This question is in some ways too large for this (or any other forum), but I thought I'd give it a try anyway:
How does one design a Narrativist game (i.e. a game that encourages a Narrativist mode of play)?
With Simulation based games I know I have to focus on what I'm going to simulate and then look for mechanics that replicate this as closely as possible. I got a decent intuitive graps of dice probabilities and enough experience with some to make a decent guess as to how a certain mechanic would feel like, and I can make decisions on skills and abilities based on the setting of the game and so on and so forth.
When it comes to Narrativist mechanics, I draw a complete blank. For instance, suppose I wanted to make a game inspired by T.H. White's "The once and future king" with the premise "Does might make right?". The setting is going to be medivial but light-hearted, with a decent dose of Disney's "The Sword in the Stone", with talking animals a distinct possibility. In fact, for this particular premise it doesn't matter what level of power someone has, as long as there is someone below and possibly someone above the character.
I can imagine scenes framed by players that address the premise on all levels, and I'd like to encourage spill-over effects, such that one characters decisions affect those on lower levels, forcing them to make decisions themselves or simply deal with the consequences.

I'm absolutely clueless as to what the next step does look like.

I don't know _what_ decisions to make next. How do I decide what skills to use, or how to weigh drama, fortune and karma in the resolution mechanism? Of course, I could copy mechanics from games I'm familiar with, but I wouldn't know why to choose one over the other.
Any ideas on how to find out what questions to ask next?

OK You have a premise already. I would phrase it more as "does might make or serve right?" or perhaps even "Does might make right or right make might?" But I digress.

What you need to do is focus on this and what would bring this out. The bit about power levels not being important is true. Don't focus on that because focusing on that could lead to gamism. Instead worry more about the kind of story you see the players creating and what tools you could provide to aide that.

Another bit of advice. If you're going to write, read often. If you're going to write music, listen to and play other people's music. If you're going to design an RPG, read and, better still, play often. You have to see what's out there. What works and what doesn't

Hello,

Jack's advice is generally correct except for two things.

1. The "could lead to Gamism" comment is entirely, completely wrong. The only issue that concerns Gamism vs. Narrativism, in terms of design, is the reward system (and its integration with all other parts of the system, as reward systems cannot stand alone).

2. The "focus on the story you see the players creating" is a very hard concept for many role-players. Some immediately get enmeshed in The Impossible Thing, as they have to process "focus on it" vs. "someone else creates it," and get all boggled. Others perceive "create story" very much in the sense of GM-style addition and management of the entire Explorative context, and imagine a bunch of GMs all arguing about what happens with anyone playing.

Have you read both "GNS and other matters of role-playing theory" as well as "Narrativism: Story Now"? The second one in particular lays out very specific parameters for historical Narrativist game design. There's even a table which compares a whole bunch of different games in these terms. I'm not sure what I have to add to what I said in these essays, and I'd like to be sure you've processed what's in them before continuing here.

Best,
Ron

"can lead to gamism" in the sense that walking east can lead to China. WHich is why I qualified in in the could should would might, might no sense. My point was that at this point if montag continues to think about elements such as power levels, he'll most likely fall into design habbit which as presented here sounds gamist to me. But montag is interested in stretching a bit into a different direction, which means understanding what he's trying to do that's different from what he had done before so that when he considers items like power levels he won't fall into old habbits. That is, I wasn't saying considerations like power levels are gamist but that considering them at this point may lead him back into design habbits and possible lead to incoherent design in the end.

