seldom be placed so that they create a bottleneck in the story (Laws 2009). The most difficult puzzles in the game need to be used carefully so that they help progress the narrative instead of halting it.

Tool Two: Beat Analysis

The second thing I want to look at is a tool that lets us map the emotional results of each scene in our games as well as providing a way of checking to see if all of the structural bits we need to tell a coherent story are in place. In “Hamlet’s Hitpoints”, Laws (2010) establishes a model for the analysis of stories. He takes an actor’s trick, analyzing the beats of a script to lay out the emotional landscape of a piece and uses it to map out potential flows of game-play. A beat analysis can be done at different levels of focus, from the individual lines of a story to the larger divisions such as scenes or chapters. Each beat is composed of two parts, the category of beat, and the emotional direction the resolution of the beat moves the audience in. Laws chooses hope and fear as his emotional poles, though for specific needs other emotions could easily be used.

The categories of beat that Laws identifies are:

Procedural

A procedural beat plays on the character’s practical, external goals. Procedural beats tend to be defined by action, with physical dangers taking a prominent role., and verbal conflicts are resolved through external results instead of emotional states. Procedural beats, along with dramatic beats make up the majority of beats in most fiction, and most beats of these types .

Dramatic

A dramatic beat works with the character’s inner goals. They involve negotiation and concessions between characters. Conflict in a dramatic beat fulfills a character’s emotional needs. They are, almost always, dialogue beats

Commentary

A commentary beat takes the focus away from the protagonist’s pursuit of their goals and reinforces the story’s themes. This is often done by a third party narrator or as an aside by a secondary character.

Anticipation

An anticipation beat creates the expectation of a upcoming procedural success. It is often a preparatory scene without any threat to the focus character. Anticipation beats allow us to enjoy some of the thrill of an expected victory before it actually occurs.

Gratification

A gratification beat is a positive emotional moment that is unrelated to the main plot of the story. It is often an in joke or musical interlude. A gratification beat must be used with care because they can very easily break immersion in the story.

Bringdown

A bringdown beat helps set the mood by providing a negative emotional movement that is unrelated to the main story. While unrelated to the narrative, they should be used to mirror the narrative’s emotional direction. A bringdown beat should hit the protagonist (and thus the audience) while they are down.

Pipe

A pipe beat presents information to the audience that will be needed later without making it obvious that the information is going to be important. Pipe beats can make reveal beats much stronger, and as we will see later, can play an important role in setting the audience’s expectations of the logic of the world and are an important part of writing an honest mystery. This sort of beat is placing the gun on the mantle place.

Question

A question beat poses a question about things that have already happened that the audience is unaware of. This is separate from a procedural beat that introduces a question about what is going to happen. These beats generally move the emotional state of the audience downward because they increase uncertainty, the lack of information presents an obstacle to the protagonist and engages the audience. Question beats are an essential part of establishing a mystery.

Reveal

A reveal beat answers a question that the audience has been asking, either as a direct result of an earlier question beat or because one or more pipe beats have created a subconscious question in the audience’s mind. Often the release of tension from obtaining an answer causes a reveal beat to end on an uplifting note, though if the reveal shows us a larger danger or greater stakes for the protagonist, the beat can certainly lead to a down note.

In a traditional game with a mostly linear narrative, a beat analysis of the script can give us important insight to what is happening in the story. A string of down beats needs the occasional upbeat or the player is likely to be swamped in the emotional morass created by a constant downward slope. On the other hand, if we see a string of up beats without any down beats, there is likely to be no tension, no stress, and thus no engagement with the story being told. If a game has a dozen question beats, then you know that you need to reply to each one with a reveal at some point, and if you intend for a character to fire the gun on the mantle, it is essential that there is a pipe beat earlier in the game that shows the gun is on the mantle.

It is, of course, possible to create an interesting and engaging narrative while breaking the rules, but you should know what you are doing and why. One of the easiest mistakes for a person working in a creative field to make is to see the amazing work of the previous generation who does spectacular work by breaking the standards and rules of the format they are working in. The problem is that those prior masters first learned the rules they broke before breaking them broke them. That is why we have a thousand emotionally dead imitations of Van Gogh, Dali, and Pollock.

There are some interesting things that can be done applying Laws’s beat analysis to the construction of a less linear digital narrative. If your story is being presented in discrete chunks, then instead of analyzing the compiled narrative, it is possible to analyze each of the pieces of story. Each piece of each story has a type and a direction, and given the tools provided by working in software, you can do some interesting tricks with that. If you write each scene as a beat, and you know which beats are upward beats and which are downward, then you can use that information to influence the flow of the game.

