Character Death in Torchbearer

[Shreyas Sampat]

This has been nagging me for a while: players of Torchbearer (my mythic fantasy game; there are two threads on it a page or two back at the moment of this posting) have near-complete control over the longevity of their characters.

Simply put, a character has a pool of a resource (Fuel), and the player's authority over his Fuel is uncontested.  Only that player determines when Fuel is spent, and the player can easily have the character act so as to replenish the pool.

So long as the character has any Fuel, his player is the only person permitted to narrate the character's death.  Once the character runs out, any player may narrate that event; the player effectively gives up the exclusivity of his authority.  (Narration privileges are given to the participants of conflicts; his narration becomes the case who won the conflict, when it is finally resolved.)

Now, my question is this:
Is there a reason for the sacred cow that is involuntary character death?  How am I impacting my game by removing it?  How am I impacting the game by giving players the right to say 'My character dies', whenever they should so choose?
I have worries about what this means to antagonists, as well.  When command over death is this powerful, there are some things that simply can't be done - assassinations, particularly, must be spectacularly difficult in the world of Torchbearer.

[Andrew Martin]

Is there a reason for the sacred cow that is involuntary character death?  How am I impacting my game by removing it?  How am I impacting the game by giving players the right to say 'My character dies', whenever they should so choose?

Is your game intended to tell an interesting story about the PCs? Then involuntary PC death doesn't fit. PCs die when the players feel their time is up.

Is your game a simulation of people in the setting? Then involuntary PC death does fit. After all, accidents and assasinations do happen.

[Jonathan Walton]

I really think this is a non-issue.

Andrew's right, if PC death doesn't do anything to improve the tone or setting of the game (like in "Paranoia" or other games that live on PC death), I don't see why you need it.  In fact, I don't see why they need PC death in a good many games that include it, simply because it will rarely, if ever, come up in the game.

In any case, I think you've decided to handle it in a rather appropriate fashion, but I think it might make players feel like they have to hold on to their last point of Fuel, so they won't die.  I think that actually works against the purpose of encouraging them to take risks.  Perhaps what you want is to say that ONLY THE PLAYER (or another player with the permission of the playing involved) can narrate a PC's death.  Players without Fuel can be brutally injured by the narratives of the other players/GM, but the final death is ultimately up to the player.

That, I think, would be the way I'd run Torchbearer.  Still, I could easily just make a house rule if that's not what you want.

Later.
Jonathan

[contracycle]

Yeah - the Sacred Cow is that "danger lurks just around the corner" and is, by its nature, dangerous.

That said, I think theres room for player-inspired character death in primarily Sim systems; this is because sometimes the Sim would and should lead to the characters death, and when "death power"(tm) is held exclusively by the GM, the player is not free to author their characters demise as they feel it should happen beyond committing suicide in one form or another.

It also allows another Sim-consistent option to be exercised; I have often seen intra-party homicides arise from what are effectively irreconcilable differences between the characters.  In real life, most of these issues are not worth dying for, and one party would usually walk away - but, in effect that is the same as character death in RPG as the decision probably denotes removing the character from play.

In both cases, I think making a tool that the player can use available, and signposting its presence as not just an optional extra but a significant consideration in play, would give players in these situations a relatively uncontentious device to employ to minimise any possible bad feelings.  Its just highly probable that the reflexively sim player will need to be severely hacked off about something before they are thinking about the game rather than in the game.  Hence, I would expect it to be used infrequently by habitually Sim players; but that it might also be a very healthy safety valve.

[Ron Edwards]

Hi there,

What an awesome set of responses to the initial post. I have very little to add, except for this observation.

In the past year, I have been playing in an aggressively Narrativist fashion with a lot of different people. You know what shows up a lot? Player-voluntary character death. I'd venture to say that I've "lost" more characters in the past year than I ever did in using the traditional means of character death, largely because in the traditional games, the GM would fudge to keep characters alive so that his story could continue (or so that people wouldn't leave the game, pissed off).

Based on this experience, and also on my observations about group-negotiated outcomes (informally, Hero Wars and The Pool; formally, Dust Devils, Alyria, and Trollbabe), I think the existing Torchbearer system will produce some outstanding results in terms of characters dying or not-dying, both player-owner-determined and other-player-determined.

Best,
Ron

[Shreyas Sampat]

Thank you, all.
I finally pinned down my nagging problem: The structure of the narration system: a player only has the authority to narrate an outcome where he succeeds.

This presents a problem; most characters in fantasy stories die of failure, not suicide.

So, a change in the narration system is in order, one which allows a player to narrate his character's death when he fails.  I see two obvious options here:
You can hand death-narration rights to another player; this is a house rule I'd probably use anyway.
Or, you can say something like, "the player at advantage can choose which outcome to narrate; the other player narrates the remaining result; when each outcome is equally likely each player narrates his character's success".

