choice design · Detectives · Narrative Design

Choose Your Own Mystery: Who Killed Harlowe Thrombey?

The growing interest in choice-based stories in games and television, as well as my own interest in the perennial popularity of mystery and detective stories led me to finally read Who Killed Harlowe Thrombey? (1981) by Edward Packard, an early title (#9) on the original Choose Your Own Adventure books.

How do you make a choice-based whodunit interesting? One of the things that keeps us reading or watching a whodunit is that, after the crime is discovered at the beginning, the reader / audience is given snippets of information, which they try to piece together to be able to anticipate the solution to the crime before they reach the end of the story. Reportedly, this is what led Agatha Christie to write her first mystery novels, challenged by her sister to write a story that the readers could solve before reading the solution. But how does giving the reader choices change the challenge? (Partial spoilers ensue.)

Turns out that Packard was actually pretty smart when writing his book. The mystery itself is a paint-by-numbers mystery (millionaire gets poisoned in countryside house, and all the suspects are people who could benefit from his death), and the reader can choose different places to go to, different people to talk to, or different stories to pursue, as the blurb of the book describes. More interestingly, at a certain point the reader can actually give up on the investigation, thus forfeiting the chance to learn the solution to the case.

My battered second-hand copy of Who Killed Harlowe Thrombey?

And that’s what makes the book compelling–a whodunit in the end tells the reader / audience what the solution to the mystery is, which can be satisfying even if one did not get anywhere near figuring out the solution. But here the reader has to pursue the solution, and in order to get all the details of how the murder happened has to actually re-read and choose different paths in order to get all the information that they’d traditionally get by just continuing to read or watch. At a certain point, the reader is told “if you’re sure you’ve solved the murder, go to page 122”, which would seem to be a shortcut for them to learn the solution even if they don’t really know it. But it is really a red-herring – the paragraph just tells them that they’re calling the police inspector Prufrock and say they have the solution, without providing any additional information or solving the case. The reader must actively pursue all the pieces of information.

For my first read-through, I figured out who one of the culprits was (there’s actually more than one) and then the protagonist was immediately killed, which was not very satisfying as a reader, but it was a very clear message telling me to try again. And this is one of the achievements of the book – it does encourage re-reading in ways that choice-based games don’t often encourage – the recent Overboard being an exception – at times because of their length, at times because there is not a sense that what the player may be missing is that interesting at all.

The book also manages to make some of the choices very dramatic – one of the early choices involves joining the victim to the dinner where he will be killed. This allows the reader to witness part of the events directly, including hearing Harlowe Thrombey’s last words. In one of the branches, one of the culprits misrepresents those last words, which makes the reader realize they are lying – but the reader has no agency to act upon that information, it is a revelation for us instead of the detective protagonist. One of the options leads the protagonist to say he has a recording of Thrombey’s last words, which feels wrong if the reader has actually read the passage with Thrombey dying on the protagonist’s arms. This is also not a choice – the protagonist does it for us. But it leads to a satisfying ending – the culprits reveal themselves and get arrested, while the reader learns that the recording was a bluff. While the reader does not have agency in this resolution, the gap between the reader and the protagonist gives way to a compelling dramatic moment.

The book thrives on this gap – while the protagonist keeps trying to solve the case in each potential branch, the thrill for the reader is to collect and contrast the information with every read-through, carrying out an investigation of sorts while exploring the different choices in the story. The detective work takes place at the meta-level, outside of the story, which turns what is otherwise a trite murder mystery low on characterization into and interesting and charming read. The book creates a loop for the reader, in consonance with the fashionable mysteries on a loop that we have seen in traditional fiction as well as games, but in a work written almost four decades ago.

choice design · Narrative Design · Reviews

Bandersnatch, or why Choosing Your Own Cereal is important

Bandersnatch, the first interactive episode of Black Mirror, was recently back in the news because of the lawsuit launched by the Choose Your Own Adventure publishing company has been settled, which has made me realize I didn’t get around to write about the episode yet. As anyone who works in narrative games / interactive narrative, of course I have opinions.

Hearing some of my colleagues in games and narrative design has been interesting, particularly those who made angry pushbacks and talked about it with derision. Much has been made of the first choice of the story, which lets the audience choose what kind of cereal they want to protagonist to have for breakfast, for instance. However, we should not dismiss the cultural impact that Bandersnatch has had on general audiences–thanks to Netflix, many people have learned the title of an actual unreleased game from the 1980s, plus it has made choice-based narratives accessible to others in ways that no other interactive television, let alone videogame, has before. Bandersnatch is very accessible–users do not have to install a new app or learn a new interface by making it available through one’s streaming subscription. Many videogame designers wish their work was as accessible and popular as this piece of interactive television.

