A Question Of Agency?

[pemerton]

For those not as versed in mathematical infinites I think this analogy might help.

There are an infinite amount of numbers. There are also an infinite amount of odd numbers and an infinite amount of even numbers. If I remove the infinite amount of even numbers I am still left with the infinite amount of odd numbers.

(Odd numbers representing player agency and Even numbers representing the force that takes away agency).

I don't understand the relevance of this. Given that every RPGer is a human, and all humans are mortal, every RPGer will only engage in a finite number of moments or episodes of RPGing. Only a finite amount of authorship will take place.

It's not strictly zero-sum between GM and players, because collaboration among humans doesn't work quite like that. But every moment in which fiction is introduced by a unilateral decision of the GM is a moment that the players were not exercising agency.

 

[pemerton]

The GM rolls no dice to fudge, has no enemy stats to obfuscate what they are doing, and is beholden to no prep.

<snip>

Another feature which makes it pretty difficult is that Blades resolves player intent and makes success incredibly transparent. There are no DCs, TNs, or Obs. You look at the dice and you know. Blades also provides for resistance rolls to overturn consequences. Playing games there is trivial to see.

I think the issue of setting difficulties (or not, in systems that don't use them) is very interesting. And also how this interacts with building NPCs (where that is a part of the game).

In 4e D&D it's pretty transparent in virtue of the DC-by-level chart, and the design of PC-build-elements is fairly baroque in parts precisely to maintain a solid mathematical interaction with those DCs. (There are a few areas of breakage - I will nominate Sage of Ages epic destiny as one - but not that many that I have encountered.) There are also very solid rules to govern NPC/monster building.

In Prince Valiant setting DCs is easy because they generally range from 1 (easy) to 4 (pretty hard) but opposed checks are a bigger deal because those depend on NPC stats, and that requires GM good faith and fidelity to genre, prior NPCs encountered, etc. Some of the episodes in the Episodes Book (not Greg Stafford's original rulebook) in my view suffer from some overdone/gerrymandered NPC builds. I've had to correct some of these for my own play.

Classic Traveller has transparent NPC build - just follow the lifepath rules like the players had to - but setting other DCs can be quite opaque. I do my best to explain my reasoning to my players but I don't know if they always follow it. What does help a bit here is the relative transparency of a PC "stat block" which means that, basically, PCs are good at what you would expect them to be given what it says on their tin! So you don't get weirdnesses that can happen in more convoluted PC build frameworks (like one time in a 2nd ed AD&D game where my 1st level cleric was a better combatant than the 1st level fighter - the player of the fighter got a bit of a shock).

Burning Wheel uses both set difficulties and opposed checks, like Prince Valiant. It needs NPC stats to make some of this work, like Prince Valiant and also Traveller. And like Traveller, the setting of difficulties by the GM is meant to be one way that the GM makes the world "come alive" to the players. The Classic Traveller rulebook advises the referee to keep notes to help ensure consistency in this respect. BW doesn't give such specific advice, but the rulebooks list many more particular difficulties than are found in Traveller, so maybe those lists are in part a substitute for GM notes.

Anyway, I think this range of approaches and the various principles that should govern them are a bit under-discussed.

 

[FrogReaver]

I don't understand the relevance of this. Given that every RPGer is a human, and all humans are mortal, every RPGer will only engage in a finite number of moments or episodes of RPGing. Only a finite amount of authorship will take place.

It was being used as an introduction to a few important concepts. It's not so much relevant how we get to these concepts as what the concepts the thought experiment about infinite duration games brought to light.
1. How should we measure agency?
2. What does more agency actually look like?
3. Is a game with alot of force necessarily a game without alot of agency?

It's not strictly zero-sum between GM and players, because collaboration among humans doesn't work quite like that. But every moment in which fiction is introduced by a unilateral decision of the GM is a moment that the players were not exercising agency.

But we have established that a lack of agency in a moment or even many moments doesn't mean that there are necessarily less moments that the players have agency. As you say above, it's not a 0 sum.