Are these real solutions, or just weird patches to the rules?  I fear that my second option would result in a lot of weird ripples that I hadn't expected.

[Jonathan Walton]

A player only has the authority to narrate an outcome where he succeeds.

I would think the easiest solution to this would be a rule like:

"The player CAN choose to narrate a failue, but ONLY one in which the character will die."

This would put a more positive spin on things, giving the player a "secret power" to narrate character deaths.

"The player at advantage can choose which outcome to narrate; the other player narrates the remaining result; when each outcome is equally likely each player narrates his character's success".

Hah!  I really like this idea.  So, if the player wants to be cautious, he/she could choose to narrate the failure, but make it "not so bad" while the other player narrates a less-than-optimal success?  Sweet.

Still, I'm not exactly sure what this would do to Torchbearer.  As with the "can't be killed while you have Fuel" method, it might make players more cautious, since you've now given them the option to be cautious.  Ultimately, the idea is interesting, but I don't think it would do what you want it to.

Later.
Jonathan

[Ron Edwards]

Hi there,

I have a slightly different suggestion. Make the roll be entirely metagame, meaning that if the player wins the roll, he gets narration power. See? Narration power. That means I succeed with my roll, but I want my character to fail and die. No problem! I fail and die.

OctaNe (and in certain moments, Dust Devils) works like this.

Alternative: a variant on the Otherkind method, in which several dice are rolled. One of them indicates how well the character did in the task; another indicates how badly hurt he is; yet another indicates who gets to narrate. (I'm abbreviating, actually; Otherkind uses four dice.)

But here's the point: you roll first, then choose which die is which. So if I roll high, low, and low, I might choose to fail the task, be badly hurt, and keep the narration - in which case, I say that the character dies.

Anyway, some version or variant of the principles in octaNe and Otherkind might be worth considering. In each case, character success at the task is decoupled from narration-rights, and it's narration-rights alone which determine whether the character lives or dies.

Final alternative: don't worry about it. As it stands, all our talk about "trust the player" should include "trust the GM" too. I think you have such a strong system and concept here (based on my reading), that you may not have any play-problems at all by having the GM always narrate a failed roll.

Best,
Ron

[Shreyas Sampat]

Ron,

Now that's a perceptive idea, decoupling success and narration power completely.  Interestingly, I think that if one simply tweaked the resolution system a bit and used this idea, we'd get something Jonathan might find interesting: Traits that allow the character control of the story, rather than providing pure Effectiveness.  I imagine that with some changes in wording, this could become a very elegant way of making an I-game, if I understand correctly that the term I-game means 'a game in which the players play game versions of themselves".

Allow me to explain.

Currently, conflicts in Torchbearer are resolved much like in Shadows: Call for a roll, total up the modifiers (with the narration that this produces), agree that everyone's done trying to influence the result, narrate outcomes and roll the bones.
In this incarnation, Traits control which character successfully executes his action.

In FitM/narration-rights form, we get something like this:
Call for a roll, total up the modifiers, state goals, roll the bones, allow the player who wins to narrate as he chooses, consistent with one of the stated goals - this is essential if we are to give players even this tenuous narrative control over others' characters.
In this incarnation, Traits control which player gets the right to decide what happens.

Pretty interesting.

[Christoffer Lernö]

Let me offer a quick take on the player death thing.

I think that even any sim GM of worth don't let a character die until he has run out of options.

For example, even in sim, let's say I have a character and it looks like he's being overwhelmed by monsters. I can choose to run or stay and fight. If I stay any fight (abandon my options) I might very well die. If I use my option to run I survive. Now if I don't have a plan for running. I decide maybe that the safest way to go is the ork cave since all the orks are out hunting. Unfortunately the orks are just getting back because the other players led them back home. Again the character has problems. The character could try to hide and avoid detection or try to fight it out. Again there are options. Until I make certifiably (or at least very likely to be) bad choices I should not even die in Sim.

Now a bad GM would have the fleeing character being shot in the back by ork archers or something, but that's not something to try to strive for.

Anyway, the point is: In good illusionist sim there is a similar effect as the one you describe.

For the assassination things I would suggest a work-around: "Things are never what they seem" and "-I thought you were dead? -I got better" are two methods to create what seems to be death but isn't.

