GM fiat - an illustration

[Crimson Longinus]

It does seem to have at least this much to do with what the player is interested in: namely, they do not seem to be interested in having their PC accede to the Duke's request; and they do seem to be interested in whatever events might happen if they have their PC go west.

I would not be so sure of that. People decide what their character does based on other reasons too. The player might decide their character refuses the duke's offer because they (the character, no the player) dislikes the duke. Similarly the character might decide to join the pirates, because they're free-spirited, morally flexible, like fresh air and are in need of money.

 

[thefutilist]

It does seem to have at least this much to do with what the player is interested in: namely, they do not seem to be interested in having their PC accede to the Duke's request; and they do seem to be interested in whatever events might happen if they have their PC go west.

Well it depends on whether you mean it in a tautological sense.

Could there be a situation where the player is interested in seeing what's going on to the west but playing to the fidelity of their character means they 'have' to accede to the Duke's request?

 

[bloodtide]

.Trying to examine player agency in an RPG through the lens of that of characters in novels or movies doesn't really work since the media are fundamentally different.

This is very important: They are fundamentally different. Unlike most though, I expand it to everything, like the PCs being "main characters" .

I think when players complain about a railroad, it is exactly about them not being able to introduce an element of the fiction... an element where they decided what they wanted to do instead of having to do what the GM has said they must do.

Well, as the Railroad Tycoon I can tell you the number one reason players complain about a 'railroad' is simply the game play is too hard for them.

Like what sparked this branch of the conversation was @pemerton describing the bare minimum amount of player agency for a game to be considered an RPG... the player being able to ask questions and have the GM provide answers.

As a near No Questions type DM, I'd guess my game has zero player agency. Not that any player really notices. No player wastes game time asking me questions and we have fun playing the game.

I don't think any of this radically exceeds what is in trad play, except insomuch as the expectation that the material the GM introduces will generally address the input of the players via the above channels. So the GM is not going to insist on running undead infested dungeon crawls when the players all styled their PCs as pirates located on an island of pirates.

Well, a bit very radical........

Your massive list of 'session zero work' is not found in my games.

Let's start with this - can we all agree that different fictional situations (and the position of the player characters within) provide different levels of agency over the fiction? That for instance, a player who is playing a political dissident serving a jail sentence would have less agency than the player who is playing a prison guard. Less autonomy, less authority, less ability to influence what is going on.

Or that say a scenario with a lot of GM established backstory that impacts the players ability to influence outcomes but is not meaningfully discoverable would result in less agency where there aren't these backstory mines to step on?

I agree. Though I would go much farther. That the "average" character is a nobody with no real power or control over anything.

Do you look at the two examples from the OP and see how one could be concerning when it comes to railroading and the like, and the other is less so?

Wait, that OP was an example of what you call "railroading"?

 

[Micah Sweet]

I would call Clue just a pretty basic logic puzzle. I recall realizing this around age 6. If you index all the answers to all the players guesses you can derive the contents of the envelope. It is really just an exercise in note-taking at that point. It's about as much a 'mystery' as working out the product of 2 10 digit numbers, you just grind through an algorithm and get the answer.

Clue does have a game component, rolling dice and disrupting other players be guessing their character into the wrong room, but the solution to the game is just logic.

What exactly would you define as a mystery then? If you presume that an objective answer exists and could conceivably be found if the right information is discovered and correctly interpreted, how is that not a mystery?

 

[pemerton]

In my recent game something happened which very much I think speaks to the issue in your OP re DM fiat with the Alarm spell.

2 NPCs charged with tailing and bringing a PC to a named NPC of Sigil (Estavan) went terribly wrong.
PC killed one of his tailers, the other managed to paralyse the PC and took his backpack informing him if he wanted it back he could find it at Estavan's manor. The backpack had 1 particularly valuable item within.

Some hours later the PC rocked up at the manor.
Estavan had the time, resources (Consortium merchant) and the capabilities (Ogre Mage) to figure out the most valuable item in the backpack. Estavan allowed the backpack to be returned but without the item.
Negotiations were initiated. Estavan wanted the PC to remain in Sigil to find and investigate the other two doors from the 2e module Doors to the Unknown. This was an impossible task for the PC as he had to return home...
After some tough negotiations, the PC agreed to answer Estavan's questions on the 2nd door he had found and investigated in exchange for an hour or so researching the item. The PC claimed he wished to Identify + document his research and thus needed the time.

