A Question Of Agency?

[Fenris-77]

Do you guys understand that you're arguing over which parts of the imaginary reality are real?

Hah. Not really. Call it as really real as the parts of the fiction that is established at the table during play. So yeah, I guess, for a given value of real. You could use the word established if you preferred.

 

[Campbell]

I don't think we disagree a whole lot. But what I will say is I think there are different procedures and methods for determining this stuff, and some are more grounded in setting down objective details than others. But the aim of the GM is to fold all those things into the living world, and to treat them as live forces. For example a random encounter doesn't exist until you roll it, that is for certain. But when I roll a random encounter I do try to give it a sound reason for being there. And if that encounter result is an existing NPC, then I look to that NPCs motivations, etc. But this is different from a sect that I've established in my notes as existing in Flower Bridge Village, at Red Lotus Manor. That exists in my mind and in my notes before the players go there or ask about it. I may get some clarity about the sect once the players arrive for sure. I don't want to paint an overly rigid image of what is going on just to fend off some of the more extreme edge cases being used in this discussion. And that is important, the interaction with the players cause me to think more clearly about the place. But those details aside, Red Lotus Sect existed, in the way things can exist in a game world (not actual real world existence) well before it was set in play. Sometimes I have a technique I employ where I write things down so they are set right before the players go there. For example if the players go into town looking for a fish monger, I make a quick mental map of whether that is reasonable (is there a water source with plentiful fish), then I decide if the there is a fish monger, who he is, some key details like his wife is cheating on him with the proprietor of the Fragrant Word Teahouse, etc. I do that because I want that stuff set in order to create the impression of a real world and to force myself to have fidelity to that world. It is true these details are being made up, but whether they are being made up two weeks before or two seconds before, it is coming into being in the setting prior to the players experiencing it directly. Now that won't happen in every case. you can't forsee everything, and you can't rigidly run a game just to adhere to a concept. But I do find these kinds of techniques and this approach generally quite useful because it does all help contribute to the sense of a world the players are exlporing, rather than one being created by their actions.

I have used some similar techniques and will continue to do so when I run Stars Without Number, Godbound, and Wolves of God. I am just mindful of my part in it. That I am working to make these things feel real. There's no shame in that. When I act I try to embody a character as fully as possible, but am mindful that the acting process is one of creation.

I can see how adopting that mindset might be helpful while running or playing a game in your specific set of play priorities. It's not going to be very useful when it comes to comparative analysis because it locks discussion only to that specific set of play priorities. If that is the only way you will engage there's no hope of understanding Burning Wheel, Moldvay B/X, Dogs in the Vineyard, or innumerable other games. Even players who might approach the same games you play differently.

 

[Manbearcat]

On this, I think we agree.

A lot of discussion of stance approaches it as if stance were a psychological (or, as you say, cognitive) state. But clearly its not: it's possible to engage in director-stance play (eg by making a Wises check or Circles check in BW) without entering any psychological state different from any other time when one says what one's PC is doing.

Stance is, rather, a type of "logical" relationship or "authorial" relationship between player and shared fiction. It's not possible to simultaneously be in Actor and Author stance, simply because as defined they are contraries, but it's quite possible to be in Actor and Director stance (as you say) and probably also Author and Director stance.

You might remember my example of the paladin who was turned into a toad and then turned back, by the Raven Queen, as my example of simultaneous Actor and Director stance from the epic "dissociated mechanics" thread.


I've cut out the discussion of AW failure, and also of 5e - I'm not sure where exactly you want that latter example to go.

If I succeed at reading a charged situation, then - as a player - I am able to oblige the GM to narrate some new fiction. As you say, that will enrich/develop the setting, the situation, or perhaps both. (I'm not sure if that can be the case for any single question, but I might get 3 of them!)

