Realistic Consequences vs Gameplay

[Fenris-77]

Yup, and there are games where it seems to me as though the players are trading some amount of agency and authority over their characters in exchange for some amount of at least narrative authority over the world. As in, the player gets some limited direct say in the world; in exchange, the GM gets some limited direct control over the PC. I've been told that's not correct, but that's the way the rules read to me (which is why I'm being clear this is my subjective reaction to/analysis of those rules).

It's not really an exchange, but if you're talking about games like FATE then both those features often tend to occur in the same games, sure. There's a pretty wide range of what player-facing narrative control actually looks like from game to game, and the presence of 'compels', to use the FATE term, isn't always a core mechanic of those games. People get really hung up on the difference between narrative authority generally and narrative authority over mechanical outcomes, but I'd submit that both index player agency to one degree or another (again, depending on the game). Some games also divorce the mechanical result more or less from the narration of the result, often based on the level of abstraction in the mechanic itself (by design).

GM control over the character is kind of a separate issue, and the phrase 'control over' has some implications that aren't really warranted in most cases, especially in how people who aren't used to those mechanics react to their presence (generally with great wailing and gnashing of teeth). The presence of that idea in a rules set usually means the game has a particular design space in mind to represent the idea that player character decision making can be swayed or influenced by other player characters or NPCs. Even in cases where a PC can be swayed via mechanics, the player is usually at least partially in control the narrative effects. It's a mechanical choice used to provide some separation between character and player when it comes to following the fiction. It's also usually a feature of rules sets that put a greater emphasis on social interaction and decision making. From a control standpoint its really more like the improv yes and than it is anything else, you just have a mechanic to enforce certain instances of what a successful roll means.

 

[pemerton]

there's a whole other factor involved IMO: player agency also involves, in normal situations, how much control one has over one's PC and its (attempted) actions both in the fiction and at the meta-level.

<snip>

Banning obviously impossible action declarations is a hit to my player agency in the fiction; I can't have my PC try impossible things just for kicks. Note, however, that allowing the declaration and then resolving it by simply saying that such an action is impossible (and if necessary figuring out any ramifications of my attempt) is just fine.

It's a small difference, but a very significant one, between a) saying I'm not even allowed to declare the action of shooting an arrow at the moon and b) allowing the declaration and then no-roll resolving it by flat-out saying I miss the moon and now let's figure out if that arrow hit anyone when it landed. Option a) denies my player agency; option b) preserves it.

I think you're confused about what is going on in the discussion of actions that violate the credibility test.

Go back to Robin Laws's example from HeroQuest Revised:

As Narrator, you are never obligated to allow a contest just because two characters have abilities that can be brought into conflict. If the character's proposed result would seem abusrd, you disallow the contest, period. . . .​

Read it carefeully: if the character's proposed result woudl seem absurd [ie if it violates genre or fictional positioning constraints] then you disallow the contest [ie no check is made; the action resolution mechanics are not invoked].

The player is free to describe his/her PC shooting an arrow into the sky aiming at the moon. But (outside the context of some sort of epic fantasy) that does not generate a check to see if the moon is hit.

The premise of Laws's remark is that a system is being used similar to what @chaochou mentioned upthread: namelhy, that if a valid action is declared then the dice are rolled and on a success the player gets what s/he wants for his/her PC, and otherwise the GM narrates a failure.

In a system in which that is not true - ie in which there is no particular connection between use of the action resolution mechanics and what happens in the shared fiction - then Laws's point becomes irrelevant. But that takes us straight back to @chaochou's other point, that where there is no such connection players have little agency.

And this also brings me back to the distinction I drew upthread: the credibiilty test is something which can be negotiated between participants (so players get to exercise their agency just as much as the GM does); action resolution, on the other hand, is a die roll to see whose vision of what happens next is made true.

 

[prabe]

@Fenris-77

Yeah. I was trying to be as neutral as I could about a style of game that I've bounced off of a few times. The games very probably feel different in play, for people who enjoy them.

And yes, agency and narrative authority do tend to run in parallel, but they're not the same thing--and I think it's worthwhile to at least be able to talk about them separately. I don't think the trade-offs I was talking about are radically different from taking a negative trait in a game with point-based chargen, to get more points: One, the player accepts some relative disability in exchange for greater capabilities; the other, the player accepts some loss of control over the character in exchange for more control over the greater world. (Where "control" isn't about that weeping and wailing and gnashing of teeth--I'm just not thinking of a better word for it; I'm not using it to bash any games or players.)