Now Estavan is an extremely intelligent opponent, at the time of play since this was an unexpected turn of events, I had not come up with the ways the Ogre Mage would have safeguarded the asset should the PC had tried to flee with it, but I expected that Estavan would have been more than prepared. The PC would have surmised the same thing...
At the very minimum the PC knew there was an invisible creature or two in the room wherein they were negotiating.
Estavan having agreed to the PCs demands, left him in the lounge alone with the object.

The PC's plan was simple - to use the Weapon Bond class feature of the Eldritch Knight. Now the item was not a weapon but a rod (actually, part of a rod ), but I allowed the ability. It was cool and fun idea by the PC.


Sure enough, before leaving Sigil, the PC used the ability to summon the rod to his hand and then he left, plane-hopping for a number of days...

After the session, I did some research about magically "marking items"
I came across Arcane Mark and Instant Summons.

• It would make sense that Estavan would have made the necessary preparations for the safeguarding of the rod. He was also the type of person to scry on the PC and use the Instant Summons once he made his way home.
• Now, long story, but we had gone back to flesh out this Sigil + plane-hopping storyline, so the actual campaign timeline is some weeks beyond this point. In that future timeline, the inference of a rod was not present. That is not to say that I would be "taking" this rod because of that, but should I exercise it, it would make sense with the storyline as it progressed.
• The future plan was and is for the entire party to eventually get to Sigil, so this Estavan and the rod would likely be part of a future campaign.

How would you have handled this?
I mean this Estavan meeting and the inference with the rod came out of nowhere. So I had no time to prepare nitty-gritty stuff, besides the main crux of the session was to see how long (and if) the PC made it home back from Sigil to the rest of the party.

[There was a fail-safe which was player-facing, should the character have died, since we were dealing with a past timeline, but that is beyond the scope of the issue we are addressing]

I'm not sure what you mean by "inference of a rod" - do you mean that, in the campaign timeline, which is ahead of the events you describe in your post, the PC doesn't have the rod? Or at least hasn't been revealed to have the rod? So that having the NPC instant summons it away wouldn't cause contradictions with established events of the fiction?

In any event, even with that uncertainty on my part I can say that what you are describing seems exactly like the sort of thing I have in mind. Setting aside the timeline aspect, this sort of thing would come up all the time in my RM play: the players would have their PCs go somewhere, and do or take something, and it would make sense that a NPC would have wards or protections or the like, but I have no prep telling me what those might be, and so stuff has to be made up on the spot, and sometimes retrospectively.

I would try and be fair and reasonable - preserving verisimilitude (eg powerful mages will have magical wards) while not just arbitrarily hosing the players (and their PCs) - but I found it hard. And sometimes pretty arbitrary. And sometimes it would frustrate the players, although they could generally see why I was doing what I was doing (ie that it made sense in the fiction). And at the time we didn't really have any more robust techniques available to us.

The systems I play now tend not to raise these issues in the same way, because of the ways that they call for rolls to be made and the ways failures on those rolls are narrated. I can't think of an especially high magic example off the top of my head; but in my Torchbearer game, one of the PCs had an Elfstone which was very valuable to her (she was cursed by the magic gem, which caused her to be obsessed with it). When a roll was failed while trying to purchase an item in a village, I narrated the failure thus: when the PC got back home (to her mother's house in the village), she noticed that the Elfstone was missing! Someone had stolen it from her.

Instead of having to reason about the skills of the thief, and their opportunities to pick the PC's pocket, and so on - I followed the pacing dictates of the game, and narrated the unhappy thing when the game opened up that possibility for me.

 

[pemerton]

Play in narrativist RPGs is fundamentally different than play in non-narrativist RPGs.

As someone who has plenty of experience GMing and playing across a variety of RPGs, I think this claim is simply false.