For example: the killing blow that isn't killing. Oh, sure he stabbed you in the chest but you can even take. You just shake it off and ordinary humans stand in amazement. Hey you're a legendary hero, you can do that stuff.
(Yggdrasil tie-in: In Ygg when you receive a serious wound you get penalties, but when you get a mortal wound you don't get any penalties at all, you simply roll for how long it takes for you to die. It might be 1 second or 1 hour, until that happens, no penalites - but you unlike the serious wound there is no way to survive it. You can go out in a glorious way though)
Or "what you just killed was not me but my doppelganger" or "the corpse I dressed up to look like me"

[Jonathan Walton]

In this incarnation, Traits control which player gets the right to decide what happens.

So, basically, what you're saying is that the player with the higher modifyer in any given situation is the one most likely to gain narrative control.  Interesting.

Of course, Traits and achieving the higher modifier aren't directly connected.  Varashi might have "The Compassion of Angels," but if she is trying to kill Neranja, her player probably wouldn't want to bring that into the conflict, even if it would give her a higher modifier.  Traits provide you with an advantage, but they would also limit how you could narrate the results, after you won narrative control.  This, I think, would make for interesting choices.

Additionally, this suggests some interesting side mechanics.  Since conflict in Torchbearer is completely dualistic as far as I can tell (one thing conflicting with another thing), you could have ways for other characters to serve in "support" roles in a given conflict, adding to a specific character's modifier and helping that player gain narrative control.  Additionally, there could be ways for characters to support GM-control in some cases, siding with the environment or NPCs.

Just some thoughts.

Later.
Jonathan

[Shreyas Sampat]

The assistance mechanic is worrying me, actually.  I want to maintain the duality of conflict and exclusivity of narration for the winner, but still allow for assistance and group tasks to occur.  How do I propose to do this?

Here is my thought.
When a conflict is declared, anyone may decide to involve himself.  Involvement takes two forms:

Interfering: The involving character narrates a subconflict with the character he is Interfering against.  If he wins the conflict, then for each success he has he may give the victim a Smoke that applies for the duration of the scene.

Nimeshtim Scarlet Lark is travelling through the wood with her friend Ereshta Radiant Crane.  They are accosted by a spirit of the land, a huge tiger with teeth of white jade.  It cocks its head and looks at them, then leaps toward Ereshta.
Conflict!
Nimeshtim, thinking quickly, rips the silver chopsticks out of her hair and throws them at the tiger.  They strike it in the eye and throat, and it rears up in pain, giving Ereshta the instant she needs to dodge out of the way.  Now the battle is joined in earnest.

The other option is Assistance.  This is mechanically a little strange - there is nothing for the assister to oppose, and without that mitigating factor, assistance becomes obviously a better option.

[Jonathan Walton]

The other option is Assistance.  This is mechanically a little strange - there is nothing for the assister to oppose, and without that mitigating factor, assistance becomes obviously a better option.

Well, in the interest of unity, why not make Assistance work exactly the same as Interferance, except, instead of giving a Smoke to the opponent, you'd nullify a Smoke of your ally (for a scene).  You could even make this an opposed roll of some kind (against the Smoke Trait?).

Nimeshtim, knowing that Ereshta is "Slow to React", rips the silver chopsticks out of her hair and throws them at the tiger.

"MOVE!" she screams at her friend.

A combination of the tiger's momentary distraction and Nimeshtim's warning enables Ereshta to scramble safely out of the beast's path.  Now the battle is joined in earnest.

I'm not sure if that would work in ALL circumstances, but you could probably play with this idea a bit to make it more universally applicable.

What d'ya think?

Later.
Jonathan

[Shreyas Sampat]

That's not a bad idea; I was considering allowing the creation of Flames instead as well.  That somewhat parallels Trait reversals.

My next concern is cooperative actions where the characters initiate the action intending to work together; here I think I will allow them to simply operate as one, as far as that goes.

Edit: Missed a reply to Chris which I intended to include here.

Yes, the more I think about it the more I believe that character death should be a tactic of very last resort.  In a game that's developing into Sim Exploration of Character (I think), it doesn't make sense to remove the primary area of interest.

[Jonathan Walton]

My next concern is cooperative actions where the characters initiate the action intending to work together; here I think I will allow them to simply operate as one, as far as that goes.

:)

This just means I'll have to refrain from yelling "Wonder Twin powers, activate!" during large battle scenes...

Seriously, treating a group of cooperating characters as a single meta-character could resolve complex battles with a few rolls.  You could break things down as much or as little as you wanted.  A game that was really focused on the narrative might have one roll for an entire battle, where the victors would get to co-narrate the results, applying various Traits in turn.  Or, if you wanted to make things more dramatic, you could roll for each round of "X combined heroes" vs. "big supernatural nasty horde," with different individuals cooperating in different instances ("Hey, Varashi!  Can you give me some help with this big one?").  Very cool.

"Let our powers combine!"

Later.
Jonathan