Forum changes: Editing of posts has been turned off until further notice.

Main Menu

Need help with Aware "narrative" mechanic.

Started by Demonspahn, November 25, 2002, 12:08:36 AM

Previous topic - Next topic


Hi everyone.  Just thought I'd post this, looking for some fresh new ideas.  I'm talking about my game Aware, where you play intelligent animals.   I'm looking for what I can best describe as a controlled narrative mechanic for the game.

Let me try to explain:  I know what sort of narration I don't want---I don't want the players narrating for other player's characters and I do not want the players to be able to significantly (detrimentally?) alter the overall continuity of the current story/adventure.

Task resolution is based on a check, with standard chances for success or failure.  I want the players to be able to narrate at least the successes to explain why an animal such as a chicken can open a locked metal door---as of now this depends largely on which stat is used (the player chooses).  Frex using Might, the player could narrate that he pecked away at the lock with a hairpin, using Wits, he might have started a human vehicle and managed to somehow drive it through the door or using Bearing he might have coerced a larger animal such as a bear to batter the door down for him.

Now, this sounds good (at least to me) in theory and keeps things pretty open and freeform as long as the desired outcome of the task is already known---the PC wants to open a door; he wants to climb a tree; he wants to swim a river; he wants to locate a reference to a pesticide in a human library, etc.  

Where it "breaks down" is when the player doesn't know what would result from the use of a successful task.

I'll take an example from play.  The fox and the clam were looking for a dog's favorite bone (which happened to be the thigh bone of his cruel master).  They were looking for anything out of the ordinary in the general area of where the bone was hidden so I had them make (perception) checks.  The clam succeeded and the fox failed.  Now, given that I would ideally like the players to narrate their own successes and failures, the game came to a crashing halt and I'll explain why.  

The fox could of course just snuffle through the brush and not find anything, which is boring but at least not too problematic.  The clam on the other hand did not know the bone was there, so how was the player supposed to narrate his "success" at finding the bone, much less make it relatively interesting narration?

I know that I could just say "you found the bone" and let the clam's player narrate this, but this also seems somewhat contrived and limiting (although right now that is the mechanic I am leaning toward, which unfortunately limits narration only to certain situations).   What I'm looking for ideally is some sort of guidelines to make that unecessary so as not to disrupt the flow of the game or of the overall story.

I want to incorporate guidelines to avoid situations where the object of the story is to say find a bone, and the player immediately makes a perception roll, succeeds and narrates that he found the bone under a nearby rock, when I had planned for it to be hidden in a cave guarded by a cougar.  Similarly, I want to avoid just using mega high target numbers when the character is looking for the bone in the wrong place because that would lead to player knowledge of where the bone is based on fluctating TN.  I suppose I could just disallow rolls when they are not in the correct area, but that leads to a whole slew of other problems, besides player knowledge again, including the ability to turn up clues in other locations.
And lastly I want to avoid the player suggesting a narration, with the GM having the chance to veto it since I am not sure but it seems like this would bog the game down a bit more than I would like.  

So, in summary, and I'll try to ramble a bit less here, I want the players to be able to narrate success and failure, in a way that is interesting, yet with guidelines so that it does not alter the current story, but in a way that possibly leads to other story/adventures either now or later.  

Damn, I hope this makes some sense.  I think it basically comes down to how one would concisely phrase the kind of guidelines I am looking for, but I might be missing a lot more than that.   Any thoughts or feedback would certainly be appreciated.  

(who's starting to think maybe he ought to just make this game, Aware d20)

Ps - that was a joke.  :)

M. J. Young

I'm not sure, but I think part of the problem is that you're trying to fuse player narration to a gamist situation. It's a bit like playing Battleship, and letting one player say, "I'll shoot B5, which is a hit, then B6, another hit, B7 is also a hit, and then with B8 I sink your battleship." You want the player to be able to narrate, but you don't want to give him the information on which he can base the narration.

How to fix it? That depends on which way you want to go.