If you were a player in my Prince Valiant game, you would not find what you are doing "fundamentally different" from playing in (say) @Crimson Longinus's D&D game. In both games, you would declare actions, which would be resolved, with the GM telling you what happens next.

You might notice some differences, though, in the ways that the GM decides what to tell you.

I think I might summarize the difference here as player agency over their character vs player agency over the rest of the fiction.

This putative contrast breaks down as soon as you, as a player, declare of your character who is drinking in a tavern "I throw my mug to the ground!". Now you've made it true that there is a mug on the floor, perhaps broken; that there is drink spilled on the floor; perhaps that some other people - NPCs - are surprised or shocked; etc.

The PC is not in a hermetic bubble. The point of declaring actions is to change things in the fiction.

This is why, upthread, @hawkeyefan stated that "narrative control" mechanics (which would presumably include things like the example he posted upthread, of the PC in Spire who can declare their familiarity with a local pub) can be set aside for the purposes of the discussion.

 

[bloodtide]

The narrativists have essentially created a new form of agency for their style of play, which is fine. What's not fine is to then say that our style of play denies or reduces agency, because it doesn't. Their agency doesn't exist in our playstyle, so there's nothing to deny. Neither one is better/greater than the other. They're just different.

Hear, hear. No good player complains about the 'railroad' in my game....assuming they know about it, and often they don't. After all a great many players are just Players and don't know or care about all the Gamer Stuff.

No, there's no real mystery. There is a pretend mystery in both games. The way that this pretend mystery is "solved" is different in each game.

You have to let go of this idea that because something was made up prior to play it's somehow more real make believe than make believe that's only revealed through play.

They're both make believe.

This is the big difference between the games. Your game has a make believe pretend mystery.

My game has a real mystery. That is me, as the DM, creates a real mystery. Then the players, for real, must learn the details of everything. Once the players get all the details, they must for real put it all together to try and solve the mystery.

It is MUCH more real to have a player or group have to solve a real mystery for real using their own abilities and skills. It is much more real then the "you rolled an 11 and solved the mystery".

Agency in a game is a product of inviolable rules which players know and can rely on to achieve known goals.

But the most prevalent style of roleplaying features GMs claiming they can obfuscate the rules (or design them in the moment hidden from the players), undermine the efficacy of any known rules (in the name of their assumed understanding of the games' greater good), all while concealing the goals of play (in the name of 'mystery') and call it agency. The only agency the players have in such games is the choice to participate at all. It's why such a playstyle is called 'participationism'.

This just is such a sad way of looking at RPGs.

 

[soviet]

I have 100% agency in the real world

Everything in your life is exactly as you wish it to be? You are unconstrained by social pressures, the law, financial considerations? You could choose to end global hunger or global conflict but you don't want to?

 

[pemerton]

I'm not getting into the definitional argument here again, but I do want to point out, many of us wouldn't consider this a narrative or a change of narrative. I'm not saying, you shouldn't describe your campaigns this way, but I wouldn't agree with this premise that if a character rejects the duke's concern, and they go west, that means is a shift in narrative. I'd be much more likely to call it a shift in focus or direction of the campaign. It may seem pedantic, perhaps it is, but I also think when stylistic concerns are being raised, it matters

To me it seems utterly pedantic, and pointless. Particularly as I think the only reason @hawkeyefan was talking about "narrative" was because you introduced that term into the conversation.

I think railroad complaints is about players not being able to say things like "Okay we tell the duke to screw and go west instead".

I don't understand: why could a player not say that? It's not against the law. They're (presumably) not gagged by the GM.

Do you mean, the player can say it but the GM will ignore it?

You lose meaningful choices when the players can alter things in the setting/narrative outside their character.

Really? Always?

Suppose that the PCs are journeying, and the game is played in the style of D&D or Rolemaster c 1988. So the GM rolls a random encounter - the PCs are assaulted by bandits. A fight ensues, and the PCs are victorious. Now one of the players, reflecting on the fact that the PCs are journeying to assault a black dragon in its lair, thinks that it might be handy to have a spare pair of boots, in the event that there are problems with swamps or acid or both. And so the player declares "I (as my PC) search the defeated bandits - does any of them have a pair of boots that will fit me?"

