« Revenge of the Jedi Play Report | Main | Almost About Roleplaying Games (part 3) »
January 13, 2005
Narrative Control
First off, a basic principle: narrative control comes from consensus of the players. This is probably just an application of the Lumpley Principle. It is not an instantaneous action which is subsequently fixed in stone; it's a constant process. There can be and often is fluctuation in the amount of control during the course of play. Sometimes we notice it, and sometimes we don't.
Now on to the fun stuff. Some of it is fairly basic but I want to lay out my assumptions.
Buckets
For the sake of argument, there are three common buckets into which control can be placed. The first is the players themselves. The traditional implementation of this is "you say what your character does." In this implementation, a given player has narrative control over most of the actions a specific character takes. The player may or may not control the narrative outcome of an action; consider the difference between "my character sits in a chair" and "my character sits in a chair which is delicately balanced over the Abyss, without falling in."
There are also plenty of examples of player narrative control which extends beyond the actions taken by a specific character. We won't catalog all the possibilities; let's just nod to troupe play a la Ars Magica, narrative control as implemented in Trollbabe, and so on.
Second is the GM. This is a specialized case of player narrative control that wouldn't warrant mention except that the GM/player split is a strongly canonized element of roleplaying. Whether or not there's a good underlying reason for this is a question way too big for this little post, so I'm just going to say "the GM typically has massive amounts of narrative control delegated to him or her" and leave it at that.
Third is the rules system. This bucket has the same validity and credibility as either of the other two buckets. Narrative control is granted only by consensus of the players as a whole, and it is only credible as long as the players as a whole continue to accept the outcomes dictated by the system.
I think it's useful to split system narrative control into two categories. Passive system narrative control is the things you can't do. You see this a lot in character generation; "No, you cannot make a guy who's really good at hitting things with swords and casting fireballs." This tends to be fairly light narrative control -- even if a D&D fighter doesn't have the Use Magic Device skill, the player can still narrate the attempt to use a magic wand -- but it does determine some possible outcomes. Active system narrative control is what happens when you roll dice and the outcome of the dice determines what happens. (For dice, read also cards, coins, and yarrow sticks.)
Again, these are not hard and fast categories; a GM is just a special case of player, and all narrative control springs from player consensus. In some ways, in fact, these categories are illusory. More on that later.
Degrees
Narrative control is not an all or nothing deal, even when it comes down to single actions. You narrate that your character attempts to hit a foe with a sword in an old school traditional game like Hero. Dice are rolled; the outcome of the dice determines -- because you as a group have agreed to allow it to determine -- whether or not the sword connects. The GM maybe describes how the hit looks. Maybe you describe some of how it looks. Maybe the GM describes the foe's immediate reaction.
In Primetime Adventures, the player determines the nature of the conflict: "I want to kill the foe," or "I want to scar the foe so he'll always remember me," or "I want to get past the foe and disarm the bomb." The dice determine which outcome results. The player with the single highest die roll narrates the outcome of the conflict. Maybe other players or the GM chips in ideas. The ultimate outcome of the conflict is determined by several different factors.
So What?
Well... let's see if the assumptions catalogued above seem generally valid before we take that next step. (Yes, it's taking me weeks to get this whole post written.) But, you know, if someone wants to say "look, dice can't ever have real narrative control," I want to have that discussion before confusing the matter with a lot of conclusions.
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8342026ca53ef00e5501e03ae8833
Listed below are links to weblogs that reference Narrative Control:
» Narrative vs Free-Form: 2-Axis Analysis from Elements
Periodically in conversations with Rebecca I find myself in an argument about narrative games versus free-form games. Inevitably I come down on the side of narration, because I like to tell stories and consider myself the one telling the story, with p... [Read More]
Tracked on Jan 18, 2005 2:35:51 PM
» Fighting Pavin hangs in - FOX SPORTS from holds a two-shot
Pavin hangs in FOX SPORTS- 5 hours ago COREY Pavin, chasing his first PGA Tour [Read More]
Tracked on Jul 31, 2006 11:22:24 PM
» GEEK: E3 Cancelled? from show, which it
access is one of the problems, with the onslaught of Web 2.0 opinion happy bloggers, [Read More]
Tracked on Aug 7, 2006 11:29:06 PM
» Oil Prices Rise After Refinery Snags from in domestic oil
domestic oil refineries reignited fears of short oil supplies during the summer season. Earlier, prices [Read More]
Tracked on Aug 8, 2006 11:40:21 AM
Comments
Sign me up for "dice can't ever have real narrative control."
The only case I know of where the rules contribute directly to the imagined events in the game is Rolemaster's crits, and then only if somebody reads them out loud and that's what happens, unedited. Otherwise, rules provide constraints on the players' (including the GM as a player) narrative control, but have no narrative control themselves.