If the player experiences a string of mostly downward procedural beats due to their choices, then your program can interject a positive gratification or commentary beat that is tied to the set of beats they have already chosen. If a player has a series of beats all moving in the same direction, you can use that to choose and modify music selections. As things go well for the player, the music can become broader, more inspirational. As they descend the emotional scale, the music can become darker, more discordant, helping to ratchet up the stress. The music then provides a sort of subliminal feedback, helping the players become more immersed in the story, but also encouraging them to make choices that cause the music to match the moods they wish to experience in game play.

A beat analysis of the story components of your game can also provide an important check. If a particular beat is a pipe beat or a question beat, you know that, no matter what sort of story you are telling, you need to make sure to close that beat off with a reveal or possibly a commentary. If your game has a place where there is a continual emotional downturn, beat analysis will show you where to insert an anticipation beat to mark the eventual upturn.

Tool Three: Establishing Scenes

The third tool I want to employ is the establishing scene. Early in the game, while the emotional stakes are relatively low and the world that is presented is still not overly complex, the player should get to see the sorts of solutions that are going to be prominent in the game. Just as the introduction is used to show the audience who the protagonist is through his or her early actions, a game should introduce the rules of the world. the logical foundation of all of the solutions the player is going to have to discover throughout their time playing in your world. This way, when the player finds themselves facing ever increasing challenges, they have a toolkit of previous expectations to draw from. If your game breaks the logic and the tone set in the establishing scene, the likely result is that the players will be frustrated and lose their sense of immersion in the narrative. Many adventure games fall prey to this problem, be it due to puzzles that don’t fit the tone of the game as in the Laura Bow mysteries, or wild shifts in tone, like the whiplash between grim mob violence and slapstick comedy in Runaway (Kalata 2011).

There is great storytelling power in worlds that follow their own logic. There is a reason that Lewis Carroll’s Wonderland remains a favorite book for many people into adulthood. It paints a consistent but strange world, a place where the logic of the ridiculous holds sway. While hitting a player with random and nonsensical situations can break immersion, if you are careful to establish the rules through early interactions, you can bring them delight and wonder in a story based on rules that would be nothing but frustrating had they not been presented in smaller doses early on in the game (Plotkin 2010). Just like Alice’s introduction to Wonderland establishes how the world works in smaller safer interactions before she ever has to deal with the stranger and more dangerous denizens of Wonderland, if you wish your game to hinge on strange logics, establish them early on.

Sam and Max Hits the Road establishes the sort of solution that “makes sense” in their setting through a series of option-limited early scenes where the title characters do things like throw a bomb out into the street because “nobody we know is down there” or reach down the throat of a cat to retrieve an important piece of paper (Kalata 2011). In both these cases, the game guides the player into making strange decisions by limiting the choices they have, both for actions and for places they can go and explore. there is never a moment of analysis paralysis because the player is not confronted with too much weirdness at once. Those two procedural beats act as pipe beats as well, presaging the sort of strangeness that will be required of the player in later sections of the game.

Tool Four: Easter Eggs

. While I have said that the narrative of a game should not be hindered by puzzles and implausible riddles, that does not mean that you shouldn’t include difficult puzzles and hidden riddles in your games. Instead, it means that those things should never halt the progression of the game. Moriarty (2002) explores the seductive nature of Easter eggs or hidden secrets. Finding secrets that most people don’t know is a profound source of pleasure for many people, and solving a wickedly difficult, optional puzzle is a huge source of fiero.

By putting your hardest puzzles into a game in places that they do not have to solve them in order to complete the scenario, by making their solutions optional and the act of finding them, itself, an emotionally positive beat, you create a chance for players to feel like they are achieving great things while minimizing the frustration of being forced to break your immersion in the narrative while you beat your head against a unobvious puzzle. We are creating unnecessary hard work that is completely optional, part of McGonigal’s (2011) prescription for a fun game.

Easter eggs are also a good place to put chunks of world building and background information that would break the flow of the narrative, even had they been put in a commentary beat. The rpg Baldur’s Gate did this. Scattered throughout the entire setting there were hundreds of books, each of which could be read by the player. Each book contained setting information, histories and biographies and discussions of the nature of reality in the setting. Each time you found a new book, a little bit more of the back story of the game was revealed, but these pieces were completely optional. You could collect and read all of them, but if you did, it was on your own time, by your own will, not because the game required that you know any of the information scattered throughout the books. Being hidden and optional made the books a source of fun instead of the frustration they could have been had they held the solutions to otherwise insurmountable problems.