 

[Fenris-77]

I'm not bashing anyone either. Players who come from games like D&D, where narrative authority rests heavily with the GM and the authority they do have is very much concentrated in their character, it's pretty natural that idea of giving some of that authority up doesn't sit well. All I can say is that in most cases that 'give' is located within a larger redistribution of authority that usually results in more player agency, not less.

 

[prabe]

I'm not bashing anyone either. Players who come from games like D&D, where narrative authority rests heavily with the GM and the authority they do have is very much concentrated in their character, it's pretty natural that idea of giving some of that authority up doesn't sit well. All I can say is that in most cases that 'give' is located within a larger redistribution of authority that usually results in more player agency, not less.

Felt more like an even-up swap to me, but that's me.

I just wanted to be clear about not bashing Fate (or BitD), because ... well, I have in the past. I might again in the future. But I'm really not now. And I haven't felt any bashing from the Canadian Arctic, so far as I know we're good.

 

[pemerton]

Well I don't really think I should be left guessing about what you don't understand about it. Maybe elaborate a bit on what doesn't make sense to you about the stance?

I don't understand why you distinguish between genre inappproriate fiction and genre inappropriate action declaration. Eg how would one get the first except as a result of the second?

It's very possible I don't know enough about those other games, but I'm not seeing any such distinction from the examples of them you have given. Every example I read sounded like it was saying "use auto failure to keep the game in genre, etc.

Notice how I distinguished between (i) establishing what counts as "within genre" and "consistent with the fiction" - that is something for the table to decide on, with the GM playing a leadership but not an authoritative role - and (ii) employing the action resolution mechaincs which generate an outcome in exactly the way that @chaochou has described: if the check succeeds the player gets what s/he wants for his/her PC; if the check fails the GM narrates the failure.

Your idea of "autofail" is a unilateral GM decision about genre or fictional positioning - it does not involve the negotiated/consensus aspect of (i) but also does not involve the "leave it to the dice" aspect of (ii). It is just the GM unilaterally deciding what happens next. That is what @chaochou is pointing to as negating agency. It's what the games I referred to are trying to avoid.

You may not have the same desire to avoid it as did the designers of those games. But I hope that actual distinction is clear enough. I find it a pretty fundamental distinction in actual play.

That a GM may delegate authority doesn't negate the fact that it was his authority to delegate.

<snip>

Even when the mechanic (aka the process of determining success or failure) is "GM Decides" it's always the GM's job to apply and abjudicate the action resolution mechanics.

I assume that you're talking here about your approach to GMing 5e D&D. I'm pretty sure that 5e D&D will also work OK if you approach it having regard to my distinction between (i) and (ii) - I know that 4e D&D works just fine that way, and I don't think 5e is radically different in this respect.

One important consequence of taking seriously the approach I'm setting out is that you will avoid problems like the OP's, because you will not have actions automatcially fail on the basis of elements of the fiction that are not shared but rather exist only in the mind of the GM.

I don't believe a chess player's agency is taken away because he can't move his knight as a queen. I believe the agency of a player is defined by the moves they can make inside the game and that those moves cause unique feedback inside the game.

<snip>

All styles allow players to change the shared fiction by having their PCs do things. This even occurs in an instance of auto-failure. Player had PC attempt to do something -> auto-failure -> shared fiction changes.

I don't think comparisons to chess are very helpful here. In chess players take turns and make moves according to rather rigid rules. In a traditional RPG - which all the games I've talked about in this thread are - there are quite distinct roles with very different powers and responsibilities when it comes to contributing to the shared fiction.

If a player makes a move - it might be better to say attempts a move - that is consistent with the established shared fiction and the broader genre and logic of that fiction, but the GM unilaterally declares it a failure, the player has not exercised agency over the fiction. S/he might have prompted the GM to use the GM's agency - eg now the Burgomaster is mad at you and calls his guards to take you away - but that is not player agency over the fiction. The GM is writing that story. The OP is crystal clear about that much at least.

 

[pemerton]

So the DM should never say yes? Everything must be rolled for?

If you ask me to pass you the salt, and I do, I haven't negated your agency. I've facilitated it.

If a player says (speaking for his/her PC) I go to the shop and buy some rations and the GM answers OK, write them down on your equipment list the GM has not negated the player's agency. The GM has allowed the player to decide what happens in the fiction.

If the GM says, instead, Rations are hard to come by in these parts. Make your Acquiring Stuff check! then we have an example of the action resolution framework that @chaochou has already described upthread.