Like this: narrative control means asserting and negotiating "this is what happens." Rules provide "these sorts of things can happen, this is what should happen, that particular thing oughtn't happen." Can, should, ought - what does happen is always in the hands of the people playing.
Posted by: Vincent at Jan 14, 2005 9:21:32 AM
Oh and otherwise, right on and I'm psyched to see where you're going with it.
Posted by: Vincent at Jan 14, 2005 9:22:11 AM
The other thing that rules do is apportion narrative control. They are the agreements among the group about who gets to make creative contributions, they don't make the contributions themselves. They can at best constrain, as Vincent mentioned.
If I swing at you with my sword and roll against my agility to see if I hit you, the die roll decides which of our desired outcomes happens. Some person has to narrate what exactly happens. The dice just point to who gets to make it up.
If that's part of your definition of narrative control, then all good. But it's problematic to say that your third bucket has the same narrative control as the first two.
My other comment would be that, as you indicate, the second bucket is a fancy souped-up version of the first, which, of course, is created by application of the third bucket.
Posted by: Emily Care at Jan 14, 2005 10:47:44 AM
If I swing at you with my sword and roll against my agility to see if I hit you, the die roll decides which of our desired outcomes happens. Some person has to narrate what exactly happens. The dice just point to who gets to make it up.
Ah, see, I think there's a level of granularity here. In a straight up game of D&D, if I roll the dice and I hit you, I do not have the option to go back and say "Hm, no, I didn't hit you." I can't pull my damage. In the story, it is unalterable fact that I dealt 10 hit points to the goblin.
Now, that's hardly all of the narrative. I can describe the hit pretty much however I want within the social contract of the group. The color surrounding the result is up to the players. But -- and maybe I'm confused here -- how is the amount of damage done not part of the narrative?
(Side note: I am wholly sympathetic to the stance that one might not /want/ to let the dice determine this, and if someone says "I don't like D&D because the dice control too much of the narrative," I'd say "Yeah, I can totally see that." Part of why I've been thinking of the randomizers as a source of narrative control is precisely because of that line of argument.)
Posted by: Bryant at Jan 14, 2005 11:35:43 AM
Vincent: I have to side with Bryant here. The rules/dice, the third bucket, do have some element of narrative control. If I the PC roll a natural 20, I hit your NPC. It says so in the book. Not I should hit you, or I ought to hit you, I hit you. It says so right in the SRD. You the GM can't say no, I miss anyway, or we aren't playing the same game any more; if you do you're robbing me of the power our social contract (to abide by the rules) grants me, the power to hit on a natural 20.
Posted by: Jeff at Jan 14, 2005 11:45:31 AM
Ah, see, I think there's a level of granularity here. In a straight up game of D&D, if I roll the dice and I hit you, I do not have the option to go back and say "Hm, no, I didn't hit you." I can't pull my damage. In the story, it is unalterable fact that I dealt 10 hit points to the goblin.
It's easily altered: the players can easily say, "We never went to that dungeon" and ret-con it out of existence. Poof. It's gone.
Posted by: Neel Krishnaswami at Jan 14, 2005 11:47:16 AM
It's easily altered: the players can easily say, "We never went to that dungeon" and ret-con it out of existence. Poof. It's gone.
If that constitutes lack of narrative control, then nobody has any narrative control, no? Anything can be retconned, no matter how it was determined.
Posted by: Bryant at Jan 14, 2005 11:48:16 AM
Yes, I think that's basically right. Here's a real example:
At one time, I didn't like mecha anime. A friend of mine did, and wanted to run a Gundam-esque game. Because he was my friend and a really good GM, I agreed to play, and mentally substituted "space fighter" whenever anyone said "mecha", and otherewise played normally and had a really great time.
Question: Did mecha exist in the game?
Posted by: Neel Krishnaswami at Jan 14, 2005 11:55:54 AM
Dice don't have narrative control, but they provide (or should I say "force"?) narrative limitations when they determine details, thus removing narrative control from others. That is, they place limits on what players can tell as the story when they decide success or failure, reducing narrative freedom. It isn't the same thing as having narrative control themselves, but it does reduce the narrative control others have.
In other words, they eliminate certain results from the possible range, perhaps even limiting you to a fairly narrow range. For example, a critical failure can say you drop your weapon and it lands in a certain spot (probably based on another die roll). There's a range of ways to tell that story (my grip slips and...; my opponent's quick response disarms me and...; overextended from my thrust, my opponent's arm smacks my forearm and...;...my sword lands point-down in the sand, quivering; ...my sword skitters across the stone, spinning to end land with the grip away from me, damn it; etc.), certainly. But the dice have taken away a significant degree of narrative control (eliminating the descriptions where the sword ended up somewhere else, for example), though they do nothing with it (that is, they tell no story themselves).