We have puzzles to challenge our audience and to manage the pacing of our game. We have beat analysis to map out the emotional journey of our game from scene to scene. We have our establishing scenes to help our players engage with the world and to let them know what to expect. We have Easter eggs to provide additional challenge and a sense of reward for engaging with the game world. We can bring all of these tools with us and hopefully use them to tell better stories.

These tools are not the only things are needed to tell a good story in an adventure game, and they certainly aren’t the only route to constructing an interesting and fulfilling story in a game format. Using them won’t guarantee that your story will be interesting, engaging, or challenging. But, if you keep them in mind while you work on your game’s story, they can help you avoid common pitfalls of game design. You can use them to manage difficulty and the emotional flow of your game, and thus to construct a more satisfying narrative. Remember that our goals are the same as those of every storyteller since the dawn of humanity. We want to tell a good story that pulls our audience out of their daily concerns, that informs, entertains, and engages them.

References
Adams, E. (2010). Fundamentals of Game Design, 2nd ed. Berkley, CA: New Riders
Anderson, L. (2010) Heavy Rain Review. Retrieved from

http://www.gamespot.com/heavy-rain/reviews/heavy-rain-review-6251617/
Kalata, K. ed. (2011). The Guide to Classic Graphic Adventures. North Charleston, SC: Self

Published, Printed by CreateSpace.
Laws, R (2009). The Essoterrorists. London, UK Pelgrane Press Ltd.
Laws, R. (2010). Hamlet’s Hitpoints Roseville, MN: Gameplaywrite Press
McGonigal, J. (2011) Reality is Broken New York, NY: Penguin Press
Montfort, N. (2004). Twisty Little Passages. Cambridge, MA: The Mit Press
Moriarty, B. (2002). The Secret of Psalm 46. Lecture Transcript. Retrieved from

http://www.ludix.com/moriarty/psalm46.html
Plotkin, A (2010). A Writer’s Guide to Interactive Fiction. Lecture Notes. Retrieved from
http://www.eblong.com/zarf/essays/if-for-writers.html

Initial thoughts on procedural generation of narrative:
What we have to be creating here isn’t the epic story arcs, but the daily stories individual lives. The ebb and flow of a community of software entities that interact with the players. Now given a sufficiently flexible system, that could itself generate a wider story arc with the sums of the actions of all of the entities yielding a final story, but for the time being I’d rather set my sights on something with a little narrower scope.
So each “character” in our story is going to have a bundle of wants and needs, each of which are acted upon only when they reach a certain threshold. Interactions with other characters will modify those wants and needs in both positive and negative manners. The characters will need to be able to interact with each other as well as with the user’s character, and each of these interactions will lead to differing responses based on the needs and wants of a given character.

 

For example, Given the characters Alice, Bob, and Carol.

Bob currently wants two things at sub threshold levels. He wants a pet cat and he wants Alice to like him. \

Alice has her own wants, but she really hates cats.

Carol has a pet cat.

Whenever Alice sees Carol’s cat, she becomes upset.
When Bob interacts with Alice, his desire to make her happy increases.

If Bob sees Carol’s cat, his desire to have a cat increases.

But if Bob sees Alice disliking Carol’s cat, and his desire to be friends with Alice is stronger than his desire to have a cat, he will further lower the priority of cat ownership.

On the other hand, if bob’s desire to own a cat has increased past his desire to be friends with Alice, finding out that Alice dislikes cats will further decrease his desire to be friends with Alice.

This tension can be used to develop interactions with the player.

Every character will have a pool of dialogue that is ranked in accordance with how it interacts with the wants and needs of the given character. If Bob’s desire to own a cat is high, he will be more likely to use cat related dialogue options, and he may well be willing to exchange things he has with the player for things that lead toward his eventual ownership of a cat.

On the other hand, if his desire to own a cat is lower than his desire to be Alice’s friend, he might ask the player to intercede with her on his behalf.

This could lead to a situation where the player helps Bob come to terms with Alice’s dislike for cats. They might also increase Alice’s friendlyness toward bob, possibly even taking actions to mitigate her dislike for cats, which would allow Bob to get all of the things he wants, and allow Alice to become better friends with both Bob and Carol (who really likes her cat, so is a poor friendship prospect for Alice.