If the GM says, instead, There are no ration venders around here - it's a wild and desolate place we now have two paths we might go down. Is this the GM taking the lead in establishing constraints of fictional positioning and genre? That's something where the players can participate in the negotiation, thus exercising their agency in reaching a consensus.

Is this the GM unilaterally exercising control over the content of the shared fiction, based on his/her prior conception of what that fiction does and doesn't look like? Then we have no player agency, as @chaochou has said. In this case it's obvious that the GM is the one who is controlloing the content of the fiction.

Again: it may be a good thing or a bad thing for the GM to unilaterally exercise control over the content of the fiction. But it can't be true both that the GM is exercising unilateral control and that at the very same time the players are exercising their agency.

What if the established fictional constraint results in no chance of success?

I've made many posts about this already. The established fiction is not something that the GM unilaterally imposes. It is something that is understood and can be negotiated by the whole table. It is established by the interaction of all the participants, as in any other shared or collective endeavour.

This happens all the time in my experience: different participants make different suggestions about what might be the case in the shared fiction and we work it out. Eg in my game on Sunday the application of action resolution mechanics dictated that one PC had fallen off the boat into the water. The player of that PC then wanted to use his sword to fight the dragon that was responsible for tipping over the boat. But that can only happen if the PC still has his sword on his person!

For Prince Valiant - the game we were playing - that's generally a matter of fictional positioning, ie there is no canonical resolution procedure for retaining or losing possession of one's gear. (Contrast, say, Marvel Heroic RP which does have such a canonical resolution procedure via the Gear limitation.) We quickly agreed that the player did still have possession of his sword, and hence could declare actions about using it to fight the dragon.

In the same session the PC who is Master of the PCs' military order lost an argument with a NPC count about who would lead the charge in the next day's battle. None of the PCs - and none of their players' - was happy with this outcome. They wanted to circumvent it. I had to remind them more than once that the argument had been lost and conceded. They therefore ended up circumventing it by leading their forces out for a night-time raid on the enemy, with the goal of making it be the case that there would not be a charge the next day. This is an example of negotiation and consensus on what exactly is or isn't consistent with the established fiction.

None of this involves unilateral GM decision-making about what might happen next.

 

[pemerton]

It confounds me that someone would only consider it to be player agency over the fiction when a player can exert control over the outcome of a fictional action. That's definitiely one kind of player agency over the fiction but it isn't the only one.

In this case it's apparent you are talking player agency in relation to fictional outcomes. I think that your point about that is so self-evident that no one disagree with it. The problem arises when you shorten that to player agency as if there's no other kind.

Let me illustrate the other kind by introducing another mechanic in addition to yours above. On a 1-3 the player decides the PC action. On a 4-6 the DM decides the PC action. This example also illustrates an additional loss of player agency. The kind that is in relation to declaring PC actions.

I think the reason that @chaochou is not having regard to this other mode of player agency that you are calling out (and which @Lanefan has also called out) is because he is assuming that there is no RPGing in which players don't get to declare actions for their PCs.

Therefore he is focusing on what is variable among different RPGs - both across systems, and across actual instances of play in those systems. And what is variable is the amount of player control over what happens next in the fiction.

Now maybe @chaochou si wrong to assume that in every instance of RPGIng the players get to declare actions for their PCs. But in the spirit of using terms carefully, I would tentatively suggest that such a game - whatever exactly it is - is not a RPG. Because the players have no role to play in it at all.

 

[pemerton]

In context with "The GM Decides"--and (I thought) specifically talking about action resolution--I thought it was clear, but I'll try to be clearer.

Talking specifically about action resolution:

"The GM Decides" is the GM deciding that there is no doubt about the outcome of an action; either it cannot succeed or it cannot fail.

"The Dice Decide" is the outcome is in doubt and the dice determine the outcome.

I don't understand what you mean by "the dice determine the outcome". Can you give an example?

All the examples of RPG mechanics that I'm thinking of at the moment involve a player declaring an action for his/her PC and then using the dice to find out whether or not the action succeeds. So it is as @chaochou has said: if the player wins the dice roll s/he decides what happens (because his/her desired change in the fiction comes true) and if the player loses then the GM decides what happens.

I guess there's one exception I can think of: at least on one interpretation of the relevant rules, in AD&D and Moldvay Basic the GM should roll a reaction roll to establish the starting attitude of NPCs and monsters encountered whether or not the players have their PCs initiate interaction. That is not a mechanic that gives players any agency - it is a device the GM uses to randomise the framing of the situation. But that can't be an example of what you have in mind precisely because it is not about action resolution. As I just said, it's about framing.