I apologise for the poor phrasing concerning the power levels. What I meant to say was, that it doesn't matter one way or the other whether a character is a king or a beggar, how strong, clever etc. they are, as long as they have or create some meaningful opportunity to "exercise might" on someone/something else. That sounds safely removed from gamism to me, though I appreciate the concern.
On with the mission:
Ok, so I need to think about how I can facilate adressing premise. ("the kind of story you see the players creating and what tools you could provide to aide that.") Fine.
But when I set out to analyse the games I'm familiar with, I don't always come up with satifying results.
For instance, the mechanics for rerolls and the use of relationships in Trollbabe certainly facilates narration, though I'm not sure, what it does for premise in particular. The mere fact of playing a Trollbabe seems to help more in that respect. The number the character is given OTOH doesn't make any sense to me.
Next: Inspectres: The basic rule system with fortune-in-the-middle does facilate addressing premise mainly through assigning who gets to narrate successes and failures, the Confessional provide opportunity for additional input, but other than that, I don't see anything which actually helps adressing premise. The franchise rules provide various "states" of the business, but beyond basic variations I don't see anything in there.
The Pool seems mainly concerned with narration rights, a basic mechanism for most Narrativist games, but beyond that ...
My life with master seems to make more sense. The setting clearly helps, as do the stats, but I must confess I haven't found the time yet to analyse the mechanics in detail.
In The Riddle of Steel, the SAs and the high degree of risk involved in both sorcery and combat bring premise into focus, and the SAs also help to address/create it.
The Devil in DustDevils helps create character based premise as far as I can see and also helps bring it into sharp focus when the Devil replaces any stat that's gone to zero.
However, in each of these games, the rest of the rules seems arbitrary. Aside from colour I see little reason not to use dice for dustdevils or use some other combat system or setting with TROS. Instead of a Trollbabe any between-the-worlds character would seem to work.
So, again, on what basis are these decisions made?

Maybe it's more helpful towards illustrating my problem if I continue with the example:

Going along with the Disney variant (which I like because it seems to facilate a playful, slightly humorous game), that would mean that:
a) each character must be an animal. Players are encouraged to come up with a type of animal that represents the most important part of their character (a lion if they want to be royality or especially loyal) but "casting against type" (a dumb fox) is permissible.
> This should facilate adressing premise by providing a readily accessible character concept, a thematic link to the fables and by revealing the individual players choice on what aspect of the central premise they'd like to address. Plus, I think it would be fun.
b) the setting will be a blend of the animal kingdom with medivial social structures, transferring aspects of the structure of the food chain onto connections of loyality, inferiority and superiority.
>This strict social structure in the setting should facilate addressing premise in that everyone obviously exercises might on those below them in the food chain/hierarchy.
c) for stats we will need only might, specified according to the means by which it is executed: prowress, rank/social, wit, with the "attacker" specifying what method they want to use, I've no idea what resolution and reward mechanic to use though. At this point I'd say players can freely choose their chracter's overall might, though some balancing might be necessary to encourage choosing a low level.
d) Narration rights: I'd like players to frame scenes for each other, especially since that would give them the opportunity to actively work for the aforementioned spill-over effects, e.g. by framing scenes for others whose consequences are likely to affect their character one way or the other. I've no idea yet on what basis to decide how narration rights pass from player to player.

Hi montag:

montag wrote: Inspectres: The basic rule system with fortune-in-the-middle does facilate addressing premise mainly through assigning who gets to narrate successes and failures, the Confessional provide opportunity for additional input, but other than that, I don't see anything which actually helps adressing premise. The franchise rules provide various "states" of the business, but beyond basic variations I don't see anything in there.

Hello Markus (montag),

With respect, you haven't played any of these games, have you? If you have, which ones?

Because you are looking at the Narrativist issue too broadly. No game, or for that matter, no instance of play can "be Narrativist" in a fully representational sense of the entirety of the mode.

Let's take Trollbabe. The narration mechanics are intended to deal with a source of frustration I've often observed in a particular demographic of role-players, who shall remain unnamed. That frustration specifically lies in seeing their characters' failures in the hands of a person they do not trust to identify with the character, specifically a person belonging to another demographic (also unnamed).

Hence that potential source of frustration is removed. Please note that narration in Trollbabe does not include some of the outstanding powers of Director Stance that you would see in, say, Dust Devils or Universalis. It's simply about having your character fail in a way you find palatable. The game is also based on building more trust as you go between GM and players, so that even the danger of that potential source of frustration is diminished as time goes by.

