Many years ago I stumbled across a copy of Ars Magica, which is a roleplaying system/setting set in the dark ages revolving around cloistered magi doing arcane research. The recommended/standard adventure is one in which one of the magi ventures into the mundane world for obscure supplies, a task for which they are poorly equipped. So they take helpers with them on their errands, thieves and warriors and the like to help them navigate the cruel world outside of their ivory tower. The game system does some clever things with the players and the characters, suggesting that the players rotate who plays the magi and who plays the helpers.
I really liked the setting, and I particularly liked the tone of Ars Magica. It does a good job of establishing the mages as reclusive, socially-inept scholarly types of different persuasions, and the world outside the walls as being a dark and intimidating one for them in spite of their powers. I’m always looking for variations on my usual template for campaigns, and I thought that a more quest-based approach where all of the PCs were novice mages at a cloister would be a fun change of pace. I decided to keep the dark ages tone, and set it in Rumania during the time of Vlad the Impaler’s reign and the Third Crusade.
The idea was to send the PCs out into the world on three introductory missions, one with each of three head magi at the cloister. When they returned from the third mission, they would find the cloister pillaged and destroyed, its inhabits all dead or missing. This would then lead them on a quest first to rescue the captured magi from the school, and then to pursue the evil rogue mage that left the school many years ago. Somewhere along the way, they would figure out that one of the three head mages had betrayed them, and have to figure out which one.
I say “the idea was” because this campaign utterly failed. Part of the reason it failed was a near-TPK event where one of the four PCs was killed and another rendered insane. But the bigger reason was that a party of nothing but mages poses some serious problems.
How Magic Works
In pretty much any RPG where magic exists, it works in basically the same way: the wizard has great powers, but he has some limitation on how much or how often he can use them. In GURPS, my system of choice, every usage of a spell costs fatigue, of which the character has a limited supply at any given time. In D&D (at least pre-4th ed), the wizard/cleric has a certain list of spells that he could cast each day, and a number of times that he can cast them. In other systems, the wizard can cast as much as they like, but there is a chance of them taking harm from their own magic.
All of these systems serve the same purpose, to balance the magi against the non-magical PCs. The wizard can cast a giant fireball, but he gets tired easily and needs the big meat shield to protect him while he gathers his strength. Without these sorts of limitations, the wizard would outclass his companions and make them obsolete (and playing a useless character is pretty much never fun). You commonly see this occur in balanced systems as power levels get higher. As the wizard’s abilities grow, they get better at circumventing their limitations and begin to outpace the other players of equivalent experience/point levels.
Magic: It Slices, It Dices
Magic is, by its nature, kind of a trump card. Let’s say that you have a PC in your party that can cast a spell that opens locks. That PC now has an automatic solution to every lock they encounter. This puts the GM in something of an awkward spot. If you really want an inaccessible room in your story, you can’t put a lock on the door. You have to make the entrance to the room inaccessible in some other, probably less plausible, way. The knowledge of the lockpicking spell trumps all locks. Similarly, if your wizard can fly, or become invisible, or read minds, or walk through walls, these things all provide specific trumps to various situations.
In a traditional party, with one or two wizards and a few guys with other, more physical skills, this is a problem that the GM must work around, but it’s not crippling. So the wizard can open locks. You’ll put all of the cool stuff at the top of giant walls so that the thief gets a chance to climb up there and drop a rope.* If you’ve got a guy who can make himself invisible, you defend something important with wolves that can still smell him, and so on. That way, the wizard still needs the skills of the other party members.
When, however, you have a party full of magi, they are going to have a wide array of powers, of trumps that they can play in various situations. One can fly, one can become invisible, one can see around corners, one can heal wounds, and so on. The players have too many trumps. Game play turns into a series of puzzle solving sessions between the PCs, followed by logistical concerns. “Okay, we’ve scouted the area with Invisible Wizard Eye. We’ve scryed up the exact position of the ancient staff of Amun-Ra. Roger, you cast Flight on Steve. Then I’ll give him Invisibility, then he can fly over all of the guards to the tomb, where he can magic open the lock and grab the staff. Then we’ll all go hide and rest for 4 hours and sneak away.” What’s worse, if the magi encounter a problem that they don’t have a trump for, they don’t have a meat shield to hide behind. This tends to end badly.
