Dealing with agency and retcon (in semi sandbox)

[ZebraDruid]

I think it's safe to say that there's a lot of contention over what constitutes making a meaningful decision (vs fateful) and things like player agency. Jumping into a dark hole is probably fateful, but a lot of people would say it's not meaningful because it offers no sense of what the consequence will be so the apparent meaning of the decision to jump in is obscured. Thus the character doesn't really have full agency despite being in control of their PC the whole time.

That said, I'm in the camp that player agency does not require a full understanding of the implications of any decision being taken. In the OP case, the PCs knew that without intervention there would be a ritual performed over a dead body to raise him and that it would have some negative consequences for the dead guy's heir (delay of inheritance if nothing else) and were suspicious of the whole affair. I don't know why, if I were a player in that game, I would need more to make a decision in which I felt I was in full control of my PC, whether I decided to intervene or not.

I think that brings a little light onto the nuance of the issue. Also this is what I'm getting so far. (not specifically addressed to you)



----
Excuse me if I sound like stereo instructions.

Player agency doesn't matter (as much) if there are no meaningful consequences to be had by their actions/inaction.

Player agency doesn't require specific knowledge of potential consequences to be had by their actions/inactions, but

Player agency is best respected when knowledge of consequences can be inferred or discovered.

Real violation of agency is based on the relative degree of consequences/rewards caused by their actions/inactions, and further relative to a reasonable level of knowledge of consequence obtainable by the player prior.
----
In the case of the dark hole.

The player is exploring in a cave, and finds a hole. The dark hole appears to look like certain death. The player inspects the hole, throws down and torch, and it fades away into darkness. The player jumps head first and after a long fall dies upon impact.

(Knowledge) A dangerous non descript looking hole that appears to be almost bottomless. With no emerging reason to jump in.
(Action) The player jumps in.
(Consequence) The player falls hundreds of feet in darkness (having no magic to slow themselves, or otherwise slow their decent) impacts on the ground and dies.
----
In this case


(Knowledge) Shady noble wants body taken from crypt, later discovered to be necromantic in nature (Runes, skulls, blood, zombies).
(Action) Body is retrieved, and party follows noble.
(Consequence) Body is taken into another graveyard crypt

(Knowledge) Noble doesn't want people attending ritual, besides suspected evil cleric after being convinced.
(Action) Party waits outside while ritual is performed
(Consequence) Body is resurrected

(Knowledge) Dead father is walking out of crypt looking dismayed and quiet, noble is trying to shoo people off
(Action) Party takes money and leaves for Inn, and leaves the necromancer and his group of hooded figures to their business
(Consequence) Necromancer goes to family home and with their pact bound fathers help, slays/captures the family.

(Knowledge) ?? Paladin feels something is off? Feels guilty?
(Action) Goes to his church and confesses crime to high priest the next morning.
(Consequence) High priest consoles him and says he will go investigate the family.

(Knowledge) Family is discovered slain (father and son missing)
(Action) Paladin confesses to guard, and the diviner investigator wizards, and divulges ritual location.
(Consequence) Guard, and wizards use divination magic to investigate the ritual site and hunt down perpetrators (including the party)
----

Does this seem like a rational progression of events?

While I want situations to be mysterious at first, I never want to engage in whole heartedly into Shmuck bait

 

[pemerton]

Having agency is based on there being meaningful consequences of your actions, not being able to declare the consequences of your actions

I didn't say anything about declaring the consequences of your actions. I used the phrase capacity to establish what is at stake.

I also pointed to some RPGs that exemplify what I have in mind (Burning Wheel, Dungeon World). Of course they're not the only ones.

Conversely, if everything that is at stake, and everything that flows from both successful and unsuccessful actions, is established by the GM, then the players are not exercising much agency.

The obvious counter-example to your post is a coin-toss to determine whether the imaginary universe endures or comes to an end: the GM could set up that situation, and then call for the players to toss the coin.