The notorious “choose the cereal” moment is actually there for a very particular reason: while seasoned videogame players are used to making choices in games, much of the Netflix audience is not. Anyone who’s shown games at venues where you come across non-gamers knows how intimidating interacting with a game can be for them. Something as innocuous as choosing a cereal and seeing how it changes the advertisement on television helps putting users at ease, so they learn how the interaction of the story works. They make a choice, and soon after they see a consequence, and see that they have not broken anything. It makes them comfortable with interaction, which is good for a story that goes quite bonkers later on.

And before I go on explaining what I like about Bandersnatch, let me state that not every choice has to change the plot or not be trivial, because anyone who’s juggled branching narratives knows how much of our time is spent creating the illusion of choice or giving players small expressive choices so they feel they’re in control when they really not. Rant over.

The episode tells the story of a young man who is working on what he intends to be the ultimate choose-your-own-adventure game for the Spectrum 48k, one of the most popular computers in Europe in the mid-80s. The protagonist is marked by the trauma of losing his mother in a tragic accident, and is trying to gain control of his own life by working on this game.

Bandersnatch was made for me: I love Black Mirror and its frequently cynical approach to humanity (although current events make the show feel less science fiction and more speculative drama). The world of the episode is quite familiar–I work in videogames and mostly hang out with indie developers; on top of that, I have taught a class on European videogames of the 80s and got to learn a lot about the bedroom developers in the UK like the protagonist, as well as the story of the original Bandersnatch. Even without the interaction, the story resonated with me (yeah, indie game development is a challenge to one’s mental health), and I had a blast by figuring out all the easter eggs and references.

The interactive story is very meta: it’s a choice-based story about someone working on a choice-based story. The self-referential aspects help smooth over one of the main challenges of designing choice-based narratives–each choice creates an alternate timeline, which makes it easy to fall into inconsistencies and complicates writing a branching story (at least when branching is understood in an oversimplified way). The editing of the episode is really smart, reusing shots with different sounds or dialogue lines spoken off-camera, as a way to avoid having to shoot large segments of different content at the same time it shows how there are different timelines and realities. To some the episode seemed revolutionary–in the past, branching narratives have been either poorly designed because they are written by screenwriters with a poor understanding of interactivity, or by game designers who have a trite understanding of storytelling. Charlie Brooker is the rare person who seems to understand games, and also happens to be a pretty decent screenwriter, an unfortunately rare combination of skills.

The self-reflective nature of Bandersnatch turns the story into a gimmick–our decisions as an audience/user create alternative timelines that some of the characters seem to be aware of; in certain scenes, the protagonist behaves as if he’s aware that someone is forcing him to do things he doesn’t want to do, turning us into the incarnation of his mental problems. But this is only true in certain threads of the story, while others turn the story into a conspiracy, while some others tell us a more intimate and ultimately tragic story. But the triumph of the story is bringing all of them together–this is a story about ontological instability, about worlds that have multiple versions with characters that are somewhat aware of that multiplicity. The huge poster of Philip K. Dick’s Ubik in one of the character’s living rooms is a pretty big and unsubtle giveaway. On the other hand, the metafictional gimmick also makes it really difficult to reproduce the success of Bandersnatch–it’s a one-trick pony that can get old fast, and not all choice-based narratives need to comment on the unstable worlds that they create.

(Excuse me if I don’t deign to discuss the “breaking the fourth wall”, because I’m tired the belief that any narrative should make “immersion” prevail, as if there is only one way to tell stories. Only one rant per day.)

Thus, the choice design of Bandersnatch is not about branching nor about agency or control over the story, but exploring the possibilities of the world, focusing on the “what if” and surprising the audience rather than making us feel like the story is at our service. The point where you can tell the protagonist about Netflix, or choose to go on a rampage are part of the fun of the story, which goes in many different directions and story genres–from personal drama to horrible murder to action movie to nostalgia trip.

Yes, choice-based narrative videogames are way ahead of this–see the wonderfully sophisticated works that Failbetter Games or Inkle Studios have been producing over the years, for example. But even with the smartest narrative designers out there, their fantastic works are still intimidating to a large part of the audience because they’re perceived as “videogames”. These are players and users who would be head over heels playing these games, but they seem too complicated to them. For example, I had someone who loved playing Her Story at a film festival ask me if he could download it through the streaming service Kanopy. The technological barrier is overwhelming to many.