    [*]You can create the player/character dichotomy that works well for narrativist games, and tell the players at the beginning of play that the story is about searching for a bone which no one knows is hidden in a nearby cave. Then when the clam makes his roll, the player already knows that the bone isn't here because he's in the wrong place. He would then know to narrate something like "I see something! There, under the leaves, there old umbrella, not the bone. Sorry, got excited." Down side of course is that the players might have the feeling of creating the story, but they don't have the feeling of finding the bone--they knew where it was all along. It also requires dedication to the narrative to avoid making an excuse to go right to the cave, and a sense of good story to know when it's time to stop searching all the places it isn't and go to where it is.
    [*]You can provide the players with the information they need when they make the roll, e.g., "The clam is certain that the bone is not here," and let them take the narration from there. This is something less narrativist, a bit more gamist, but still gives them the ability to narrate effectively. In essence, it says that a successful perception check in the wrong place will result in perceiving that it's the wrong place.
    [*]You can not tell them what the target numbers are or whether they were successful in the roll, merely letting them know what they rolled and whether they found anything. This is the most gamist; it may encourage them to search the same area again if they think they may have rolled poorly (and you may wish to allow that--if they both roll poorly in the cave, they might never find the bone). They could still narrate what they're doing based on what you tell them, but not have any idea of whether they have been ineffective searchers or are looking in the wrong place.[/list:u]
    As I say, it depends what you want to accomplish.

    --M. J. Young


    Hi M.J.,

    Thanks for the reply.  I think your #2 bullet is more what I'm leaning toward right now although ideally I would like to somehow get an (impossible?) combination of #1 and #2.  Still, it is starting to seem like a square peg/round hole dilemma.  I see I'll have to think some more on this, but thanks.


    Mike Holmes

    Yep, I've run into the same problem in my Synthesis playtest and elsewhere. I'm glad you brought the issue up.

    A common Narrativist solution (the one Josh would like me to implement, for example) is just not to have perception checks. That is, if there is information that the GM want's players to have, he should just narrate them finding it. If the GM doesn't want the players to know something, he should just not tell them, or even allow them to roll.

    Perception checks are somewhat Simulationist. In that the usual rationale for why you are having the check is "well, the character would have a chance". This sort of Cause and Effect thinking is Sim, and doesn't support Narrativism (and as MJ, says, it's Gamist to the extent that the player wants to know in order to advance his character's protagonism).

    This all said,  I like the idea of using the randomness of a Sim result as a catalyst for story creation. After all, all resolution rolls contain this element, do they not? If a resolution system can produce more than one result, either it's anti-thetical to Narrativism (in that it might produce a result that's bad for the story), or we can see our way to using the outcome as inspiration from which to further the story.

    If we go with the last assumption, then there is nothing particularly about perception checks that are more or less anti-narrativist than any other sort of resolution. As such, I suggest that it would be cool to find a way to overcome the problematic nature of these sorts of resolutions.

    The most powerful solution to me is to use the Dunjon approach. That is, allow the player to narrate anything. Then the GM just changes the backstory to fit the player's narration. If the player is successful, they can narrate the bone being there. Even if it was "supposed" to be somehwere else to begin with. This requires a lot of flexibility on the part of the GM, but it's not impossible.

    But I suspect that's not what you're looking for.

    Let's look at MJ's solutions. The first bullet is problematic. Essentially, you have to tell the entire backstory before you play it out. This could be laborious, and might canalize play. "Playing before you play" as it were. So I'd avoid that wholesale.

    The second bullet is OK, but as noted, means that the GM can in certain sorts of circumstances take away the player's creative authority in order to protect the "reality" of the world. Which dampens the ability in these cases.

    The last bullet is very sim. If you did go this way, I would suggest conflict resolution such that a failed roll meant that the PCs had exhausted their ability to persue that rout of advance. But I fail to see how any player power is maintained in this circumstance.

    I agree that a combination of one and two would be cool. And I think I have a way. Note how a player who knows something of the world is required to narrate so as to take those facts into account. This is an unstated truth, but an important one. That is, if PC Bob encounters a troll, and it's determined that the troll is always angry, then Bob can't in future resolutions ignore the fact that the troll is always angry. This is why bullet one would be effective; the player can't ignore what has been established as fact. But how do we establish fact in a cool manner, and only for the scene in question? So that we don't have to "pre-tell" the entire backstory?

    Cut-scenes. I haven't tried this yet, but I think it would be effective. Have you ever played Final Fantasy? Every once in a while, they cut to something that the characters are yet to come across, and show the secret thing as it happens. Thus, when the characters arrive on the site of that event, when the secret is uncovered there is some context for it.