While you verbally acknowledge it's not 0 sum, It sounds like you are basing your conclusions on the premise that agency in a game is a 0 sum scenario. Your thoughts come out as if they are contingent on agency being 0 sum, such that there is only so much of it and if the player is deprived of any that means the game has overall less agency. But as you just mentioned - it's not a zero-sum between the GM and players. If it's not 0 sum then pointing to a moment and saying - see the player has less agency here doesn't actually mean that game produced less overall moments of player agency.

 

[FrogReaver]

No. What character types might be available isn't a matter of agency whatsoever. If you think it is then when you say 'agency' you means something other than what the other people in this thread mean.

You disagreed. Could you elaborate on why you are disagreeing? What do you see being different in those scenarios?

 

[FrogReaver]

It's literally in the there you quoted earlier:

Judgement calls vs "railroading"

on this failure, did you author that the brother was evil or did the player? Finding cursed arrows, even knowing they were made by the brother, would not prove guilt to me. There are a ton of other explanation that fit, especially in a game where facts are ephemeral until a failed search for a...

www.enworld.org

I found the below quote from that thread very interesting...

Well, in the sort of game I run it's my job to frame the PCs into situations where the various aspirations of the PCs intersect.

I cannot understand how this is not force?

 

[pemerton]

I am not accusing you of doing anything, though I feel you really do not appreciate how much the information the GM provides will impact the direction of play. In your explanation I see multiple points where the GM clearly either influences or has an easy opportunity to influence the outcome. But I do believe that you do not use this to consciously direct the game in any preplanner direction.

And things like reading the player reactions, taking account their character's motivations, carefully nudging the game in certain directions, inserting preplanned cool elements where they naturally fit, drawing players' attention to certain things are not dickish behaviour to me; these are things that a good GM should do! Now I understand that this is not within the spirit of the Blades, so that is another reason why I wouldn't be interested in that game.

I have never GMed Blades in the Dark. Nor have I read the rules. My knowledge of it is based on (i) others posting about it, and (ii) its resemblance in certain respects to Apocalypse World and Dungeon World.

That said, I am pretty confident that a GM of BitD is expected to have regard to the motivations that players establish for their PCs. I think @Ovinomancer was doing that in the actual play report he posted (about the haunted house and the painting). But the GM is not expected to nudge the game in certain direction, and doesn't really have the resources to do so. I'm not sure about preplanned cool elements - there is a pre-established setting (Duskvol) and so I imagine that does at least suggest some cool elements. But I don't think they're meant to be secret from the players.

Most importantly, in the context of a discussion of participant agency in RPGing, I don't think the GM is entitled to declare that a declared action fails by reference to the GM's unilateral conception of the fictional situation. If the GM isn't going to say "yes" then I think the action has to get put to the test (which is what we see in @Ovinomancer's game: the PC attunes to the painting to try to work out if/how it is enchanted).

The player is still forcing their desired outcome to the reality, it just is a gamble. But there really is no mystery. "Is this painting magical" is not really an interesting question any more. "It is because I decided to examine it, though it may eat my face." or even "It has certain chance of being magical because I decided to examine it" are ultimately both answers that the player just directly produced

What makes a good game is actually being able to study the situation and make either tactical or dramatic decisions based on that.

What makes the action meaningful is not whether the things are codified in the rules, it is the existence of objective base reality against which you can make decisions. Rules are one (and often good) way to communicate such reality, but not the only one.

<snip>

What makes it low-agency is the player not being able to gain meaningful information or make meaningful choices regarding that goal, at least according to your definition which discounts flavour. The player could have latched into any item, any time, anywhere, and interrogate it in the same manner than the painting to force the check. The rest is RNG.

<snip>

Right. So there actually is some independent reality you can learn about. You actually need to study it more to progress your quest. Meaningful choices can be made.

<snip>

There needs to be some reality against which to make decisions for the decisions to matter. Sure, getting to tell a bit of the story and randomising who gets to do it is a form of agency, and if you like that sort of agency good for you. But it is not really making meaningful choices, except perhaps flavour wise, and this is something you had low regard earlier.

What all this suggests to me is that (i) you are not very interested in character-driven or character-focused RPGing, and (ii) you much prefer what I call "RPGing as puzzle-solving" or "RPGing as learning what is in the GM's notes".