Consequently, their plans either account for every possibility and succeed flawlessly, or miss some critical factor and fail disastrously. Counterintuitively, the second outcome is usually the one that is more fun. When everything goes to plan, it can be anticlimactic. When things go wrong, that’s when people have to get resourceful or lucky, and that results in better play experiences. None of this is to say that puzzle-solving isn’t fun. I, like most nerds, love me a good puzzle. But if people just want to solve puzzles, there are better, faster,ways to do that than RPGs. When a campaign breaks down to a series of puzzles, it’s underdelivering on the potential of the medium.
So How Do You Fix It?
The concept of a setting where the PCs are all wizards of some stripe is too appealing for me to just give up on it. The first thing that I’ve been thinking about is Magic systems without constraints. No fatigue costs, no finite number of spells per day, no percentage chance of catastrophe. This is magic more in the Harry Potter or Gandalf model – the wizard can cast as much as he wants. I would never consider this in a traditional setting because it would make the wizard so much more powerful than the other PCs. But if all of the PCs (and, consequently, the important NPCs) benefit from this lack of restriction, then power-balancing is kind of a non-issue. This turns the “anyone without magic is underpowered and irrelevant” from a bug into a feature. I hypothesize that if the PCs could cast endlessly, they will be far less inclined to carefully plan out each encounter, and more willing to just go in spells blazing (which lends itself to more Rockmost moments).
This is not to say that I want magic to be the equivalent of just having super powers. The fictional work that I really want to model is Lev Grossman’s book “The Magicians.” In it, the characters can cast spells freely, but casting requires a lot of concentration and calculations based on your current environment. So a character could, for instance, levitate a book off of a table. If they have time to concentrate and think about what they’re doing, and they are under little stress, they can do it pretty much every time. If they’re under stress and in a strange place, then it gets harder.
This captures, to me, the way that it feels like magic would work, were it real. It’s like computer programming. If you’re sitting at your usual machine with your familiar tools and settings in your office working on something, you’re pretty fast and efficient at it. If you’re about to give a major presentation at a conference and the system crashes and you have to try to fix it on some strange laptop and you only have ten minutes, it becomes much harder to concentrate and execute effectively. I want magic to play like reprogramming reality. This is, in practice, much more cumbersome than it sounds. If I want the players’ ability to perform magic to depend on the affects of outside factors, then all of those factors have to be delineated and assessed for every casting. But I digress.
The other factor that I think would help to prevent the “sit and plan, then watch everything unfold” outcome is restricting information. The PCs in the Cloister campaign had some pretty powerful information-gathering spells at their disposal, and this enabled them to scout and plan instead of being forced to act. If the PCs can’t scry up the exact location of the item they need, and then invisibly survey its location, then they are forced to act with incomplete information, which leads to more “well now what are we going to do?” sort of moments.
The Takeaway Lesson
I think that people learn more from their failures than they do from their successes; what I learned from this campaign is that chaos and uncertainty are fun. When the players don’t know what’s going to happen next, or how they’ll possibly escape from certain death, they’re excited, involved, on the edge of their seats. And when they come up with that amazing solution or that lucky roll that saves them – those are the moments that people tell stories about years later. Of course, sometimes they don’t find a way to save themselves, but that’s a topic for another post…
*Alternatively, you could just tell the PC that they can’t take the open locks spell when they are building their character. I find, however, that these sorts of restrictions tend to remove some of the enjoyment for the players. It’s fine to say that they can’t take Necromantic or Mind Control magic because it’s illegal in the setting. And it’s fine to say that the arts of healing were lost to time. But to say, “Oh yeah, you can fly, and walk on walls, and throw fireballs, and talk to the dead. You just can’t open locks.” tends to break suspension of disbelief. It’s obviously a cludge to stop the player from breaking the campaign, and it jars people out of the experience.
Author’s Note: I’ve attempted to tackle this topic before, but I feel like the approach that I took was ineffective. I’m trying again because I feel like it’s an important, foundational piece of the way Ben and I approach role-playing, and also is full of win. Hopefully this will be a more successful endeavor.
There are three components to role-playing campaigns: story, setting, and characters. Most of the medium is dominated by setting-based role-playing. The GM and the players agree on what system they want to run, and that system comes with a pre-built setting. That setting then suggests what sort of characters and stories should result. Traditional D&D is a high fantasy setting laden with monsters and treasure-rich dungeons. The characters will therefore be adventurers, and (most) of the adventures will consist of slaying said monsters and exploring said dungeons. Shadowrun is a cyberpunk setting with magic and hackers where giant corporations run things. The characters will therefore be shadowrunners who work for/against these corporations for money. Star Wars, In Nomine, Vampire the Masquerade, etc all follow this pattern. They provide an interesting world to play in with some obvious built-in role-playing scenarios, and then leave the rest to the players.