A more low-scale version: the GM tells the player they stumble on an assault (A vs B) and thereby prompts them to intervene. Perhaps A is the heir to the throne, who if they survive the assault will go on to murder their parent and take over the realm, establishing a dictatorship. Perhaps B is a revolutionary, trying to kill A so as to establish justice in the realm. But if the players don't know any of this backstory, and simply have to decide whether to help A, whether to help B, or whether to pass on without intervening, the players have not exercised any agency, beyond the most minimal of declaring some actions for their PCs and then waiting to find out what the GM has decided will flow from those.

@ZebraDruid's scenario, as described in the OP, seems more like this than like high-agency play.

If I attempt to lockpick open a chest it is my agency to declare that i am trying to open it, it is not determined by the hypothetical contents of it, If it is trapped, If someone sees me doing it or even if i actually succeed the check but my ability to attempt to open it and the fiction following meaningful logical consequences from my attempts to do so.

Picking locks and searching for traps in a context where the GM is establishing everything that matters about the chest, whether or not it is trapped, etc, is classic low-agency play.

 

[pemerton]

I can write an agenda for an NPC that can then intersect with what the PCs choose to do. Just as PCs can react to player actions, so to can NPCs. It’s only a problem when PCs come up with reasonable things to prevent something and the DM forces it through regardless.

I'm not posting about what is or isn't "a problem" in general.

I've made three points in this thread.

One is that the OP describes low-agency play.

The second is that their are reasonably well-known RPGs that will cover the same sorts of tropes, themes etc @ZebraDruid seems interested in that will better support high-agency play.

The third is that, in low-agency play (aka railroading), the game will tend to break if the GM doesn't make clear to the players what decisions they are expected to make for their PCs. The OP is an illustration of this point.

To relate the third point to your post: suppose that the GM decides that NPC A, if unthwarted, will bring the world to an end. And suppose also that the GM keeps this secret from the players. Suppose the GM even goes out of their way to present NPC A to the players as a reasonable person (they are the friendly quest-giver, the sponsor of the PCs' exploits, etc). So the players never declare actions for their PCs that would thwart NPC A's plans and undertakings.

The upshot will be a game in which the players go along, doing as they think they are meant to do, only to have the (imaginary) world end. At some tables that may be satisfying RPGing. I think at many tables it might be experienced as a variant on "rocks fall, everyone dies". (The OP suggests something closer to the latter than the former for that table at least.) However the table experiences it, it is clearly super-low player agency RPGing.

 

[ZebraDruid]

I didn't say anything about declaring the consequences of your actions. I used the phrase capacity to establish what is at stake.

I also pointed to some RPGs that exemplify what I have in mind (Burning Wheel, Dungeon World). Of course they're not the only ones.

It sounds like semantics. When a PC chooses to unlock a chest, they have the capacity to establish what is at stake. Their health if it's trapped, their reputation if they're stealing, their potential reward if it has treasure inside.

If the consequences of a player's declared actions are decided by someone else (eg the GM) by reference to stuff that is more-or-less outside the player's control (eg secret backstory that only the GM is privy to), then where is the agency?

In high player-agency RPGing, the players have a lot of capacity to establish what is at stake. In the OP, the players seem to have had almost no say over what is at stake.

It also seems like you said both things? If not, then who but the DM declares the consequences of your actions? Another player? I'm not sure of the DM's purpose if that is the case.

Conversely, if everything that is at stake, and everything that flows from both successful and unsuccessful actions, is established by the GM, then the players are not exercising much agency.

I'm confused, who else would establish what flows from a successful or unsuccessful action, besides the DM? If the player determined what followed their successful or unsuccessful action. Then a scenario like this could exist.

Player is grabbed by zombie.
Player attempts to wriggle free from zombie.
Player rolls athletic check to see.
Player fails roll.
Player decides zombie released them from their grip anyway
Player is disappointed of not being able to escape on their own, but they never give up!
Audience cheers.

Player opens chest
Player says the chest has 1000 gold pieces in it.

The obvious counter-example to your post is a coin-toss to determine whether the imaginary universe endures or comes to an end: the GM could set up that situation, and then call for the players to toss the coin.

That is tough, but I think my point above actually stands to this very well.

"Real violation of agency is based on the relative degree of consequences/rewards caused by their actions/inactions, and further relative to a reasonable level of knowledge of consequence obtainable by the player prior."