I certainly think that is a type of participant - in this case, player - agency. As I've already posted, I see it as being a certain cleverly-structured form of making suggestions to the GM which the GM is not free to disregard. My use of the concept "making/taking suggestions" comes from this post by Ron Edwards:

​

I'll expand those authorities I talked about into a list, with a key addition and with the order changed for greater clarity:​

​

Content authority - over what we're calling back-story, e.g. whether Sam is a KGB mole, or which NPC is boinking whom​

​

Plot authority - over crux-points in the knowledge base at the table - now is the time for a revelation! - typically, revealing content, although notice it can apply to player-characters' material as well as GM material - and look out, because within this authority lies the remarkable pitfall of wanting (for instances) revelations and reactions to apply precisely to players as they do to characters​

​

Situational authority - over who's there, what's going on - scene framing would be the most relevant and obvious technique-example, or phrases like "That's when I show up!" from a player​

​

Narrational authority - how it happens, what happens - I'm suggesting here that this is best understood as a feature of resolution (including the entirety of IIEE), and not to mistake it for describing what the castle looks like, for instance; I also suggest it's far more shared in application than most role-players realize . . .​

​

I'm suggesting that you look at it from the total opposite viewpoint - that these four things are separate, they will always be separate . . .​

​

Do they have causal relationships among one another? Of course. The easiest version is top-down reductionist: because content is consulted, a plot authority decision is made, and then a situational authority decision/presentation must be made, and finally narrational authority must be exercised. I assume that for you, this is the most easy and familiar construction, and you're used to conducting them (or at least constructing them, idealistically speaking) as a single causal sequence in this order, with one person in charge - it's a "thing," perhaps the thing you call GMing.​

​

As a side note, other causal relationships exist, putting the authorities into a different order (to preserve the top-to-bottom causation, for clarity). For example, you can reverse them entirely, and remarkably it is very easy, although it's harder to catch oneself doing it because memory typically rewrites the act into the more familiar sequence I described above. We'll have to work on this idea later, because, for instance, Kickers and Bangs in Sorcerer rearrange the sequence far more drastically, putting situational authority at the top/starting position. Please don't get distracted by this paragraph. It's intended to be a distant signpost to future discussion.​

​

The real point, not the side-point, is that any one of these authorities can be shared across the individuals playing without violating the other authorities. . . .​

​

Well, let's look at this [ie another poster's bad RPG experience] again. Actually, I think it has nothing at all to do with distributed authority, but rather with the group members' shared trust that situational authority is going to get exerted for maximal enjoyment among everyone. If, for example, we are playing a game in which I, alone, have full situational authority, and if everyone is confident that I will use that authority to get to stuff they want (for example, taking suggestions), then all is well. Or if we are playing a game in which we do "next person to the left frames each scene," and if that confidence is just as shared, around the table, that each of us will get to the stuff that others want (again, suggestions are accepted), then all is well.​

​

It's not the distributed or not-distributed aspect of situational authority you're concerned with, it's your trust at the table, as a group, that your situations in the SIS [ie shared imaginary space, or shared fiction] are worth anyone's time. Bluntly, you guys ought to work on that.​

​

It seems to me that, if the GM is obliged to take suggestions, then even though the GM has formal authority over (say) situation, the players clearly have a great deal of agency. Edwards also gets at this in an earlier post in the same thread:

GM: "All right, you guys see the lamps of Apple Town up ahead. You've arrived."​

​

Player: "I spend a Story Token. My uncle lives here! We go to his cottage."​

​

In games with such mechanics (or, in fact, games without such mechanics but in which such suggestions are welcome as suggestions), the GM pretty much has to be ready for some footwork, once in a while. If his prep, for instance, includes the assumption that no one in this town knows any of the PCs, well, he might have to think a bit.​

​

But on the other hand, and presuming that the group is fully aware of these mechanics or these suggestions, it's really not as prep-destroying as you might think. The GM might have been wondering how the hell to get these guys into the conflicts of the town, and the uncle will be a much better entry into an informational scene than the random encounter with a talkative pickpocket the GM had been planning. Or maybe he can make the big villain of the scenario into the uncle! Perfectly fine and more fun to GM, frankly.​

​

The shift from suggestions being welcome to being obliged to have regard to suggestions is real, but in many ways I think a matter of degree.

Now to return to your post - whether I would agree that being able to make suggestions in this way counts as manipulating the situation and/or setting as a game piece I'm not sure. What payoff am I getting from going along with you here?

Unfortunately, I'm a bit pressed for time so I'm not able to answer all of the questions/counters that you and @Ovinomancer have. I don't want to short-shrift them though so I'm going to put together a meaty post tonight or tomorrow or the next day when I have more time to read in full, digest, and formulate my thoughts (because I'm working my own way through them in real time).