A lot of the new Indie games are trying to change this model, focusing less on setting and more on story. The products suggest new and interesting guidelines for how to build and tell the stories, with an emphasis on collaborative storytelling. In both of these structures, however, the characters are kind of an afterthought. That’s not because people don’t like characters, it’s because it’s much more difficult to make the characters the primary emphasis.
I have a friend named Dann, and he was able to come up with a template for campaigns that successfully put the focus on the characters, and allows them to drive the story(consequently, we will often refer to this as a Dann Campaign). This template relies heavily on some of the techniques from The Way We Do Things, particularly the ideas of people being out of the room when their player isn’t present, and campaigns with definitive story arcs and endings.
Making the Characters
In a character-focused campaign, each Player Character should feel like the star of their own movie. The story should be about things that happen to them, and how they respond to it. This is qualitatively different from the traditional, party-based, mission-based approach. The traditional approach is to say “You’re a party of archaeologists on a quest to find the true tomb of Tutankhamen. Go make characters.” The character-based approach is to say “You all live in ancient Babylon. Let’s meet to make characters.”
But wait, you might say, didn’t you just start with the setting? Nothing gets past you, savvy reader. For reasons that will be explained as we go, a character-focused campaign requires a confined setting. One of the goals of this approach is to allow the players as much freedom as possible in what sort of character they wish to play, but some structure is required. Depending on the campaign, the setting may be as specific as a particular pirate ship, or as vague as “the modern world.” After the GM has informed the players of the setting, s/he meets with each of them individually to create their PC. Since the whole story is going to be built around the characters, good PCs are a must, and the PCs must be rich and complicated characters with deep back-stories.
Building the Story
After the PCs have been created, the GM goes and figures out how to work them into the plot. In most cases, the GM already has some sort of vague story in mind, and it’s a matter of figuring out how the events of that story would effect the PCs and draw them in. In other cases, new branches or elements need to be added to involve them.
I GMed a campaign set in a somewhat traditional fantasy setting, set in a walled city called Bakad*, that happened to be governed by a vampire named Toruf Tar. The city was near a massive, Roman-style empire, but had up to this point remained independent. The empire had tried to seize it many times, but could not. Bakad was unusual in that it was located on the edge of a vast untamed jungle, and had complicated rules surrounding legal assassinations.
The back-story of the setting (which the PCs did not know) was that long ago Toruf Tar and a gifted doctor competed for the affections of the same woman. The woman chose Toruf Tar, and the doctor decided that he lost because he could not offer eternal life. He devoted his life to this quest, and created two rings that made the wearer cease aging. He gave one of them to the woman who spurned him, but she never wore it (nor did she choose to become a vampire). The other key factoid was that Toruf Tar had two trusted lieutenants, the only other vampires allowed in Bakad, and one of them tried to turn on him a long time ago. They fought, the lieutenant lost, and was thought dead. After many years of hiding, he’s made contact with the emperor and struck a deal – he’ll give the emperor the ring (and eternal life) if the emperor makes him the new ruler of Bakad.
There were four PCs: a bounty hunter with a kid to feed and a job with the mob, a spy who worked for Toruf Tar and was obsessed with becoming a vampire, an assassin who considered killing his art, and a scholastic wizard who specialized in plants and animals.
So how to tie these PCs into the same story? The spy was easy – since he worked for the government of Bakad, he could be sent to investigate things. With the assassin, I used an old favorite; I framed him for murder (in this case, killing without a contract). He inadvertantly killed one of Toruf Tar’s spies in the city, so the spy PC was sent to track him down. Their investigation led them to a wealthy merchant inside the city who was collaborating with the Emperor to overthrow the city.
Tying the mage in was far more difficult. As a scholarly character, he was reclusive by nature, so he didn’t really care what was going on in the city. I started his story by having him learn that his mentor had been brutally murdered, and that he was the inheritor of his possessions. One of the things that he inherited was a key, which was supposed to open a trunk. The trunk, however, was missing. Then I had the bounty hunter go to collect on a debt for the mob that lead him on a quest for the same trunk. As he followed the chain of ownership, every person who had owned it had been murdered in the last week. This lead him to the mage, and eventually led the both of them to the renegade vampire hiding under the city. Eventually, the two pairs of PCs encounter one another and figure out that their stories are related.