How does that fit with the other mechanics of the game? (1) Risk assessment; the player decides how much danger the character is in, per conflict. (2) Conflict announcement - people have to recognize that anyone can place their character into conflict, and that they'll have to deal with it; there's no "safety" in Trollbabe in the sense that your character can avoid trouble. (3) The single number: there's no tweaking, strategizing, or layering about the quantitative features of play. It's a statement about your character's ability to succeed, and it's a single statement. The point is to cut to the chase of conflict, not to consider how well you can do in the conflict beyond the initial decision about the number. Note as well that the number may change quite painlessly from session to session, 1 unit at a time. (4) Since changing the number indicates change, but not improvement, actual character effectiveness is only increased via adding relationships and by increasing the overall Scale of play. It's very hard to discuss these issues with someone who hasn't seen them in actual play, but both of these issues are fully integrated with #1-3, in terms of players' emotional connections to the relationship NPCs.

I hope that makes a little sense.

Best,
Ron

montag wrote: For instance, the mechanics for rerolls and the use of relationships in Trollbabe certainly facilates narration, though I'm not sure, what it does for premise in particular. The mere fact of playing a Trollbabe seems to help more in that respect. The number the character is given OTOH doesn't make any sense to me.

Wow, thanks everybody, that helped a lot.
Obviously there's aspects of the respective designs I missed (I wasn't attempting a complete analysis anyway) but the broader problem, as Ron correctly pointed out, is that my focus was/is indeed too broad. In a way, this was the point of my initial question, namely how does one break down this huge concept "Narrativist game" into smaller sub-questions, which can then be addressed or answered by various aspects of the game. Breaking down the question about "the kind of story you see the players creating and what tools you could provide to aide that." is still daunting, but one step closer already.

Summarising what I've understood so far, I'd say a tentative list might include the following.
- What forces players into conflict/addressing premise?
- What unnecessary elements _must_ be removed, so as not to distract from the actual focus of the game?
- How can I make the choices the player makes concerning premise actually meaningful in play (while at the same time providing freedom of choice were premise is not concerned)? What consequences will these choices have in mechanical terms?
- Who narrates what under which circumstances.
I'm sure there's a lot of questions I've missed, but I think I'm getting closer, so again, thanks everybody.

To address some additional questions:
@Ron: I've only gotten to Actually Play InSpectres, Trollbabe is next on the list.
I don't get the "certain demographics" stuff you're talking about, but wanted to say thanks for explaining the number-thing.
@Michael: I noticed that (and plain forgot about it when posting) but didn't realise its importance for premise. Thanks.
@ WDFlores: thanks for explaining the franchise stuff in greater detail. I just re-read the rules and noticed I also missed the teamwork rules, which also help premise by allowing for a degree of self sacrifice.

markus

Hi Markus,

A better way to look at the focus of most Narrativist aimed designs, is this:

-Players MUST have some form of input into "what happens" in order to address premise, which can range from balls out everyone has diretor stance power such as Universalis to player proposes, GM narrates of traditional rpgs.

-There cannot be a predetermined story/outcome(this precludes player input, above)

-There must be some impetus, either from the group itself, or the text to suggest to the group, of addressing premise.

That's all.

Trollbabe pushes for premise because you are constantly weighing the value of a relationship to success. Sorcerer forces you to weigh the context of your actions in light of Humanity. Riddle of Steel makes you define the issues at hand through Spiritual Attributes, then focrces you to decide which are more important, and HOW you intend on using them. Heroquest forces the players to affirm, deny, or reinterpret the tradtions and expectations of the setting through the heroes' actions. Dust Devils puts premise smack dab in its Devil Mechanic and begs for resolution.

As you can see, there are many ways to put that impetus in there, but the key point is making it possible through system and techniques.

Chris