To answer your last question (that I've bolded) and then propose a quick example though. Hopefully:

* A means for people to understand (a) how Protagonist Agency is made manifest (at the system level and at the play level), (b) how it can manifest DESPITE having considerably less actual Tactical or Strategic Agency that a player might enjoy in another game while not enjoying much, if any, Protagonist Agency (contrast My Life With Master with many forms of D&D 5e), and (b) why some games/play produces it and why some other games/play doesn't produce it.

I feel like a lot of the problem of these conversations is that Protagonist Agency just gets folded into other two and then ASSUMED that if you have x amount of Tactical or Strategic agency, then its inevitable that you have x amount of Protagonist Agency. This is fundamentally not true and I'm hoping that a matrix serves to make this clear.

* I want to clarify (for others but also myself) when, at the actual game machinery/interface level, Character and Setting are just conceptually discrete things...but not actually discrete things. When is it not possible to "pick up the Character Piece" without simultaneously "picking up the Setting Piece".

It appears that in these conversations we've had over the years that some believe that its possible for "on the Venn Diagram of the Vector/Piece/Medium" that (at the actual GAME LEVEL) a player can nearly always just exclusively pick up the Character Piece and make a move without picking up Situation or Setting pieces.

I'm confident that isn't true so I'm trying to build a matrix to (a) show when this is true and when this is not true and (b) therefore show that exclusively picking up the Character Piece is artifice, self-deception, illusion in the course of any given segment of play. I'm presently thinking this is likely only possible to maintain this persistently in very specific low level, highly structured Dungeon Crawls (eg Moldvay Basic) when you're playing a Fighter or a Dwarf...but I'm not certain.

* I want to discretize Tactical and Strategic Agency and how this system/design actuates this in play. Yes, there will invariably be interdependence, but there are enough degrees of freedom at the design level that games like 4e D&D, Mouse Guard, and Dogs in the Vineyard (both predominantly Tactical games) are (a) meaningfully different than games that feature both (either in equilibrium like Blades in the Dark, elegant crawl games like Torchbearer, or wildly out of equilibrium at any given time like 1e/2e/3.x D&D) and (b) how this manifests at the design level of Classes/Playbooks (eg why is the Classic D&D Spellcaster so much more powerful than the D&D Fighter).

* I think this matrix (or something similar) will have explanatory power of how to design games (Torchbearer and Blades in the Dark) which have all 3 of Protagonist, Tactical, and Strategic Agency and none of these things ever become at tension during play. When the design around these 3 things is not robust (but it aims and/or alleges to be) and the play becomes unwieldy, it gives rise to Force manifesting in play as a participant (typically a GM) arrests 1 or 2 of those so that the third can be prioritized and survive that "contact with the enemy." This paradigm shows that there is an actual apex priority in play and that, when push comes to shove (because system hasn't been able to maintain equilibrium and its offloaded on the GM to juggle this), it will win out (because the GM expresses their authority to make it so...typically with sleight of hand/illusion to keep up the pretense that all 3 of these things are actually still in equilibrium...when they're absolutely not).

This, in my opinion, is a HUGE issue with D&D and it hasn't been forensically broken out as to how/why this happens. The Forge tried to tackle this with its "incoherency" model, but that didn't do enough work (or at least the right kind) with most people but it absolutely is a real thing and understanding it would be very good.

---

Alright, a move in play to take apart. I think the Dungeon World Spout Lore move shares a lot in common with a Blades in the Dark Flashbacks, so I want to discuss that move.

Dungeon World - Spout Lore

Vonn's Player Action Declaration - "We've been on the run from this Orc Horde for two days straight. We have to get out of this valley! I don't know about you guys but my Fighter Vonn is exhausted. I've got 2 Debilities (one of them Strength!), I'm at half HPs, I've got 1 Ration Left and 0 Adventuring Gear left! And my Scale Armor is damaged so I'm at 1 Armor instead of 2!

Vonn turns to the Elven Arcane Duelist.

Triellia, your people lived in these mountains before the Orc Horde crashed over them like a tidal wave breaking a shore. You must have heard of a a secret way out!