In a sense, systems that do this serve to force the narrative into certain repeating patterns, a sort of "systematic cliché" that less detail-determining rules systems would avoid. This is particularly true of systems with a small range of special results (all critical failures in combat mean weapon dropped, for example), but that's just calling attention to an unusually precise example of the normal behavior of system, which is to provide structure and guidance for the narrative (by describing character abilities, determining success/failure, etc.).
Posted by: Ghoul at Jan 14, 2005 1:17:29 PM
Bryant, Jeff: in practical terms, we agree, of course.
If you're going to be drawing theoretical conclusions based on the dice having the same kind of narrative control that people have, though, practical terms aren't good enough. When the group says "the dice say you hit, so you do," there are two very different things happening: the dice are providing an event and the players are making it so. Is this practically the same as the dice making it so? Sure. Is it really the same? Not at all.
It's not a matter of taste or wanting - I'm for following the dice. And I really don't think it matters at all, unless you're building a theoretical case. Bryant, does whatever you're building toward really depend on the dice having a voice at the table?
Posted by: Vincent at Jan 14, 2005 2:35:07 PM
Hmmmm....
Let's say we're playing a role-playing game. There's 5 of us.
3 players, a GM and Bob.
Whenever the players want to do something (hit a guy, find a clue, study a spell), they say: "Hey Bob, can I do this?" And maybe "I really think I should because of X" or "This will be really tricky because of Y".
And Bob thinks about it and says "Yeah, you do it" or "No you don't". The answers aren't totally random, if you give a really good reason why you should or shouldn't be able to do something, Bob generally agrees with you. But it's not a given.
The GM also has to ask Bob about a number of different things.
"Hey Bob, is it raining?", "Hey Bob, how many orcs are there?", "Hey Bob, did the dragon spot the thief?", and so on.
And Bob doles out the answers.
Sometimes Bob gives very specific responses and sometimes Bob just says "Yes" or "No". Sometimes the player interpets Bob's answers and sometimes Bob has the final say.
Does Bob have any narrative control?
(I now have a crazy desire to write a mini-game where the Fortune mechanic is another player.)
Posted by: Tom at Jan 14, 2005 3:16:47 PM
"Sometimes the player interpets Bob's answers and sometimes Bob has the final say."
Sometimes Bob doesn't have narrative control and sometimes he does.
Posted by: Vincent at Jan 14, 2005 3:43:31 PM
Tom--Why does what Bob says matter? What if nobody ever asked Bob anything? What if everyone ignored what he said?
If he was a person, he could argue in favor of his input. How do dice argue? They can't. Some person has to invoke them. How can rules assert something in play? They can't. Someone has to refer to them.
People use rules & mechanics to back up their statements. They use them to constrain what other people say. But it's always a person who brings a mechanic into play, or describes what the game text says.
Posted by: Emily Care at Jan 14, 2005 3:45:52 PM
I think that the point is that the rules and the GM are methods by which the players may choose to mediate their control of the narrative. That is, you roll the dice, and they combine with the rules to say "You hit." Of course, what actually says that you hit is that all of the players reach consensus that you hit. The rules and the dice provide an influence on that reaching that consensus. In some games (a D&D dungeon crawl with 0 character immmunity), the step from "Dice say" to "We say" is very small and very few. In other games (low-mechanics "High roll good, Low roll bad" games), the steps from "Dice say" to "We say" are many and long.
The same holds for the transition from "GM says" to "We say."
It also holds true for the transition from "I, the player, say" to "We say."
That is what I take Bryant to be saying. The buckets don't control the narrative, the buckets are what the players can agree to use to control the narrative.
Anyway, a tight constraint on the narrative is a control on the narrative.
Posted by: Charles at Jan 14, 2005 4:17:03 PM
This seems like a really piddling point to be arguing about; as Vincent observed, I think we're all basically in agreement and talking in circles.
But Emily: when we sat down to play we agreed that Bob's decisions would have weight. We ceded some control to Bob.
Consider, say, the Lone Wolf adventure books, the solitare game ones. It's an analogous case: there are three contributors to what happens. There's the author, the GM, who declares that if you use Huntsmastery you can turn to section 41 and read that the relcher you're tracking went left at the fork. There's the player, who decides whether to go left or right at the fork and whether to take Huntsmastery or Weaponmastery. And there's the rules system, which decides whether the player survives the combat with the relcher.
Or say that we're playing D&D. The GM is using a Dungeon magazine adventure in which it's written that a particular doorknob is trapped and will disintegrate anyone who touches it if they fail a saving throw. A player, not knowing the doorknob is trapped, touches it. You've triggered a disintegration trap, the GM says. Make a Fort save. The player rolls a 1. Everyone around the table winces as they imagine the player's character getting turned to dust, and then the party's trap-disarmer goes to work on the doorknob. Who was controlling the narrative when the player's character was disintegrated?