In my experience, the GM will not know the answer to this question in advance. So a decision has to be made by the GM. One option is for the GM to say yes: "You find one pair of boots that is a good fit." This is an example of the player "altering things" in the setting/narrative outside their character - the player's question prompts the GM to make a decision about the bandit's footwear that had not previously been made.

What meaningful choice did this remove? I mean, it obviates the need for the player to declare that their PC goes to a town and visits a boot-maker. But that's true of anything at all - eg the PCs finding some treasure on the bandits obviates the need for the players to declare that their PCs get work as labourers in order to earn some money. The GM having a NPC approach the PCs and offer them payment to perform a task - a pretty common "adventure hook" technique - obviates any need for the players to declare that their PCs go up to someone and ask them for a job.

So I am not seeing the removal of the meaningful choice in the bandit case.

I think interrupting the PC is a place where they could keep pestering the party with stuff so the spell was impossible to cast, but I also think this would be super obvious. There is a fine line here too because one of the reasons stuff like this matters is circumstances are sometimes meant to be a balancing factor with spell casting. So there may be times when it would be appropriate for alarm to be hard to cast, but that shouldn't be because the GM wants to stop a particular encounter or wants a particular encounter to happen a specific way.

<snip>

Maybe the Assassin gets around the alarm and surprises the party, but more likely the assassin sets it off and he is the one who is surprised and starts the fight out on uneven footing.

<snip>

I think if the assassin has a ranged attack and that is how they would launch their assault anyways (and it can be made from outside the range of the spell) fair enough.

<snip>

if the GM is manipulating NPC tactics using his own knowledge of the spell and not the NPCs to get around the spell, that is an issue.

I guess my question is that you present "maybe" and "more likely" and "that is how they would" and "the NPC's knowledge" as if these are external things, that operate as constraints. But they are not. An answer has to be reached, somehow, about which possibility is the one that actually becomes part of the fiction.

I do things like track NPC movement. If there were a question about when they arrive, I might leave it to some kind of roll. But a GM having having the NPC show up five minutes after in order to stage the encounter the way he wants, is bad GMing IMO. I could maybe see an argument for the assassin knowing the spell has been cast somehow and waiting, but like I said in my post about the precision here, I think there are a lot of Ifs in that scenario so if the GM wants to play the Assassin smart or give him a sporting chance, he should really set some ground rules and procedures and follow them.

And as I've posted repeatedly upthread, the upshot of those ground rules and procedures and "some kind of roll" is a procedure broadly analogous to Torchbearer's.

Which could then be built into the spell rules.

I mean, it's not foreign for 5e D&D spells to include mechanical information. The old version of Pass Without Trace says, "A veil of shadows and silence radiates from you, masking you and your companions from detection", while the new version says "You radiate a concealing aura". But both than go on to say that this grants the masked/concealed characters a +10 bonus to DEX (Stealth) checks.

So Alarm could similarly refer to warding a place, no larger than a 20' cube, raising the DEX (Stealth) DC to sneak into the place by 10. If the NPC assassin has knowledge of magic then that can be applied via the standard skill augment rules (whatever those happen to be for the GM in question - 5e D&D is a bit ad hoc in that respect, I think). The benefits of using a ranged attack then simply get bundled into the benefits that confers on a DEX (Stealth) test, compared to attacking from melee distance.

The changes/additions I've just described wouldn't deal with the issues around timing (for casting, and for warding duration), but would deal with the issues around precise tracking of distances down to the last inch.

The spell is kind of pointless if they can't at least roll initiative and not be surprised by the assassin.

So why not express the spell's effect by reference to the surprise rules, initiative rules, etc? As I've just posted, it's not as if D&D 5e resolutely refuses to incorporate mechanics into its spell effect rules.

 

[Bedrockgames]

I don't understand: why could a player not say that? It's not against the law. They're (presumably) not gagged by the GM.

Because in a campaign where the GM is railroading and agency is being thwarted, maybe you can say it but the GM is likely to do things like thwart you by throwing up obstacles, make the adventure he had planned all along come to you, or just have it be completely pointless and boring to go in a direction that isn't pointed in the direction of the duke