Triella's Player Action Declaration - "When we fought and lost to these orcs there must have been an Alamo with a secret tunnel or a portal or a chimney to climb to the top of the mountains so the noncombatants could escape. Elves always have contingencies!

Rolls Spout Lore with +1 due to Warfare.

Now if this player gets a 7-9, they're going to get something interesting...its on them to make it useful. If they get a 10+, its going to be something both interesting and useful. Given the games ethos, structure, and the principles that bind and inform the GM's response, a 7+ response is likely going to come from either (a) the GM using this exact prior conversation/speculation by the players (use the answers) as an input or (b) they'll ask further questions (and use the answers). So the player is going to pick up the Setting Game Piece here. If they get a 10+, it will bind the GM to give them something useful as well as interesting, which, combined with ask questions and use the answers, will have the player picking up the Situation Game Piece here.

Interestingly, on a 6-, the constellation of ethos, structure, and principles that inform the GM's response will ALSO have had the player "pick up the Situation Game Piece by proxy (at least...its possible that new Setting may be generated and established as well here)" because their situation will change adversely with the context of prior play!

Obviously they will have picked up their Character Game Piece in order to make the move. So some formulation of Character + Situation or Character + Setting or Character + Situation + Setting will 100 % emerge as a play byproduct (even on a failure!).

Flashbacks are similar.

Thoughts?

 

[Campbell]

As someone who does value a sense of things feeling tangible within the game one of the hardest lessons for me personally to come to grasp with as a GM is that I have limited cognitive threshold. I can only contain so many things in my head at the same time. Given that I lack infinite cognitive capacity and cannot hold an entire city in my head and notes much less a world I need to decide which elements to focus on mentally and in my prep. Many GMs make the decision to focus on physical space and moving through it while not focusing much on characters as like people. I take the opposite approach. I tend to elide space in the interests of focusing my mental attention on the characters closest to the players' characters (emotionally close that is).

My NPC constellation approach to prep for character focused games is born from that - acknowledging my own limitations. I think when it comes to game design and technique discussions that we are able to acknowledge our cognitive limits. We cannot find the best way to utilize that cognitive threshold until we really know that we have limits.

Speaking as someone who has acting experience even trying to contemplate a single person's lived experience that is different from my own is mind boggling. Like there is so much of my own lived experience that I do not remember or have suppressed. Doing that for millions of souls is absolutely impossible. So I have learned to pare down and focus my efforts on creating basic profiles I can use to frame scenes with the players and really focus on what an NPC wants right now.

 

[FrogReaver]

On shared fiction and establishing details I think the very word shared precludes anything that's not been shared with the players from being established in the shared fiction. But let's really take a look at what shared fiction is and how it comes into being.

1. I would say in a game each player has their own fiction that they are imagining. There's a process to establish what major and important details are the same amongst them. That's the shared fiction. The minor and trivial details work differently. I will discuss them later. I think the key take away is that the players each have their own individual fictions even when they are basing them on the shared details that we call the shared fiction.

2. The question then arises, what are the various processes to establish something in the shared fiction?
The one process I want to focus on is the one where the DM maintains a separate fictional space that's separate from the shared fiction and reveals details from that fictional space as the PC's encounter them. The PC's actions form a feedback loop into that separate fictional space that affect change there which then gets revealed to they players as the encounter those elements and so on.

In this particular area, the fiction does exist as fiction apart from the players. It preexists their shared fiction. So while it's not yet established as shared fiction, it's still the fiction that their shared fiction is based upon.

3. The question arises, since the above is just about the important details, what happens with all the minor/trivial details. There's various ways of handling them but I want to focus on one and the implications of it. In a game where the players don't have full knowledge of all the fiction that exists they cannot add a minor or trivial detail to the fiction on their own as they can't ever be certain that such a detail is actually trivial or minor.

Note: I think they may very well add trivial and minor details in their individual own fictions but pushing these details into the realm of the shared fiction can cause problems in a game using this style of shared fiction generation.

Also Note: this isn't the only process for generating shared fiction and so the conclusions for what works for this shared fiction generating method don't necessarily apply to others.