What establishes the meaningfulness of the choice made by the player in Ovinomancer's game is that the PC, as played by the character, is prepared to take a risk to find a magical item that will improve his relationship with the university. We now learn something about this character, his drives, and what he thinks is worth taking a chance on. That is (broadly speaking) theme. The fact that it involves soul-sucking is probably closer to trope than theme, thought that's not a bright-line boundary and I'd of course be happy to hear what Ovinomancer thinks about that.

The idea that choosing to stake your soul on finding something to improve your standing with the university is not meaningful and is mere flavour is - to me - a very strange one. I wasn't in Ovinomancer's game but to me that sounds like part of a cool situation leading to interesting stuff down the track. In my BW game where I'm a player, my PC Thurgon is prepared to stake his life to defend his honour, and to restore (what he sees as) the honour of his family and their estate. This is why encountering his brother Rufus as he did, and why his failure to rouse Rufus to action, mattered. It's not mere flavour - that's the game!

The point of the random number generation in BitD (and AW, and BW, and - I would say - 4e D&D) is not to deliver theme. That's built-in and guaranteed by the rules for PC gen, for establishing the consequences of action declarations, and for framing scenes. The point of the dice is to manage pacing and related story dynamics. In a good story the protagonists get what they want some of the time, and they fail some of the time. Sometimes the chances they take pay off; sometimes those chances are overreach and redound upon them. In these RPGs, that is determined by the dice rolls. Part of the skill of designing these games is to make sure the maths works to produce reliable peaks and troughs of success and failure and complication. (We can also distinguish the games along those lines: 4e D&D produces more success than failure and so - especially when this combines with its tropes - tends towards the gonzo; whereas BW produces a pretty high rate of failure for a RPG and this is part of what makes it a demanding experience on the participants - players because their PCs are suffering and GMs because they're obliged to drive home those failures.)

There is nothing in my BW game, or in Ovinomancer's BitD game, that is remotely comparable to the GM having prepared a haunted house mystery where the job of the players is to manoeuvre their PCs, via "I walk towards the . . ." or "I closely inspect the . . ." action declarations, into fictional circumstances where the GM then tells them pre-authored fiction which the players gradually piece together to solve the mystery.

I did post upthread about a recent scenario I ran that was exactly as I've just described:

A few weeks ago I ran a session like this for my family - one of my daughters wanted to do a murder mystery for her birthday.

I adapted a murder scenario from an old Traveller module, and wrote up some characters (one for each other family member, plus a couple for their entourages, plus a small number of important NPCs whom I played). There was no action resolution in any mechanical sense - the players described what their PCs were doing, and who they were talking to, and I delivered up information as seemed appropriate (eg what they found if they searched a stateroom; what a NPC said if they spoke to him/her; etc).

This is an example of puzzle-solving: the players' goal is to acquire enough information to be able to infer to the hidden bit of my notes (ie whodunnit). It is a different experience from watching an episode of Death in Paradise or The Mentalist, as there is the first-person description element to it. But it doesn't really involve very much more agency.

(One difference from those shows is that they are scripted to try and occlude the audience's access to the relevant information, whereas in our murder mystery I was desperately trying to shovel information out the door. A better comparison might be to reading The Eleventh Hour.)

That was fun enough, but involved very little player agency in respect of the shared fiction. It was more interactive than solving a crossword puzzle or solving The Eleventh Hour, but at its core was not a radically different intellectual exercise.

 

[pemerton]

Generally speaking, the division of authority starts with a model like D&D and from there you move down the spectrum with different bits of authority devolving to the players in different systems. Authority doesn't generally move the other way round. I mean the spectrum goes both ways, but historically speaking what we've seen is a steady movement toward devolving more authority on the players, which does indeed come from the GMs slice of the authority pie, which isn't a bad thing, or a good thing, just a thing.

I want to disagree with this. Early RPG rule texts expressly or impliedly conferred more authority on players than became the norm in 80s and many 90s texts.