    Well, similarly, in tabletop, one could use a cut scene as a way to inform the players about what information they need to know for the scene, but that the characters do not know. In the example, just prior to cutting to the scene where the bone is, the GM cuts first to a scene in which he describes it being burried. Then the PCs arrive on the scene, and either make or do not make the discovery of the bone. If they do, the description of the bone being burried can be used in their description.

    In this case, the Player would be limited in all cases of perception where there was no cut scene. That is, the player is free to describe anything he wants in a perception resolution except finding the bone in question. Knowing that there was no scene of it being buried in town, he is not free to make up information that indicates that it was.

    If one times these things well, it can make the whole thing much easier. If we want to establish as "reality" for purposes of the game that the bone is in the cave, I'd do a cut-scene just following the expository scene in which the quest for the bone is revealed. Sure that means that the players could play Pawn mode, and just go straight there, but we're assuming a system that doesn't promote this to start. If it's at all problematic, then putting the cut scene just prior to the scene where the information might come up can work.

    Does that make any sense?

    Member of Indie Netgaming
    -Get your indie game fix online.

    Christoffer Lernö

    For me personally after playing Aware, I'd be happy to see you taking a Donjon-ish approach Pete, but I'm also aware that this might not quite be what you want.

    The cut-scenes Mike talks about would work but of course they require some restructuring to firmly base them in the system. The advantage is that it both solves the problem as well as adding a little extra to the game. And maybe it can be used to steer the game towards that dark gritty feeling you wanted for the game?

    Actually visualizing these cut-scenes in play gets me pretty excited. If it wasn't that Ygg is heading clearly in a different direction I'd be tempted to try it out myself.

    "You happily set out to look for the Pir the Mouse, whistling a little as you go along. Meanwhile back at the creek, a dark shape floating upstream lodges a long gnawed on thing under a ledge in the creek. Another shape swims past and for a second them seem like they talk about something. Then one of the shapes dive down into the water and everything becomes quiet."
    formerly Pale Fire
    [Yggdrasil (in progress) | The Evil (v1.2)]
    Ranked #1005 in meaningful posts
    Indie-Netgaming member


    This is an interesting dilemma, and one I have no idea of how to solve. Mike's idea may be the closest to a solution at this point.  Still, I am giggling at the idea of a clam chilling with a fox, looking for bones and stuff. :) Good image, I guess...

    Without knowledge sharing there, this is an impossible question to answer. At some point, you will have to tell the players that the bone is not here. Unless you do something even different, where you require that players leave all action descriptions at a cliffhanger... i.e...

    Player: "My clam shuts up tightly and stretches out with his feelings, listening to the sounds of the earth. If the bone is here, he will find it. After a few minutes of searching, he... [pause]"

    GM: ".. does not find the bone."

    Player: "'Well, hrmph... It is not here." says the Clam to the Fox."

    The other key thing might be that if an action is successful but the intended outcome is not reasonable (the bone is not in the room), then perhaps another clue should be rewarded in, rather than just a failure?


    GM: ".. does not find the bone, although you become aware of an aging sheepdog watching the two of you from a cliff overhead."

    Player: "'Well, hrmph... No bone here, but don't look now. Old Roger is watching us from above. Perhaps he knows where the bone is. He has a thing for bones...' Says the Clam to the Fox."

    Or some such.

    Well, good luck. This is pretty interesting.

    Nathan Hill
    Serving imagination since '99
    Eldritch Ass Kicking:

    Mike Holmes

    Another option that we're all overlooking is the idea of allowing the participant that has the information to do the narration. Yes, the GM (per such games as Sorcerer, Hero Wars, and D&D). I know, radical thought.

    I guess an important question is why you want to allow the player to have the right to narrate successes? This seems to be a current trend. For example, I thought it would be a good idea for Synthesis. But now that I look back, I'm not so sure.

    The reason I included it was that it seems to me that the best way to ensure that a character is protagonized by the description of a result. But a GM can do a reasonable job. Truth be told, it's just because I'm lazy. Or, to be more compimentary to myself, I wanted to lighten my narrative load so as to be more effective elsewhere.

    Also it seems fun for players to narrate their successes.

