So I was playing Dogs in the Vineyard, and I absolutely love the Traits-as-experience system, for its foolproof nature. If you make a mistake and write in a trait that it turns out (later) you don't enjoy then you simply never bring it in. No harm, no foul. The player can miss an opportunity to do something better, but they cannot in any way screw up their concept of the character.

And I thought: "Wow... something where a person cannot, no matter how badly they screw up, ruin the fun of a person. And where, in fact, if the first person suggests something off-kilter then the second person is encouraged to be creative in finding a way to make it their own. It's a damn shame they're both the same person, because if they were different people you'd have a great safety mechanism for negotiating the SIS."

For Dulcimer Hall (a game with conspiracy-soap-opera levels of lies, revelation and speculation) once character creation is complete you may never again add Traits to your own character. You may only add Traits to the characters of other players. If and when they choose to use those Traits, it gives them a bonus to effectiveness. It also earns you resources, yet to be defined, but sufficient to drive you to try to make the Traits you add valuable and interesting to them.

Are players going to balk at this? Or is the inherent line-item veto of being able to choose which Traits you do or do not use enough power?

Players balk at nothing, in my experience. When you decide to play something, you don't generally start whining about particular rules. You might not play it a second time, but you'll try it at least once. If I were you, I wouldn't worry about any other kind of players.

While you're doing this kind of system, why not make it more colorful by tying the trait allocation to concrete in-game events? In my experience it's always better if you can interpret the mechanical actions as something happening in-game as well. Even if the players don't explicitly say it, they can make such interpretations privately.

So instead of getting to add traits nilly-willy, you have to gain play position that allows adding such traits. Your own character could witness the other character having a fit of rage, after which you can say, like "Man, you're one disturbed son-of-a-bitch", after which you add "Disturbed son of a bitch" to his character. This becomes interesting when you recognize that the guy playing your mother can give you virtually any traits at all. "Oh, I've always known that his fits of temper can get destructive..."

If you're doing a game of lies, revelation and speculation, I would think that the above would work nicely, because then the characters would be imposing all kinds of traits on each other, instead of the players doing it on the meta-level. Trait adding would be one of the main sources of action. (What is a romantic plot but characters running around, trying to generate love traits in each other?) I don't find the meta-level solution fueled by resource rewards too interesting, and could imagine it being a source of dissatisfaction: there's no rhyme or reason in the trait addition, so it becomes abstract in the same manner as Universalis. But if the characters have a stake in each others traits, that's interesting.

I think the idea is very powerful, Tony--maybe too powerful. Whether or not you use the traits somebody else gives you, they're still on your sheet (if I'm understanding you correctly). This sounds to me like a recipe for making the social contract very, very delicate, especially in a game that already concerns itself with secrets and lies.

Player A: "Sorry, I can't help you get into the museum. It's against my moral code."
Player B: "But you have incentive. You just gained the Trait 'wildly in love with Character B.'"
Player A: "What?!"
Player B: (shrug) "Come on, it helps the story. Now you have to help."
Player A: "No I don't. I don't have to use it."
Player B: "Fine. You still have to write it on your sheet."
Player A: "Well... I'd never fall in love with you, because you just gained 'vast collection of furry porn!'"

I don't think players would balk at giving Traits; they might balk at taking them. The best practice, I think, would be a two-chance veto, both at creation time and at use time. I also think Eero's idea of tying Traits to concrete events (postfix, not prefix) would be a powerful addition.

>nod, nod< I see what you're saying, Brendan: unless there's some specific flag that the Traits are provisional then the initial creation of the trait is a unilateral change of the SIS, not a negotiation.

My sense is that many Traits should be provisional until used... and that the only ones that would not be provisional are those that reflect something that has already happened in the SIS (as Eero points out).

So here's an initial schema for that:

• Traits fall into one of three categories

• Secrets• Beliefs• Image

• Secrets are things that may or may not be true, which the character has not decided their opinion of yet and may not realize ("I love Knightley")• Beliefs are traits that the character is consciously aware of and (for the moment) agrees with ("I am shallow")• Images are traits that the "public" may become consciously aware of. The character has not decided her opinion of them yet ("Emma is a sweet, kind girl")• Any player may introduce any Secret on another player. ("Knightley loves Miss Bates") This does not reflect any change in the SIS.• Any player may introduce a Belief on another player that reflects something they have observed happening. ("Knightley has behaved with consistent kindness and attention toward Miss Bates")• Any player may introduce an Image on another player ("Emma has behaved shamefully") This does not reflect any change in the SIS.• No player may introduce a Belief on a character about the character's own behavior.• Traits can migrate from one column to the other in some as-yet-undefined way. This is the only way in which new Beliefs of a character about herself may be created.• When a player chooses to use a Secret trait, their character starts wondering whether this thing is true ("My stars... have I fallen in love with Knightley?") This does reflect a change in the SIS regarding that characters emotions and thoughts.• When a player chooses to use an Image trait, the public becomes partly or wholly convinced of the Trait ("See how kindly Emma cares for others? What a sweet, kind, girl.") This reflects a change in the SIS regarding society.