In this scenario there is too much risk, for such a small amount of action. It isn't relative at all, which is why no one dice roll, or single action resulted in the party being caught, but several actions in a row.

A more low-scale version: the GM tells the player they stumble on an assault (A vs B) and thereby prompts them to intervene. Perhaps A is the heir to the throne, who if they survive the assault will go on to murder their parent and take over the realm, establishing a dictatorship. Perhaps B is a revolutionary, trying to kill A so as to establish justice in the realm. But if the players don't know any of this backstory, and simply have to decide whether to help A, whether to help B, or whether to pass on without intervening, the players have not exercised any agency, beyond the most minimal of declaring some actions for their PCs and then waiting to find out what the GM has decided will flow from those.

I'm not sure this applies. Especially when a party member in the party would support side A, and another party member would support side B.

The cleric wanted the ritual to happen. The paladin did not. I wanted to respect both of their agency by providing a situation where either of them could triumph, and both of them have something at stake. In this instance the cleric (evil) won.

The paladin didn't even attempt an action to be in the fight however.

And the more serious consequences arouse 'directly' from the paladins action of snitching.


Remember, the paladin also had a chance to attack them when they came out of the crypt, and another chance to stop them before they went to the house.

Using your example, side A won the fight, but the leader of side B fled from the battle. Side A started shouting about glory to the supreme dictator, and started to chase after the leader of side B.

The players ignored the chasing warriors who went off to kill side B leader.

Later on side A dictator passes a law saying no more adventurers in his land.
A player goes and tells the dictator he and his friends are in fact adventurers, and where to find them.

@ZebraDruid's scenario, as described in the OP, seems more like this than like high-agency play.

Picking locks and searching for traps in a context where the GM is establishing everything that matters about the chest, whether or not it is trapped, etc, is classic low-agency play.

I'd really like to know what is high agency play when regarding searching for traps and picking locks.

 

[pemerton]

In my example earlier I described a deep dark hole, and how there are consequences to jumping into it. The consequences might seem obvious, but not always. Just like any NPCs motivations may seem obvious.

If what you say is true, then a player who jumps into a dark hole should be able to determine the consequences, not the DM.

@ZebraDruid, I get the impression you are not very familiar with non-railroading RPG play.

What I say is true. I know it is true from many years - decades - of experience GMing high-agency RPGing, and paying close attention to what distinguishes it from low-agency RPGing.

In your example of a "deep dark hole" whose raison d'etre is established by the GM, whose nature is established by the GM, and in which the players make an essentially blind choice as to what to do - then the play is low-agency. It may be good play or bad play - that's a further question - but it's not play in which the players are exercising meaningful control over the content of the shared imagining. At best they get to prompt the GM to say one thing (if they declare their PCs jump into the hole) or another (if they declare their PCs walk on by the hole).

if the backstory is discoverable before the consequences manifest themselves, but the players choose not to, do not find, or otherwise ignore that information, are they still being robbed of agency? Is player knowledge of consequences required to respect the players agency?

What you are describing here is a type of puzzle play: the secret backstory is a puzzle, and the players are expected to unravel it via their action declarations.

The most classic example of this is a Gygaxian dungeon. It relies on many features to work: the environment is near-static until the PCs interact with it, and hence scouting and planning are highly viable strategies; the environment is sufficiently austere that it is possible for the players to learn it; the environment conforms sufficiently to certain shared tropes and expectations (eg around architecture, furnishings, NPC/creature motivations, etc) that the players can make reasonable conjectures about it. Also, the players are largely in control of framing and pacing (they choose which doors to open, when to leave the dungeon, etc).

The more that a GM's puzzle-backstory varies these parameters, then the more unreasonable it becomes to expect the players to solve it. The most obvious variation in the OP is that the players are not in charge of pacing and framing: play begins with a "quest giver". The players, not unreasonably, treat this as a GM hook that they are expected to "bite" at.

If I use my trap metaphor. If they walk down a hall without doing a perception check, and spring a trap, and get hurt, are they robbed of agency?