The reason that I used this example is the trunk. The trunk, of course, contained the ring, as well as several other possessions of Toruf Tar’s long-dead wife. Though the trunk was just a plot device to obfuscate the ring, it became an object of importance in it’s own right, and the scene where the PCs managed to sneak in and open it was heavy with rockmost. I filled it with several magic items as red herrings to disguise the nature and value of the ring, and the items became cool things for the players to use later (when there’s no plunder or dungeons to raid, any magic item becomes A Big Deal).
When this comes together smoothly, it feels very natural for the players. Their character feels like a well-developed person with their own life, friends, and goals. Something happens to threaten one of those three things, and they are forced to respond. Along the way, they meet some other remarkable people dealing with problems that seem to be related, and they work together to discover the root cause and deal with it. Then, once the problem is solved, they go on their separate ways, hopefully changed by the experience.
Since the various PCs start off singly, and then slowly come together as the story leads them, you need a story structure where various seemingly unrelated things all end up having a common cause. The easiest way to accomplish this is a conspiracy-based plot. That allows the GM to create branches of the larger conspiracy that will reach into the realms that the players inhabit.
Aside: I’m a big fan of mystery-based plots, mostly for the reason that they are hard for players to short-circuit. In quest-based campaigns (retrieve the sword, kill the guy, destroy the ring, save the kingdom, etc), the goal is immediately obvious, but seemingly very difficult. Clever players, however, can come up with solutions that the GM didn’t think of that effectively ruin the campaign (“Why don’t we just have the eagles fly to Mordor and drop the ring in Mt. Doom?”). If, however, they don’t know what’s going on, they are forced to tackle intermediate goals before they can take on the final bad guy.
Since the players will be doing a lot of solo time at the beginning (Ben has written a couple of posts on this topic), and the GM has to create individual stories for each player, it’s important that there be a small number of players. The most that I would attempt is four, and three is really optimal, as it makes for shifting dynamics between the PCs. I would sooner try two (a buddy cop story, for instance) than five.
Putting it All Together
So based on the above, we’ve got the formula for a Character-focused Campaign:
- Each Player Character should feel like the star of their own movie
- Rich and complicated PCs with deep back-stories
- A confined setting
- A conspiracy-based plot
- A small number of players
This is not to say that this is the only way to execute a character-focused campaign – it’s just the only one that I know of that works. I really like criteria 1, 2, and 5. I feel limited by numbers 3 and 4. The problem is those pieces are deceptively important. Quest-based stories, by their very nature, take a lot of the focus off of the PCs. They become interchangeable cogs in a larger story. Even if your players create really great, interesting characters, most of the decisions being made by the players are the ones necessary to accomplish the quest. Ben and I have talked a lot about trying to tell stories that are neither mystery-based nor quest-based, but they frequently run into the Circular 4th Wall problem.
Similarly, the confined setting is deceptively important. It’s very difficult to tell a story with multiple locations and still allow the PCs to each have their own story. One of the incidental benefits of quest-based role-playing campaigns is that it’s easy to send the PCs from one place to the next. If the PCs don’t start out together, and the characters are making the decisions that drive the story, getting them to go to the next location all at the same time often requires railroading them, or giving them extremely unsubtle hints. One of the real virtues of this campaign structure is that it gives the players a high degree of authorship in the campaign, and any sort of railroading undermines that.
There aren’t very many examples of stories told in other media using this sort of structure, but there are a few. Most of William Gibson’s novels follow this sort of structure, where the story follows three or four seemingly unrelated people who’s stories turn out to be related. In Gibson’s books, the characters don’t usually come together until near the end, whereas the ideal structure for a campaign of this type has the players meeting around the end of Act I (more on campaign structure here).
Another example of a story using this structure is the first season of Heroes. The characters don’t know one another, and are all dealing with their own problems caused by their powers while solving a larger mystery/conspiracy. The big difference, of course, being that the characters are not confined to one geographic location. You can see, however, the devices that the writers use to overcome that limitation. Prophecies, time travel, serendipitous coincidences. It might have been simpler just to start all of the characters in New York City and then have them meet through more organic means.
The third, final, and best example of this structure is LA Confidential. Three cops, with completely different personalities and goals, all pursuing different cases that turn out to be related through a conspiracy. The characters are proactive, taking actions that actively move the story, instead of just reacting to the bad guys. And at the end, each of the characters has been transformed by the experience.
*Ben, our resident linguist, informs me that I should spelled it Bakhad since the accent is on the second syllable.