There are things that Bandersnatch could do better, certainly. The aspect of social commentary is not as solid or resonant as other episodes, for example, probably because it’s more of a trippy period piece than science fiction. Since the story is about itself and the choices, there’s little room for anything else. There are issues with gender representation–the women in the episode are mostly wives/mothers or helpers and have a supporting role, for example. Mental health issues created by trauma are associated with psychopathy and murder, which is something that we storytellers should be better at–although it is part of the storyline exploration of the episode, the murder storytline reproduces stereotypes that do not need reinforcement. Although my particular pet peeve of the story is how the protagonist is creating a game for the ZX Spectrum in BASIC, and there’s no way that a game of that scope, with graphics and all, would fit in the paltry 48K of RAM that computer had. But while there are issues of representation that could be better addressed, we also have a few lessons to learn about how to engage a wider audience that may recoil at the mention of videogames, while they may probably love playing choice-based narrative games. And at times all it takes is starting by letting them choose their own cereal.

Narrative Design · News and Events

The Narrative Innovation Showcase 2020

For this year’s GDC, Matthew Weise and I had prepared the 5th consecutive Narrative Innovation Showcase at the Game Developers Conference (GDC)  in San Francisco. It was going to be the opening session of the  the Narrative Summit on Monday March 16th. Since GDC was cancelled as one of the early casualties of the pandemic, we didn’t get our wonderful set of presenters to show their work in California, but we would still like to highlight the showcase we had prepared.  Here we explain why we selected the games for the showcase, in a year where there have been plenty of games made by very talented narrative designers. Our panelists are also putting together videos of what would have been their presentations, which will be made available in GDCs YouTube channel. 

Astrologaster 

Katharine Neil for NyamYam.

Game Website

Astrologaster is a particularly apt game to play these days, since it takes place during the plague in London in 1592. We are Doctor Simon Forman, an astrologist / astronomis / physician who can cure any illness and solve any problem by looking at the stars – he’s a true renaissance man. By solving the plights of our customers we can gain reputation and finally achieve the proper doctor’s license that Doctor Forman doesn’t really have. The historical setting is unusual in games – though not the only game set during a plague released in the last year. There are many things that we love of this game that we find truly refreshing – from its Renaissance songs, to the bawdy themes. This is also one of the rare cases where we have a writer who is on top of her craft, the care in the language in the dialogue as well as the lyrics of the songs is a joy to see in videogames. 

Before I Forget

Chella Ramanan & Claire Morley, 3-fold Games

Before I Forget is a poetic first-person experience attempting to capture what it is like to live with dementia. It would be unfair to peg it as a mental health awareness game because, though it clearly serves that function, it uses its thoroughly researched mechanics design to push first-person storytelling in new directions. The use of shifting color to signify something half-remembered, the way everyday objects provide fragments of memories that can suddenly cascade into full-on recall, or the way space folds in on itself to capture the experience of a memory just out of reach: these are at once fresh techniques for creating evocative spatial stories and also beautiful ways to drive the subject matter home. We love how Before I Forget portrays dementia’s effects, be they good or bad, with nuance and complexity by pushing the medium forward.

Cris Tales

Carlos Rocha, Dreams Uncorporated

Game Demo

 

Cris Tales is a great example of a group of talented folks taking a familiar genre to new personal, political, and mechanical heights. It is a fantasy RPG where the player has the ability to simultaneously experience past, present, and future at every moment of the game, allowing you to see the impact of actions across time instantly. Yet rather than just being a fun gimmick for another general save-the-world story, this core conceit is used to explore the cause and effect of systemic environmental collapse, the ensuing civic crisis it causes, and how that crisis exacerbates existing class inequalities in a fantasy world based heavily on Colombian folklore, architecture, politics, and economics. What we love  about Cris Tales because it is the very essence of artistically engaged genre work, using familiar forms to say something about real life through metaphor, via a fresh core mechanic that is impossible to divorce from its storytelling function.