We can't answer this question in the abstract. In Gygaxian play, if this is a hall in the dungeon then traps are fair game, and players are expected to declare various precautions taken by their PCs. On the other hand, in Gygaxian play, if this is a hall in the tavern where the PCs rest betwee forays into the dungeon, it looks more like a "gotcha" to me.

In a non-Gygaxian game, suppose that the quest-giver invites the PCs back to their rooms to spell out the details of the quest. The players, believing (not unreasonably) that this is part of biting on the hook, have their PCs agree. They don't declare that they look around for traps in the parlour, for poison in the tea, etc. In that situation, if the GM springs a trap and then responds to the cries of "unfair" or "railroad" that the players should have searched their surroundings, well my sympathies are 100% with the players. How were they supposed to know that what the GM was presenting as a hook was in fact part of play? From their point of view it's all arbitrary. And that is one hallmark of low-agency play.

I'm deciding that a tree will fall in 30 minutes due to wind. If the PCs come and brace it, it won't fall (possibly based on a roll). I'm not taking away their agency to stop the tree from falling, especially if they know it is a windy day. If they choose not to inspect the tree to see its' weak roots, or that it is swaying in the wind and it falls. And choose to do nothing as it starts to tip. I can't control that, nor would I want to.

What is the significance of the tree? Why do, or ought, the players to care about it? Why have you, as GM, narrated a strong wind? This example is so underdescribed we can't tell anything about whether or not the players are exercising agency.

if I'm purely creating situations based on the moment without context, arbitrarily make a tree start to fall as they pass just to give the players a chance to stop it. Which I don't think is very interesting or complex.

Well, I've got dozens - maybe hundreds - of actual play posts on these boards. I'll link to one thread - https://www.enworld.org/threads/torchbearer-2e-actual-play-of-this-awesome-system.691233/ - and you can tell me whether or not the play I describe meets your standards of interest or complexity: .

To be honest, this comment reinforces my impression that you are not very familiar with non-railroading RPGing.

 

[pemerton]

It sounds like semantics. When a PC chooses to unlock a chest, they have the capacity to establish what is at stake. Their health if it's trapped, their reputation if they're stealing, their potential reward if it has treasure inside.

<snip>

who but the DM declares the consequences of your actions? Another player? I'm not sure of the DM's purpose if that is the case.

<snip>

I'm confused, who else would establish what flows from a successful or unsuccessful action, besides the DM?

<snip>

I'd really like to know what is high agency play when regarding searching for traps and picking locks.

Different RPG systems take different approaches.

Here's an obvious one (and it is the approach taken in Burning Wheel): the player establishes what flows from a successful action. And when narrating the consequences of an unsuccessful action, the GM has regard to what the player put at stake in their action declaration.

Here's another (and it's the approach that I take in Torchbearer): if the GM decides what it is that the PC finds when they search the chest, the GM has regard to what the player has put at stake (via their build and their play of their PC) in deciding what that is.

Here's a Burning Wheel actual play example (taken from an actual play report posted on rpg.net):

One of the players had bought rulebooks and built a BW PC (a noble-born Rogue Wizard inspired by Alatar, one of Tolkien's blue wizards of the East). I had built a PC for another player to show him what the system was capable of - a spell-using necromancer ranger/assassin (hunter-wizard's apprentice-rogue wizard-bandit).

<snip>

Writing up beliefs took a little while. The rogue wizard, Jobe, had a relationship with his brother and rival. The ranger-assassin, Halika, had a relationship, also hostile with her mentor, and the player decided that was because it turned out she was being prepared by him to be sacrificed to a demon. It seemed to make sense that the two rival, evil mages should be one and the same, and each player wrote a belief around defeating him: in Jobe's case, preventing his transformation into a Balrog; in Halika's case, to gain revenge.

Each player also wrote up a "fate mine"-style belief: He who dares, wins for the sorcerer, and Stab them in the back for the assassin. And each also wrote up a immediate goal-oriented belief: I had pulled out my old Greyhawk material and told them they were starting in the town of Hardby, half-way between the forest (where the assassin had fled from) and the desert hills (where Jobe had been travelling), and so each came up with a belief around that: I'm not leaving Hardby without gaining some magical item to use against my brother and, for the assassin with starting Resources 0, I'm not leaving Hardby penniless .