 

[hawkeyefan]

This isn't how many of us conceive of play. Shared fiction isn't a term I've ever been able to embrace. It just seems loaded and feels like it is part of the problem in a discussion like this one (moving things towards a specific way of thinking about setting and play). And scenes are not how everyone things of play either. I never think of what is going on as a 'scene'. I have done so for certain play groups or when playing certain games (like Essoterrorists for example). But generally speaking, scenes are not how I think of a play (and you can see the difference in how people who use that language in this thread approach play, versus those who don't).

I’d rather not argue the terms used. If the game doesn’t produce a shared fiction for you, then I’m confused, but go ahead and pick a term that works for you.

And if your game doesn’t involve the GM saying things like “You find yourselves in a tavern....” or “The rain has left the road to Shang Lo a muddy mess, and by the time you arrive, you’re soaked and tired and filthy....” then again, I’d be confused. This is all I mean by “scene”. Again, insert your own term and then let’s talk.

The possibility existed before it was found. There is a reality that exists in the game outside what the players are hearing from the GM. And once found, that painting is either supernatural or it isn't. It doesn't become supernatural the moment its nature was revealed to the party (unless the GM is ad libbing everything).

But that’s exactly what happened in play. The painting didn’t exist until the GM described it. Then it didn’t become important until the player decided to examine it further.

And it’s not solely that the GM is ad libbing. It’s also that the player is initiating something.

In other words, the fiction is not solely what is established by the GM and discovered by the players.

The idea that everything is fuzzy and in flux until the players see it or encounter it, to me runs very counter to how a typical sandbox and living world GM would think of things in the setting. Much of the point is to create a sense of a real concrete world outside the player characters. And a key way of doing that is trying to be consistent with these kinds of details. Now the parts can move around (an NPC doesn't have to be in the same room waiting for the party to get there: this is actually one of the main examples used in "Living Adventure" discussions: the NPCs are alive and move around with their own goals, like PCs). But the point is the GM is meant to take things like NPC motives and goals, very seriously when deciding what that NPC is doing, and where it is (not simply decide arbitrarily that the NPC is in location Z because that is convenient for some other goal----like having a plot or something)

These are fine as principles to guide your play. But they don’t apply to all styles. In your attempt to make sure that your approach is included, you’re trying to discard descriptions that fit all play in favor of one that fits only your own.

There is no game fiction prior to being introduced to the game. Prior to that, it exists only as a possibility.

In the mind of the GM, in the notes of the GM, and in the logic that the GM uses to extrapolate details and determine accidental qualities of the things in the setting. If I decide Rupert the Elf lives at 23 Chestnut Street on Monday, and the players ask me who lives at 23 chestnut street on Thursday, his existence there wasn't fluid prior to them asking.

What if you change your mind on Wednesday and you decide that It’s not Rupert who lives there but Mortimer the Troll? Then you proceed to play on Thursday where this gets established through play.

Is there some kind of paradox? Who lives at 23 Chestnut? What is true in the shared fiction or fictional world?

 

[Ovinomancer]

On shared fiction and establishing details I think the very word shared precludes anything that's not been shared with the players from being established in the shared fiction. But let's really take a look at what shared fiction is and how it comes into being.

1. I would say in a game each player has their own fiction that they are imagining. There's a process to establish what major and important details are the same amongst them. That's the shared fiction. The minor and trivial details work differently. I will discuss them later. I think the key take away is that the players each have their own individual fictions even when they are basing them on the shared details that we call the shared fiction.

2. The question then arises, what are the various processes to establish something in the shared fiction?
The one process I want to focus on is the one where the DM maintains a separate fictional space that's separate from the shared fiction and reveals details from that fictional space as the PC's encounter them. The PC's actions form a feedback loop into that separate fictional space that affect change there which then gets revealed to the players as the encounter those elements and so on.

In this particular area, the fiction does exist as fiction apart from the players. It preexists their shared fiction. So while it's not yet established as shared fiction, it's still the fiction that their shared fiction is based upon.

3. The question arises, since the above is just about the important details, what happens with all the minor/trivial details. There's various ways of handling them but I want to focus on one and the implications of it. In a game where the players don't have full knowledge of all the fiction that exists they cannot add a minor or trivial detail to the fiction on their own as they can't ever be certain that such a detail is actually trivial or minor.