I love props at the table. I may, one day, post about props in general, but I want to talk about a specific type of prop that I think people are, in general, pretty familiar with: the map. I divide maps into two gross categories: Big Maps and Little Maps. These names have nothing to do with the size of the paper the map is printed on or with the size of the area they cover. Rather, they have to do with the importance of the map to the campaign.
Big Maps are those that are applicable to your whole campaign. In Kjemmen, which is intended to take place pretty much entirely within the city of the same name, I have a map of the city, which is my Big Map. If you were doing a campaign where PCs were expected to planet hop, your Big Map might be the whole quadrant and/or their space ship.
I love big maps because they give the campaign a sense of place. They really ground the events for the players by letting them see where things are happening compared to other things. If you provide a big map (especially to each player so they can write on it), they’ll always be asking for you to pinpoint on the map where some location is that they’ve been.
Little Maps are incidental maps. Mostly, they’ll be good for just one scene. I don’t use miniatures, but if you do, I imagine all your battle maps are like this. If I need a map of a battle space, I generally draw one up on the spot and tell the players it’s loose and not to scale. We can point with our fingers and I arbitrate distances. In fact, for most Little Maps, I draw them up on the spur of the moment. I have an idea of what the space is like in my head already, so I sketch it out really quickly.
This requires some artistic ability, of course, but I’m mostly just drawing boxes, curves and circles on notebook paper, so it’s not that you need to have a degree in Architecture or whatever. I generally don’t even know when a Little Map is going to be needed ahead of time. You might be pretty sure that there’ll be a battle at the Keep this session, but my players regularly wonder what the layout of a random stretch of street is because they want to ambush someone in an unexpected location or whatever.
In The Middle
Okay, I lied. There’s sort of a third category. I ran a paranormal FBI campaign sort of like the X-Files. I even copied the episodic nature of the show (some cases unrelated to the over-all story). One of the cases took place in a ski town for which I made a map. Another had to do with the tunnels that are under Niagara Falls of which I made another map. Both of these maps were around for more than one session and I prepared them ahead of time, so they don’t really feel like Little Maps. However, neither was applicable to the entire campaign, so they don’t feel like Big Maps. My only point, really, in bringing them up is to say that these are also fine and a thing that I do. This isn’t a strict classification system.
So What Goes On There?
Another thing that I have sometimes found difficult is deciding what goes on a given map. Sometimes that means I’m having a hard time deciding the boundaries of a map. For Kjemmen, it’s easy: the game happens in the city so the map should be of the whole city and not much more. For a game set in, say, the Star Wars universe, though, do you want to have a map of the whole galaxy? Of just the little part of the rim where your smuggling-focused story takes place? It’s sort of a hard call. The only advice I can give you is that if your map is big, it gives the impression that your world is big (and fleshed out) as well, which is good. If your map is small, the players’ attention will be more focused on the region covered. I think the happy medium lies in having a map that covers more than you’ve strictly detailed out, but not more than you could make up details about if they go somewhere unexpected.
Another quandary is deciding what sort of markings should go on a map. For a region of land, say, you might be wanting to put cities and towns and roads and forests and rivers and bays and mountains and… all with names. You also want to print it out on an 8.5×11 piece of paper. Things can get a bit crammed. A couple of solution ideas I’ve had are: Cut your map up, which basically makes it larger, and hand out more than one sheet of map; Make more than one map. For Kjemmen, I wanted to include specific locations as well as the names of noble houses that controlled a given stretch of territory (which ended up being like landscape names). I printed off two maps for every player: one was the land-scape with the noble names and the other was dots and names for the specific locations they knew about.
Finally, is an idea I’m stealing from Stewart. Sometimes, the best map is not a map. He and I like to think of a campaign being centered in a specific location and so we talk about that aspect pretty early on when discussing campaign ideas. He ran a campaign called Amunaven for which all the PCs were among the youngest generation of a family of, more or less, gods. The eldest were on the order of Ares and the youngest more like Achilles. We had crazy abilities in a bronze age world and could teleport about with magical stones. So a map didn’t make sense, really, because he didn’t really want to hand us a globe. Instead, he handed us a family tree. The family was big (maybe 30 total members over 4 generations) and very central to the plot, so the family relationships were, in essence, the location of the campaign. This is more generally applicable than just family trees, of course: the point is to keep in mind that the purpose a map serves is to make concrete some aspect of the story that is central and omnipresent, which is often, but need not be the physical space it occurs in.
My next post will explore map making in a bit more technical detail: I’ll talk about some resources I’ve found helpful, some techniques I’ve used and probably share some examples. What kind of maps do you make for your campaigns? How do you decide what to map? Let us know in the comments.