    But what's wrong with just saying that for "Uncertain" checks (to use the Megatraveller task delineation terminology), the GM narrates the outcome. Considering that these are often not particularly dramatic, it seems little to give up. And you can even keep the player in the loop using this: "Bob looks all about, but fails to note the bone wedged under the log." The player is updated, the event is narrated, and play can continue.

    This can be formalized. You can state that a player can request that he narrate, or the GM can request that the player narrate. In effect, this is what I've been doing in play lately. In Hero Wars, for example, there is no player narration rule. But I just started saying, at some point, "Hey, Julie, tel us what that success looks like." Or we even do it as a group discussion at times.

    So what I'm thinking lately is that the job should technically be the GMs responsibility. He either narrates himself, or assigns somebody to narrate. Hmm. This gives me a neat idea. That means that once in a while, the GM can assign player A to narrate the result of player B's successes.

    Note that one can also apply these ideas to failures as well. I've been refering to successes, simply because that's how Synthesis deals with it (based on the original idea form The Pool, of course).

    Member of Indie Netgaming
    -Get your indie game fix online.


    Wow, thanks for all the feedback guys.  It really helps to get the mind working.  OK, I've been mulling these posts over for a few hours and here's what I've come up with.


    I _love_ the idea of cut scenes.  The only thing is, I can see them being just as problematic in some cases.  In the bone example, the GM and the players knew what they were looking for (the bone) so a cut scene would work well to move the story forward.  However, what if say the players knew there was a mouse in the area and were looking for his hidden hole?   Or what if they just knew that something had happened in an area and were looking for clues?  A cut scene for each incidental roll like this would seem to detract from the cut scene mechanic (which I think is an extremely powerful and handy tool, having played enough video games and hell, seen enough movies that demonstrate this well).  Then again it might not, if done well.  What do you think?

    I don't want to just leave out perception rolls all together at this point because off the top of my head I can envision situations (such as an ambush) where the PCs might or might not have a chance of detecting something.  Sounds awfully Sim I know, but as a player (and this being a game I want to play), I dont care for flat dictation on what or what I don't see.  

    As to the reason behind player narration, it is a matter of protagonizing the players, but also because I wanted to do something a little different than just the basic cause and effect rolls (that I'm more familiar with) and because, like you said, it's just seems more fun.


    I think I like the idea of cut scenes for Aware as well.  I just have to figure a way to integrate them, but I'm almost certain I'm going to include them somehow.


    Your "roger the sheepdog" example sort of touches on an idea I have been kicking around---that of letting the outcome of any roll add something more to the story/adventure.  I think I may have figured out something that might work so let me try to explain and see what you guys think.

    What if (I'll continue to use the dog example here) the characters find the bone regardless of failure or success on the perception check.  All the check does is determine the events surrounding to the actual discovery/recovery of the bone.  

    So, regardless of the result of the check (as long as the PCs are searching in the right area) the GM would say "you found the bone wedged under the bank".  

    If the check was successful, the character not only finds the bone, but also must narrate a Reward (as per Aware's story elements).  A Reward is usually something simple, like a small item, the meeting of a friendly NPC, maybe an easier chance of doing something for the scene or perhaps the use of an ability (or the use of a new ability/skill) that allows you to advance that ability after the story/adventure is ended, provided you spend the (XP).

    On the other hand, if the check is a failure, the character still finds the bone, but would also have to introduce a complication, above and beyond any complications the GM has already planned.

    So in the bone example, both the clam and the fox make perception checks while on the bank of the creek.  The clam successfully makes his check and the fox fails his.  The GM says that the bone is wedged firmly under the bank.  

    The clam's player then narrates his Reward (since his check was successful) which in this case is that he spots a place where water has eroded the bank, making the bone somewhat easier to dislodge.  

    The fox player (who failed his check) then narrates that if the bone comes away, it will dislodge some large rocks making retrieving the bone somewhat dangerous.  Or he narrates that an alligator is lying on the bank near the bone, or that the water current is very swift and anyone who enters risks getting swept downstream, etc.

    This would also cover generic rolls.  If the PC was looking for a mouse hole, he finds it automatically however, a success means the mouse is inside, a failure means a cat is lurking nearby and the mouse is afraid to come out or talk to anyone, etc.

    How does something like that sound?  It's still rough I know but it seems like it might work in theory.