So, to rephrase Brendan's excellent example of social-contract dysfunction in this new schema:

Player A: "Sorry, I can't help you get into the museum. It's against my moral code."
Player B: "But you have incentive. You just gained the Secret 'Loves Character B'."
Player A: "Never going to use it."
Player B: "Then it will always be there to tempt you... oooh, free dice just for a moment of wondering 'Could I be in love'?"
Player A: "You can not make me go there. But I'll happily add the Secret 'Character A loves me' to Character B. Now you can fantasize about it for extra dice all you want. It will make the eventual let-down all the sweeter to watch."
Player B: "I've got no reason to use it yet... you've given me nothing that could be misinterpreted."
Player A: "True... okay, I've got Secret "Wants to be a bad girl" from Player C. I'd love to get enough supporting evidence to move that into my Belief column. So I will rob the bank with you, and activate that."
Player B: "And I'll activate the one you just gave me, because he thinks that's why you agreed to help."

No player may introduce a Belief on a character about the character's own behavior.

I meant it as the first one (Player B cannot make a Belief on Character A, though they can make a Secret that is temptingly useful to them, and hope that they'll promote it to a Belief).

The second one rocks for actually simulating Emma and other simliar works. I'm awed at what a strong objective role that would make for matchmakers. Player A wants Character A to fall in love with Character B. Player B wants the same thing. Neither of them can create "Secret: Loves B" on Character A. They need Character C to come along and say "They'd be a lovely match" and create the Secret (or perhaps the Image, or both) before they can perform the whole cascading dance of courtship.

But it may be further than I want to go in Dulcimer Hall. Yes, there are ideas too radical and involved for me to jump on them. The difference between "Character C can benefit hugely from jump-starting that romance" and "The romance can only happen if Character C is on the ball" strikes me as something that would create a powerful barrier to the initial process of learning the game through experimentation.

Hmm... maybe a cost-based compromise? Player B can induce the Image or Secret "loves Character B" on Character A, and gains only subjectively--ie no experience or tokens or whatever resource you end up using. Or better, Player B has to spend some of the resource to induce that, since it relates directly to his character.

Meanwhile, if Player C uses an in-game event to induce "loves Character B" on Character A, Character C gains some of the resource. She profits objectively, while A and B (probably) profit subjectively.

Of course, inducing any Secret or Image would have to be justified by / tied to an in-game event, or that puts Player C in an excellent position to game the system. Maybe put a per-session limit on the amount of resources that can be gained by third-party inducement?

Why would encouraging players to game the system be a bad thing? Do you have specific dysfunctional scenarios in mind?

Well, because it's Gamist in what I think is a pretty powerful Narrative system. The scenario I'm thinking of is

Player C: Hmm... if I had twelve more experience I could bump up the die type for my Swing Dances Like A Madman trait.
Player B: Don't look at me. Character B has never danced with you.
Player C: Okay, you just gained the Secrets "loves Character A," "loves Character D," "loves Character E" and "loves NPC 5." I'm also giving you all the same Images, but with "hates" instead of "loves." That gets me... sixteen experience! Great!
Player B: I'm never going to use those!
Player C: Doesn't matter to me. I now Swing Dance Like A Madman 8d12.
Player B: Fine! You just got the Secret "desperately afraid of Character B!"
Player C: You going to pay for that?
Player B: Never mind, you get the Image "nurses a desire to eat Character A alive" instead. That gets me up to Medical Doctor 9d6.
Player A: Whoa, what did I do? I'm spending six experience, and you both get the Image "owes Character A a big wad of money."

Ah... okay. The objection isn't (I think) against the principle of having players game the system (which is only a technique, and not actually associated with any CA), it is against having people game a system that doesn't support the goals of the game.

Yeah... the trick is to make sure that anyone who brutally, cold-heartedly, mechanically exploits the system is doing so by doing precisely what the game wants. In this instance I think the gap would be right here:

Brendan wrote: Player B: I'm never going to use those!
Player C: Doesn't matter to me.

Oh, excellent. The consolidation mechanic really, ah, puts it all together; I'd only add that the consolidating player should gain something (eg a bump in die-type, or resource points) from the consolidation, but I think you probably intended that anyway.

You're right when you say that gaming the system, out of context, isn't necessarily bad--just a tool that can work for or against the interests of the game. I should have been more specific.

This does sound very cool... especially when the distinction between "potential" and "active" traits is kept up. My suggestion would be to make players' piles limited. They take a risk by handing you one of their dice--they can't use as many anymore. They will get a reward once you actually roll that die. In that moment, they reclaim the die and gain experience, and the character who used the trait has now adopted it and made it "active." Limiting the pile lowers the chance of abuse and makes sure you make tough choices on how to distribute traits.

OR--traits are always potential. In the moment that die is used, the potential is used up, but the other player can immediately reinvest the die in the hopes that your character will act in that way again.

In that sense, a player could offer potential traits of hir character for other players to "invest" a die in.

Good points!

I'm going to close this thread, not because of disinterest in your points (which are well worth discussing) but because I've got further-refined ideas being bandied about in [Dulcimer Hall] In Search of Angsty Loner Cypher Guy.