I don’t mean having flashbacks of gaming and I don’t mean dealing with PCs who have Post Traumatic Stress Disorder; I mean having scenes every once in a while that have happened in the past. This idea came out of a discussion Stewart and I had about campaign ideas (as is so common with posts here). We were riffing on the themes from The Usual Suspects, which, without too many spoilers, is effectively told through a series of flashbacks. That wasn’t the aspect we were interested in, but we got sidetracked onto it. It’s not an uncommon trope in fiction and so I spent some time thinking about it and arguing with Stewart about it. This idea is still very embryonic and I haven’t quite figured out how to use it or whether it would even be fun.
I’m envisioning scenes where the PCs are all doing whatever, learning about the plot and trying to uncover What’s Really Going On and they unearth some clue. It seems to be an arms cache of some kind with lots of laser rifles and explosives. They open a Batman-style armor closet and you describe it, then kick everyone but the pilot out. You play a flashback scene with him which, in the end, features a bounty hunter dressed in just such a suit of armor which ruined spectacularly some job. Then you cut back in to the present and move on. You could use this method to disseminate additional clues, tie in PCs more tightly to the story, present new questions that need answering. I’m sure there are more things.
What Purpose Do Flashbacks Serve in the Narrative?
I’m not overly enthralled with the show as a whole, but I do have to admit that the folks working on Lost (over)use flashbacks pretty effectively. I’ve got the idea that it would be really cool to do something like that in a game. You’d have to get a lot of information about your PCs… maybe ask your players things like, “Tell me about a time a job went bad.” or “Describe the way your last relationship ended.” or something. The idea would be to get vague stories that you could work into the plot. If you can do it right, you make the story of the campaign about stuff that has already happened to the PCs, sort of. Like when, in season… I think it’s 2 of Lost when we meet Desmond we also learn that one of the other main characters had met him before ever coming to the Island.
What I’m getting at, here, is the idea that a flashback is generally a way for a storyteller to tell the audience (in our case the player, considering the Circular 4th Wall) that something they’ve just seen or something they’re about to see is important. In our medium, I think it’s important to include player authorship in it, hence my invocation of asking them lots of questions. Spoiler warning, but consider the episode of Firefly where the crew goes to pick up their mail. That episodes flashes back at the start to the Unification War in order to introduce the young guy and show his relationship with Mal and Zoe. That way, when he shows up a few scenes later (back in the main timeline), his presence and the characters’ reactions to him make sense to us. If the flashed back scenes are from a PC’s history, it seems like you could increase their buy-in dramatically.
There are some problems I have, or, if you want to be more optimistic, hurdles I haven’t yet figured out how to address. Chiefly, if you set things in the past, you’re getting into railroading country. You can’t have the character die or lose an arm or anything. You’ve also probably already talked over the scene with the player so they know how it ends, which is sort of the player self-railroading. I’m thinking maybe you just keep it short short short. If they (in the past) were breaking into the office building to steal files and things got hairy so they had to escape down the garbage chute, well, they know the ending is they escape down the garbage chute. I’m thinking you end before then, then. Maybe you just cut in while they’re dodging security and breaking into the secure room so they can see the corporate logo (which just “reminded” them of the flashback events in the main stream) and you end with them being discovered, which they then know goes on to the garbage chute. I’m not sure.
Also, there’s the question of how, how much and when to start tying the PCs’ histories into the main plot. Do you try to get them all to tell you about events that might have to do with your mafia plot idea or do you just try to get them to tell you stories and then piece something together given what they told you? This one, I don’t even have a clue on. You could end up with something that isn’t linked strongly enough to the PCs, with something that plays like a Mad Libs RPG or just plain tipping your hand by being too suggestive asking for stories.
Lastly, there’s balance. Not game balance of the TPK-worried sort, but in that you have to try to make sure all the players feel that their flashbacks are roughly as relevant as everyone else’s. Especially if your campaign (as is my thought) were to make flashbacks a central theme or rhetorical style, if you will, you have to spread it around evenly. Which comes back to how many stories do you get from each player and how leading are you in asking for them. Because the content of those stories is entirely player-based and necessarily when they’re least familiar with your campaign world, you might get nothing useful from one player and all gems from another.
This is, as I said at the outset, a hare-brained, early-stage idea. I’d love to hear some thoughts from readers on the topic. Have you done anything like this? Have I missed some obvious and fatal flaw with this idea? Can you even envision a way this would be fun? Let us know in the comments.