Some instincts were written up too: the ones that (sort of) came into play were, for the mage, When I fall I cast Falconskin and, for the assassin, I draw my sword when startled.

<snip>

I started things in the Hardby market: Jobe was looking at the wares of a peddler of trinkets and souvenirs, to see if there was anything there that might be magical or useful for enchanting for the anticipated confrontation with his brother. Given that the brother is possessed by a demon, he was looking for something angelic. The peddler pointed out an angel feather that he had for sale, brought to him from the Bright Desert. Jobe (who has, as another instinct, to always use Second Sight), used Aura Reading to study the feather for magical traits. The roll was a failure, and so he noticed that it was Resistant to Fire (potentially useful in confronting a Balrog) but also cursed. (Ancient History was involved somehow here too, maybe as a FoRK into Aura Reading (? I can't really remember), establishing something about an ancient battle between angels and demons in the desert.)

<snip>

Jobe, having both nobility and sorcerers in his circles, and a +1D affiliation with both (from Mark of Privilege and a starting affiliation with a sorcerous cabal), initially thought of trying to make contact with the Gynarch of Hardby, the sorceress ruler of that city. But then he thought he might start a little lower in the pecking order, and so decided to make contact with the red-robed firemage Jabal (of the Cabal). With Circles 2 he attempted the Ob 2 check, and failed.

So, as the 3 PCs were sitting in the Green Dragon Inn (the inn of choice for sorcerers, out-of- towners and the like), putting out feelers to Jabal, a thug wearing a rigid leather breastplate and openly carrying a scimitar turned up with a message from Jabal: Leave town, now. You're marked. Halika noticed him looking at the feather sticking out from Jobe's pouch as he said that: it seemed that the curse had already struck!

The player establishes what is at stake: the utility of the feather for fighting a balrog. The player also establishes the overarching context within which the scene is framed - being in Hardby looking for things that are useful for fighting a balrog.

The outcome of the failed check - the feather is cursed - is further incorporated into elements of the situation established by the player: the relationship with the Cabal, and his status as a minor illusionist with a vexed history.

Traps and locks can be handled similarly.

 

[ZebraDruid]

@ZebraDruid, I get the impression you are not very familiar with non-railroading RPG play.

What I say is true. I know it is true from many years - decades - of experience GMing high-agency RPGing, and paying close attention to what distinguishes it from low-agency RPGing.

That's great! But you seemed to have snipped out a crucial piece of context to my core question.

If I use my trap metaphor. If they walk down a hall without doing a perception check, and spring a trap, and get hurt, are they robbed of agency? The nuance, and question here, is when does it become not the same as a hallway with a trap.

We can't answer this question in the abstract. In Gygaxian play, if this is a hall in the dungeon then traps are fair game, and players are expected to declare various precautions taken by their PCs. On the other hand, in Gygaxian play, if this is a hall in the tavern where the PCs rest betwee forays into the dungeon, it looks more like a "gotcha" to me.

Using your example, at what point was their dealings with a necromancer more akin to a tavern trap, than a dungeon trap?

What is the significance of the tree? Why do, or ought, the players to care about it? Why have you, as GM, narrated a strong wind? This example is so underdescribed we can't tell anything about whether or not the players are exercising agency.

It's a metaphor. Looking at the core principle of cause and effect, player agency, and responsible restriction of the DMs carte blache. (I can use french words too...)

Well, I've got dozens - maybe hundreds - of actual play posts on these boards. I'll link to one thread - https://www.enworld.org/threads/torchbearer-2e-actual-play-of-this-awesome-system.691233/ - and you can tell me whether or not the play I describe meets your standards of interest or complexity: .

"I described Asfaloth as stopping suddenly, almost having hurt himself at a hole in the ground. They inspected the hole and decided to lower themselves down. This Dungeoneering check failed, and so they fell. Golin succeeded at the subsequent Health check, even though he was the first (so Ob 3); Fea-bella failed, and so was injured with a sprained ankle."