Tick Tock: A Tale for Two 

Mira Dorthé & Tanja Tankred, Other Tales Interactive

The way Tick Tock: A Tale for Two pushes the art of narrative design forward is at once brilliant and simple. To a certain extent, it is a conventional adventure game, where a pair of co-op players must work together to solve the unraveling, multi-generational riddle of a mysterious dynasty of clock-makers. The two find clues, solve puzzles, all while piecing together fragments of the story. The difference is both of these players are on a different screen and have access to different information, causing in-person communication to emerge as the secret core mechanic. Players have to talk to each other – discuss the story, clues, context, theories – to solve the ever-expanding mystery and complete the story. Tick Tock: A Take for Two elegantly to re-introduces the communal in-person experience to the narrative adventure game format, making it feel old and new, simple and complex, all at once.

Mutazione

Hannah Nicklin, Die Gute Fabrik

Game Website

Explaining the charm of Mutazione is not easy – it is best to play it and see for oneself. We do have a soft spot for adventure games – we have featured a lot of them in our showcase over the years. The island of Mutazione is populated by quirky and endearing mutants—everybody has secrets, hopes and traumas, which we can discover as we explore the different locations. The game lets us forge emotional connections with the characters because everybody feels real; there are no stereotypes or cardboard characters here. Its developers define it as a soap-opera, because its cast has a life that is independent from us, they have whole lives and relationships on their own – they’re not there just for the player. The capacity of the game to make us believe that it is a living environment is unusual, and we would love to see more games that explore the emotional connections between their NPCs.

(Full disclosure: I worked on the Spanish version of this game.)

You can watch previous editions of the Narrative Innovation Showcase here: 2016 (YouTube), 2017 (GDC Vault), 2018 (GDC Vault), 2019 (YouTube).

choice design · Interactive Fiction · Narrative Design · Resources

Taxonomy of Narrative Choices

A few years ago, I realized that in order to teach narrative choice design I needed a classification of types of choices. Creating games in the style of choose-your-own adventure, in the style of Choice of Games, Inkle Studios, Failbetter Games, the sadly defunct Telltale Games, or for interactive film, requires having a vocabulary that refers to the different ways in which choices can be expressive.

I came up with my own taxonomy, which I have been teaching in different iterations over the years. Inspired by works like Understanding Comics by Scott McCloud, I finally realized it’d be best to illustrate the taxonomy interactively, on top of discussing examples in the class. You can read/play the taxonomy on my itch.io page.  (Update 16-Oct-19: Spanish version of the interactive taxonomy.)

This taxonomy is part of a larger lecture, which is the introduction to narrative choice design in my classes. While games may be a series of interesting choices, they also become much more complicated when we make those decisions be part of a longer narrative. One of the reasons why making choices can be compelling is that players can see the consequences of their decisions—that is the foundation of agency, one of the pleasures of digital media as defined by Janet Murray. We like feeling that the game is listening to us, that we are in control of our actions in the game, that our choices matter.

Challenging agency, however, also has a lot of expressive possibilities. First of all, players can have the illusion of agency without having to change the content of the game. If you’ve played any of the choice-based Telltale games, you have probably seen the caption “So-and-so will remember that” after making a decision. Most of the time, it turns out they won’t—but the player can believe that their action had an effect on someone. It is a way to use the player’s perception and imagination in order to fill the gaps and not having to expand on the content.

Frustrating the player, taking away agency, can also be expressive. Games have the annoying tendency to become “entitlement simulators” ; game designers should challenge players to make difficult, painful decisions, have them realize that the world may not revolve around them, and that there may be problems that are beyond their control. Untethered agency can be problematic in dating sims, for example—boiling down intimate relationships to choosing the right things to say oversimplifies how humans connect with each other. Representing love as having agency over a person, or being able to “win” them by achieving a score, can reinforce certain unhealthy ideas about relationships, and push players down the brink of toxicity

The taxonomy I propose is therefore a breakdown of the different ways in which undermining agency can be expressive, and can help players think about what they do. (Maybe.)

combinatorics explosion.001

Using different types of choices is also a healthy way to keep projects under scope. As the image shows, if for every decision the player makes the story bifurcates, the content multiplies very fast. In order to write a choose-your-own-adventure story where each version is only four pages long, and only gives two choices to the player on each page, we  would have to write fifteen (15) pages. It is not very efficient and the combinatorial explosion goes out of control very fast. There are also models of different structures that both provide players with interesting choices without having to generate inordinate amounts of content—Sam Kabo Ashwell provides one of my favourite classifications on his blog.

Last but not least, a lot of what went into the taxonomy and my original lecture comes from the lessons I learned from working  in my classes with ChoiceScript by Choice of games (check out their game design blog posts if you’re interested in the topic).