Note: I think they may very well add trivial and minor details in their individual own fictions but pushing these details into the realm of the shared fiction can cause problems in a game using this style of shared fiction generation.

Also Note: this isn't the only process for generating shared fiction and so the conclusions for what works for this shared fiction generating method don't necessarily apply to others.

Sure. But, again, looking at this from the perspective of player agency, how can you make this argument and also say that the players do not have less agency in the game? If the GM's fiction is the authoritative one, then the GM has that agency and the players do not. If the players can contest that, somehow, then they gain some agency. Even allowing for the argument that the players can exert some agency by convincing the GM, either though entertainment or compelling argument, is it not obvious that this is a lesser agency than if the players can exert changes to the shared fiction without having to convince the GM?

This has been the point in this argument the entire time. As @Campbell, and I, and @Hawkeye have continually made the argument, this fact doesn't mean the game is lesser or not fun -- we all play in this regime for some of the games we dearly love. It's just a feature of this approach to gaming, and is offset by non-agency consideration that make it a worthwhile trade-off.

 

[Campbell]

I expect the following might be slightly contentious. A fundamental part of the understanding of a shared fiction is that nothing is true until all the players at the table have accepted it as true. That even the GM is susceptible to a basic credibility test. In D&D the DM is given broad authority to establish what is true in the fiction, but this may be overridden at the table on the social level.

I have personally seen this at the table in our D&D game.

DM : This is true.
Player: What about X? How about Y?
DM: naughty word. Actually this is true.

 

[hawkeyefan]

2. The question then arises, what are the various processes to establish something in the shared fiction?
The one process I want to focus on is the one where the DM maintains a separate fictional space that's separate from the shared fiction and reveals details from that fictional space as the PC's encounter them. The PC's actions form a feedback loop into that separate fictional space that affect change there which then gets revealed to they players as the encounter those elements and so on.

In this particular area, the fiction does exist as fiction apart from the players. It preexists their shared fiction. So while it's not yet established as shared fiction, it's still the fiction that their shared fiction is based upon.

What you’re describing here is a process where the GM decides and then the players discover these decisions through play.

This is what @pemerton describes as “playing to find out what the GM has in his notes”, which is usually a description that sees some hard pushback. The only distinction you’re making is that the GM is free to change his notes based on what the players do.

But you are right it is indeed one way that the shared fiction can be established.

 

[FrogReaver]

Sure. But, again, looking at this from the perspective of player agency, how can you make this argument and also say that the players do not have less agency in the game?

Because the game isn't the sum total of establishing shared fiction about the setting and it's non-player inhabitants.

If the GM's fiction is the authoritative one, then the GM has that agency and the players do not.

It's not that the simple. As I said, there is a feedback loop where the PC's doing things alters the GM's separate fictional space. The existence of that feedback loop does give players some agency via their PC's actions.

If the players can contest that, somehow, then they gain some agency. Even allowing for the argument that the players can exert some agency by convincing the GM, either though entertainment or compelling argument, is it not obvious that this is a lesser agency than if the players can exert changes to the shared fiction without having to convince the GM?

You are trying to look at a particular area and extrapolate that less agency in this area means less agency in the whole game. It should be obvious that such a statement isn't necessarily true and needs proven. It goes back to the whole focus on 0 sum games much earlier in this thread.

Heck, it also brings up the questions of how do we actually measure quantity of agency - which I'll note that no one saying game X has more agency has ever actually provided a method or agreed to mine.

It also brings up the question of whether measuring quantity of total agency is actually meaningful - or is it the particular kinds of agency and quantities of each kind that is meaningful.

This has been the point in this argument the entire time. As @Campbell, and I, and @Hawkeye have continually made the argument, this fact doesn't mean the game is lesser or not fun -- we all play in this regime for some of the games we dearly love. It's just a feature of this approach to gaming, and is offset by non-agency consideration that make it a worthwhile trade-off.

You may be right. I don't think either of us have conclusively proven our cases. But I think it's alot more complicated than you are trying to make it out to be - especially given some of the basic questions being left unanswered and the roadblocks that have been erected toward even attempting to think about the problem in terms of types/vectors of agency.