Piranesi and Mysteries in Created Worlds

There is something delicious about a mystery. I won't say much about this but that it continually presents the reader with questions to ponder. The narrative is generous and consistent enough that these questions can yield significant answers. The tone of the book is contemplative and yet also somehow effusive and I can't recommend highly enough the audiobook narrated by Chiwetel Ejiofor. 

The rest of this post contains light spoilers, so go and listen to this short novel before reading on.

  

Piranesi's title character inhabits a house that evokes the 18th century Italian Giovanni Battista Piranesi's series of Imaginary Prisons (pictured above). Susanna Clarke slowly feeds the reader information that is remarkably clear and consistent, allowing her audience to gradually move from guessing at mysteries to a series of revelations. 

At one point in the story Clarke references the wisdom of the ancients. The idea is that certain people operate using a kind of pre-rational thought and see the world differently. Perhaps they see objects as Platonic messengers, bearing meaning beyond their individual properties? The idea is explored further and presents some fascinating prospects for imagining different types of people and modes of thinking.

This works in part because of the authority of the writer. Clarke can show us the inside of a character's mind and because these thoughts are written down the reader can return to ponder them. It is possible to present intricate mysteries and revelations because of the the precision of the medium.

In roleplaying games things are different. GMs are limited because they must present information in the moment. Prep is helpful, but it impossible to carefully select information with the same care as an author. Read-aloud narration feels artificial in a game based around conversation and it is not responsive to player actions.

Playing a game is different than reading, too. Players receive information through auditory channels and they cannot easily return to anything they haven't written down in their notes. Most players don't take notes. Even when they identify important ideas, the words they choose to write down are an imperfect reflection of the information the GM intends to portray.

What do characters even know? GMs can't be certain what players know and players due to imperfect communication and due to limited cognition and memory. Ultimately, even players receive only an imperfect representation of what their characters know because it is filtered by what the players themselves know. They cannot see the world. An author can omnisciently state a characters thoughts, but GMs typically only go so far as to present impressions or verify players' questions. To invade a player-character's mind any further risks robbing the player of the chance to make conclusions on their own.

In order to tackle these potential issues, tabletop designs have tended to branch into one of two directions. The first is to grant some degree of narrative control to the players, allowing them to add or invent details to create a more satisfying story and to overcome some of the difficulties of imperfect communication. These kinds of games border on a completely separate genre from roleplaying depending on how much control the players are granted. The games may entail greater or lesser challenge and mechanical complexity, and often can yield suprising and creative outcomes based on player input. The issue for me is that my entire concept of mystery relies upon the fact there is an objective truth to be discovered. If I can create my own 'truth' then the discovery is robbed of value. 

The other method RPGs use to add mystery is to add detail. Clarke's worldbuilding is masterful in part because she embraces the richness afforded by allusion to the real world rather than trying to invent all her setting details from whole cloth. This quote from Alexis at Tao of D&D gives a good example of what I mean:

Let's begin with the basics.  Is your game world a sphere, or like Earth, "sphere-like?"  I ask because if it is, we've got to address the physical laws inherent in large spherical bodies with atmospheres, located — presumedly — from one or more central suns in your world's solar system.  Now, I understand that may seem completely irrelevant.  Yet we have to start from one or two perspectives: (a) either the game world makes sense, in every respective way that human beings on a planet understand, from the ONLY example they have of how a world works; (b) or the world doesn't make sense, which discards millions of useful ideas, facts and documents we might have relied upon for providing our world structure and function, for the sake of one choice.

I do think satisfying mysteries are possible in RPGs but there are limits. Since the inner workings of player character and NPC minds are mutable and essentially off limits, they do no satisfy the requirement of open information and consistency that solvable mysteries require. Any conundrum that relies on illusion or unreliable perceptions will be difficult to parse for the same reasons.

RPG mysteries need to be robust to different player actions and it must be possible for the players to fail to solve the mystery. A mystery novel with no solution is a failure but a roleplaying session in which players fail to unravel a plot or puzzle can be a springboard to adventure or something to return to later.

Careful information management is important. The GM should be willing to answer player questions and repeat important information but also try not to be too obvious. I have both played and run games where insights were telegraphed by an over-helpful GM and games where players felt utterly stymied. It is truly an art and very dependent on different groups. I think it important to be willing to tell players that perhaps they don't have enough information and let them move in another direction. Let them think of other ways of finding what they need. This is much easier in sandbox style games where players have a number of possible objectives. Don't use mysteries as locked gates that prevent progress in an adventure or you'll just frustrate your players. If they need to solve a mystery to advance you'll either need to give players enough clues to continue or expect them to be stuck.

Leveraging the advantages of RPGs, the best puzzles should have elements that allow players proactively investigate and to return to them over time. Either locations that can be revisited or witnesses that can be tracked down and interviewed, or clues that can be spread around. 

Justin Alexander has some fantastic content on implementing mystery scenarios in RPGs, with his 3-Clue Rule and Node-based Adventure Design posts a great place to start.