In another thread earlier this year (haven't tracked it down, sorry, but can if you like) I posted an illustration of this from the evolution of the rules for Traveller: one edition of the 1981 version expressly confers authority on the GM to use illusionistic techniques to drive "the story"; whereas in the 1977 edition the referee's power to drive "the story" is expressed in terms of choosing to impose encounters (ie what today one might call scene-framing). But there is nothing in the 1977 version to suggest that the GM will exercise unilateral power to decide outcomes.

My own explanation for this trend is that early RPGs were more explicitly modelled on wargames, and as there was a move towards "story" as a desideratum of play there was no real understanding of how this might be done except via GM control over the shared fiction. There was also a tendency for the mechanics to lag - so the 80s and 90s see many games whose mechanics are fairly close to classic D&D (map-and-key resolution, rules for interpersonal combat and interacting closely with architecture, etc) even though the ostensible goals of play are very different. GM agency is the device these systems use to bridge from their mechanics to their goals.

 

[pemerton]

The conversation with @Nagol was about "soft-balling" (I brought this precise point upthread if you recall) and not classical Force.

Right. I went back to reread the post and had exactly the same thought.

In the spirit of thread recursion, maybe now is the time for me to reiterate that I am a sentimental GM who finds it hard to narrate hard consequences. But I wouldn't really call that railroading!

 

[pemerton]

See, and I am one of the weird ones that thinks if the end of an adventure (and the scenes between beginning and end) is already predetermined then the adventure is a railroad. But I do realize that many people believe otherwise. Oh well, to each their own I guess.

I agree. I also think that The Alexandrian's "node based design" is also about railroading. The fact that it goes A, B, C, D rather than A, C, B, D due to player choice doesn't change the fact that all the situations were largely predetermined.

Some of the old Fighting Fantasy books also had elements of node based design, in the sense that they could handle the difference between getting to B before C or getting to C before B. But clearly the reader/player of one of those books is not exercising very much agency over the content of the shared fiction. It's a puzzle-solving experience.

 

[pemerton]

I found the below quote from that thread very interesting...

Well, in the sort of game I run it's my job to frame the PCs into situations where the various aspirations of the PCs intersect.

I cannot understand how this is not force?

Because it doesn't undermine or change the outcomes of action declarations? Because it builds on player-signalled priorities for the game and their PCs (and so is a form of "taking suggestions)? Probably other reasons too, but they're the first I thought of.

GM force is not a synonym for GM authorship.

As @Manbearcat uses the term - and I believe he's the one who brought the term into this thread - it's about the GM perturbing the (ostensible) method the system uses for turning participant inputs into shared fiction.

For my part, when I think of GM force I think of the GM making unilateral decisions about the fiction that have the effect of either blocking or altering what would otherwise be player contributions mediated via the action resolution mechanics.

Now there is a style of RPGing where what I have just described is essential: in classic D&D play the GM has to prepare the map and the key in advance, and has to refer to them to resolve exploration-oriented action declarations. The goal of the players is - quite literally - to recreate the GM's map and the GM's key by declaring exploration-oriented actions for their PCs which will prompt the GM to share his/her notes with them.

As I have already posted in this thread, I think that the style of RPGing just described breaks down as soon as the fiction becomes rich enough that the players can't use straightforward exploration-oriented action declarations to recover the fundamentals of the fictional situation from the GM. Any "living, breathing world" will cause this issue: eg it's not feasible to learn every escape route assassins might have taken in a city, or to learn of every person willing to lend money to the improvident mayor, or to learn where every source of magical shapechanging in the duchy, simply by using exploration-oriented action declarations of the sort that are so fundamental to classic D&D.

Given that I neither GM nor play in classic dungeon RPGs, and that every game I've run or played in since about 1986 has been of the "living, breathing world" variety, I'm not that concerned with the principled "force" of classic dungeoneering. My concern is with the impact that GM force has in the sort of games that I've GMed and played in over the past 35 years. And that impact is to reduce, and in some cases even eliminate, player agency over the shared fiction.

Framing a situation - you see a passer-by fall to the street, and what must be the assassin escaping across the rooftops or the mayor asks you to lend her money or OK, so you're going to the wizard's academy to try and speak to an expert in shapechanging? - isn't force. It doesn't unilaterally establish a particular outcome in the fiction outside of the action resolution procedure. It's creating the context for actions to be declared.