Posted by: Jeff at Jan 14, 2005 4:23:59 PM
"The buckets don't control the narrative, the buckets are what the players can agree to use to control the narrative."
Ah. This I agree with.
Posted by: Vincent at Jan 14, 2005 4:24:34 PM
This is awesome. I agree also with that. It's all player choice in the end. That one sentence is exactly the conclusion I've been straining towards, put just how I wanted to put it; I just haven't been able to figure it out.
Posted by: Bryant at Jan 14, 2005 4:27:26 PM
From the original thread about credibility and assent:
- Nothing's true in the game until all the players agree that it's true.
- System, mechanics, GMs and so on are all just ways to get the players to all agree that things are true.
Posted by: Emily Care at Jan 14, 2005 4:35:44 PM
I've put a piece up on my blog about how RPGs' rules work and what's the relationship between "the narrative" and "the hit points." It has silly animated gifs.
Posted by: Vincent at Jan 19, 2005 10:25:01 AM
Hey Vincent, make it a link on the front page -- it's good stuff. :)
Posted by: Neel Krishnaswami at Jan 19, 2005 11:20:03 AM
Or just cross post the whole thing. We want silly animated GIFs too!
And see also Robin Laws' recent column, Make It A Gimme, on how we routinely let dice do what no player should do.
Posted by: Rob at Jan 19, 2005 3:51:22 PM
This thread started with the idea that the three buckets (players, GM and rules) control the narrative. With due credit to Bob (which by the way I find to be a fascinating idea - I wouldn't mind playing this role in games) in RPGs the players turn to the GM for affirmation of all their actions. Game Master is an understatement. The negociation between player and GM is just that, but the GM has the final say. The rules give the players some power to say NOOOOO! But in the end the GM can just say no. The players then have to decide whither they want to lump it or move to Canada.
What I don't like about this game structure is how much work this lays on the GM. The GM's part of the social contract is to make a fun game. The players are guided through to maze. They can negociate a change here and there but not much or the GMs work is ruined and the GM needs hospitalization for stress.
I've done Matrix Games for years (see a previous write up on these games in 20x20 room) In Matrix Games the power relationship is different. The "Referee" reads off a story opening to the players just like in an RPG. The players then look at lists of characters (both good and bad) and decide who they want to champion. Unlike in an RPG players do not have to be on the same side. Now they are ready to play.
The players have a BIG BUCKET. They say what they want to have happen next in the game. These arguments for action can be about anything (their character, other characters, the weather, even to change the rules of the game.) The "Referee" decides what players have to roll for their arguments to happen and a few other things (which trigger trouble saving arguments, conflict resolution arguments and which arguments logically compete). The referee can give the players impossible rolls - "Roll six sixes in a row!" This can effectively veto an action but the ref can not then jump in and say what really happens - as a GM can.
The power is divided differently -
The players create
The referee performs technocratic functions but can also veto stupid arguments (the final line of defense against power gamers.)
The rules - well...they only tell the referee to make a few rulings on arguments and ask the players to make things up - so they don't have power. They are more like a vehicle you can drive where you want it rather than a raod you have to travel.
This shift in power makes the players responsible for their own game. The referee can do a little guiding and some editing but the players control the world. The players are the Game Masters.
This method work very well. I've used it since 1988 and now you can find traces of it appearing in lots of independent games. BIG BUCKET, small bucket (that can knock the big bucket over), and bucket with no bottem that water can be directed through.
Posted by: Chris Engle at Feb 2, 2005 4:53:53 PM
mnosbercad.com The plot emerges through Carson's meditative, elusive
http://gis-spatial.osbercad.com/gis-spatial.html gis spatial fragments, mysteriously isolated couplets, excerpts
http://kaart.osbercad.com/kaart.html kaart from versified conversations and letters, interior
http://us-geological-survey.osbercad.com/us-geological-survey.html us geological survey monologues and (as Carson's readers have come to
http://gis-clearinghouse.osbercad.com/gis-clearinghouse.html gis clearinghouse expect) digressions on matters of classical scholarship.
http://colima.osbercad.com/colima.html colima This kind of thing is imitated badly and often by
http://open-gis-consortium.osbercad.com/open-gis-consortium.html open gis consortium others, but Carson's phraseology within poems remains
http://the-world.osbercad.com/the-world.html the world her own: "Rotate the husband and expose a hidden
http://jobs-gis.osbercad.com/jobs-gis.html jobs gis side," she urges early on
Posted by: flag at Jul 13, 2005 7:13:36 PM