If we're ignoring the thing you snipped out that I said.
"The nuance, and question here, is when does it become not the same as a hallway with a trap"

If the consequences of a player's declared actions are decided by someone else (eg the GM) by reference to stuff that is more-or-less outside the player's control (eg secret backstory that only the GM is privy to), then where is the agency?

In high player-agency RPGing, the players have a lot of capacity to establish what is at stake. In the OP, the players seem to have had almost no say over what is at stake.

In this instance, Fea-bella had no agency, because she didn't know there was a hole in the ground, and thus sprained her ankle. Her sprained ankle was out of her control, as it was determined by the DM. She failed a roll, but it is irrelevant because she didn't have the agency to decide whether or not her stepping in the hole would cause her ankle to sprain.

Do you understand what I mean by nuance, and what my actual question is? In that post it was posing a theoretical discussion. Not a direct practical application. The trees purpose is irrelevant.

To be honest, this comment reinforces my impression that you are not very familiar with non-railroading RPGing.

To be honest, I'm not sure you're reading what I'm saying.

I've started DMing for a group of 4 close friends. I'm fairly new to it, but most have experience with TTRPG or at least CRPGs.

 

[Thomas Shey]

So firstly I think your NPCs backstory and circumstances are really good and interesting. Ignore folks that say you can’t play gotcha stories on your players. Players are almost going to be suspicious of all but the most banal of things… it’s good… it gets them investigating an questioning things.

When it doesn't just teach them to practice avoidance about any presented opportunities of any stripe.

 

[Thomas Shey]

It sounds like semantics. When a PC chooses to unlock a chest, they have the capacity to establish what is at stake. Their health if it's trapped, their reputation if they're stealing, their potential reward if it has treasure inside.

You're showing the problem with this by your very phrasing: "if". If they have no idea whether it will be or won't, they aren't setting stakes at all; they're interacting with a black box. If you have no way to establish the benefits or costs actually associated with an action, you're not setting any stakes at all; you're simply accepting whatever ones the GM has, blindly.

 

[hawkeyefan]

I think that often when it comes to RPGs, there's fuzziness around the idea of player agency. It's likely due to several factors, but there are two that seem most relevant to me, and I think they're connected.

First, I don't think the simple act of taking part in the game... the ability to play... is enough to constitute agency. In an RPG, this entails declaring what your character does. This alone should is a bare minimum, and its presence isn't sufficient to establish agency. Too many people seem to think this is enough to establish agency. It's not.

Second is the perceived need to separate character knowledge and player knowledge. This is something that generally doesn't exist in other games. In almost any other game, you're a participant in the game, and so you know what's going on, what may happen at any time, what you are allowed to do, what the stakes are... and so on. There may be information that you're not privy to... let's say like an opposing team's playbook in football, for example... but the range of possibilities are known. You know what the other team can do, and you likely know their strengths and weaknesses.

Look at almost any other game other than pure games of chance, and the player is aware of the odds and the stakes and so on. In chess, you can literally see the entirety of the board... yet you can still miss an opponent's move, or be unable to stop it. In poker, there are cards that you cannot see... but there are cards you can see, and you know what's in the deck, and what cards are likely, and so on... you have the ability to act in an informed manner.

That's agency. Being able to actually play the game, just as you'd expect in almost any other game.

Apply this weird restriction to any other game, and you immediately see it. If you could somehow limit the knowledge of participants in other games so that they could not know the opponents' strengths, or could not know what cards are in the deck, or were not allowd to know how many yards they had to get to make a first down, or what inning it was or how many outs there were... such restrictions would be considered awful for the state of a game.

Yet people do it in RPGs all the time.

This whole "meta" angle really needs to be dropped if you want to promote agency. Simply share information with the players. Let them make informed decisions. They are players in a game... let them be so. I know this often leads to cries of "RP is more important than the G" and "role play rather than roll play" but that's all garbage. Role playing is not diminished by sharing game information.

If you want to allow agency, you have to empower players. And as with any other game, they need to understand the circumstances, odds, and stakes at any point of play.

If that's not something that appeals, then you're not interested in player agency.

And this is before we even really get into the mechanics and processes of play. Sharing information with the players so that they can then use that information to make meaningful decisions is the first step toward actual agency.