The PCs are in a world. They are not the first characters in that world, nor will they be the last. I try to keep the world consistent and occasionally have things happen that are unrelated to the PCs.

Someone somehwere has probably played a game of D&D where this was not true, but I think it's pretty typical.

But it doesn't answer any of my questions. You're talking about the content of the fiction. I'm asking how is that fiction authored? By whom? And how is that authored ficiton used in subsequent adjudicaitons of declared actions?

If a game is set in the real world, there are going to be things the PCs won't be able to do, and some of those things will be impossible because of the GM's understanding of the real world, which might be different from how the real world objectively works

But now you're just assuming that players dont have agency. From time-to-time I GM games that take place in the "real world" - Cthulhu Dark and most recently Wuthering Heights. The players as much as me get to express views over what can be done in the real world. Eg in one of our Cthulhu Dark sessions the PCs had taken control of a tug boat and the player who knew the most about tug boats told us what could be done with it.

In my games set in non-real worlds - eg my 4e game - the players also help decide what can or can't be done. Eg in that game it was the player of the invoker/wizard who generally took the lead in deciding what was possible to be done with magical effects.

This is why - multiple times upthread now - I have emphasised that establishing constraints of genre and fictional positioning can be a matter of negotiation and consensus, in which the players exercise their agency as participants in that process. It need not be unilateral GM authority.

And the fact that it can be negotiated is a reason for distinguishing it from action resolution procedures which, in the traditional RPGs that I play, are not about negotiation but rather involve rolling dice to see whether or not the fiction unfolds as the player is hoping for his/her PC.

It's possible that I'm the only one still in this conversation who separates player agency (the ability to choose what a character does) from narrative authority (the ability to tell the story). The ability to choose--whether to try to swim across the river--is player agency; the ability to describe the result--a current or a monster or angels or a canned leafy green vegetable--is narrative authority. I have played at least one game where narrative authority was not dependent on success in task resolution--so you might fail at a task and have authority to narrate that failure.

As I just posted in reply to @FrogReaver, the definition of player agency that you posit here is uninteresting because in every RPG players have it. It's not something that varies.

@chaochou has made it crystal clear that by player agency he means the ability of the player to change the state of the shared fiction. Given that, in a traditional RPG, the way a player changes the shared fiction is by declaring actions for his/her PC and then having those resolve, the connection between player agency and action resolution procedures is not coincidental.

If a player can't change the shared fiction; if all s/he can do is prompt the GM to make such changes by describing what it is that his/her PC tries to do; then what is the role of the player in the game?

 

[pemerton]

The question about whether anyone other than the player of a PC can make it true in the fiction that the PC does such-and-such is a different one from whether or not a player is able to make changes to the fiction.

Off the top of my head I can't think of any RPG in which only the player of a PC can make it true in the fiction that that PC does such-and-such. Apocalypse World comes pretty close, but the worked examples of play make it clear that the GM is free to talk about stuff the PC does when narrating failures; though that will be mostly colour.

D&D is very permissive in this respect: there are a whole host of GM or other-player moves (in the fiction they are mostly said to be magical effects) that let someone other than the player of a PC say what that PC is doing. This all goes well beyond mere colour.

This doesn't always have to be magic, either. 4e D&D has a monster (I think a type of chained demon?) that can use its chains to manipulate an enemy like a marionette: in mechanical terms this lets the GM, as the player of that monster, declare actions for the PC.

And there are many RPG systems that have social/emotional resolution frameworks that can allow someone other than the player of a PC to say what that PC does. The earliest published version of such a framework I know is the Classic Traveller morale rules (1977). I posted an actual play illustration of this sort of thing not far upthread: in our Prince Valiant game on Sunday the player of the Master of the Order lost the social conflict with the Count, which meant that I as GM got to decide that he agreed to let the Count lead the charge in the next day's battle.

There is no particular correlation here that I can see. AW has a high degree of player ability to make changes to the fiction and a low degree of capacity for someone other than the player of a PC to make it true in the fiction that that PC does a thing. MHRP/Cortex+ Heroic has a high degree of both. A game that used GM-decides D&D to play a low- or no-magic game (thus purging out all the charm, fear, etc effects) would have a low degree of both. D&D played with the magic taken straight from the books, and using a GM-decides approach to resolution, will have a low degree of player ability to make changes to the fiction and a high degree of capacity for someone other than the player of a PC to make it true in the fiction that that PC does a thing.

As I said, there seems to be no particuar correlation.