GM fiat - an illustration

[pemerton]

It isn’t meaningless and this is, from what I see, how most gamers seem to mean it (or at least they mean it inna way that is closer to this). Agency is about being able to have your character do things without the Gm placing constraints to railroad. If the GM isn’t placing limits on what your characters can try to pursue, most folks see that as a campaign respecting player agency

I don't even know what this means.

Like, the GM starts a 1st level AD&D game, telling us that we (the PCs) are in Hommlett, and have heard rumours of the Moathouse. Suppose I decide that my PC wants to pursue the goal of becoming the ruler of the Bone March, how am I going to pursue that? Is the GM allowed to tell me that my player, born in a village near Verbobonc, has never even heard of the Bone March?

Suppose I tell the GM that I head back to my village because what I really want to do is woo Rosie Cotton and become a prosperous farmer, how am I going to do that? How does that even fit into AD&D play?

I really can't work out what you are talking about. To me, it doesn't seem to conform to any sort of RPGing that I'm familiar with.

 

[Bedrockgames]

I don't even know what this means.

Like, the GM starts a 1st level AD&D game, telling us that we (the PCs) are in Hommlett, and have heard rumours of the Moathouse. Suppose I decide that my PC wants to pursue the goal of becoming the ruler of the Bone March, how am I going to pursue that? Is the GM allowed to tell me that my player, born in a village near Verbobonc, has never even heard of the Bone March?

Suppose I tell the GM that I head back to my village because what I really want to do is woo Rosie Cotton and become a prosperous farmer, how am I going to do that? How does that even fit into AD&D play?

I really can't work out what you are talking about. To me, it doesn't seem to conform to any sort of RPGing that I'm familiar with.

I really don’t think what I wrote was all that perplexing or hard to understand

 

[pemerton]

My default preference is for 'off screen' stuff to be known about by everyone at the table, simply so we're on the same page. Yet there are some back story parts that are commonly hidden that necessitate other back story parts to be hidden.

The most common hidden backstory part is the status of a person or thing.

Michelle has been murdered

The money isn't in the safe.


The following often needs to be hidden if the above is hidden.

NPC backstory: Michelle's backstory can't include the fact she's dead if her murder has to be hidden. Likewise with Michelle's murderer.

NPC effectiveness gets hidden a lot for the same reason.


Why hide it?

We want to see the interaction between the player and the GM's suddenly revealed backstory in real time. Otherwise we have a lot of pre-play for a lot of things.

'what happens when my character discovers Michelle has been murdered?' and so on and so on.

And assuming that we do want to see that interaction, what approach do we take to revealing that Michelle has been murdered?

Vincent Baker, in DitV, talks about "actively revealing" backstory in play.

But in a typical CoC-esque game, I don't think it is a goal of play to find out what happens when the PC realises that Michelle has been murdered, or more generally to see interactions resulting from sudden reveals. The goal of play is puzzle-solving (as you have posted in the past). Which means that "actively revealing" backstory would be a poor technique.

And the reason for hiding is not the one that you give in your post, but rather because the goal of play includes the players having to "manipulate" the fiction - that is, declare the right actions to prompt the GM to reveal the right information - in order to solve the puzzle.

What do you think the benefits of having a GM are then?

Because to me, having hidden information seems one the features. although given it's the feature i'm least keen on the other reason I've found is strong scene framing authority.

One reason for having a GM is authority to frame scenes, especially by establishing conflict. Another related reason is to allow for the introduction and handling of adverse consequences.

The hidden information aspect is important in some RPGing. But if I think about my Prince Valiant or Classic Traveller game, there is very little hidden information. And generally its purpose is to allow gradual framing of the situation and the implicit conflicts - to ramp things up over time. It's about manipulating stakes and tension.

For instance, consider this bit of play:

That ended the battle scene, and we cut to the castle. With the reserve garrison and the returned troops there were 22 archers, 36 men-at-arms, 18 sergeants and 3 knights, plus Sir Andreas and his castellan Sir Satyrion, and the PCs and their 12 men-at-arms: a total of 96 defenders. The penalty for being outnumbers two-to-one is -2 dice; the penalty for assaulting the castle is -4 dice; and so the players were reasonably confident they could hold the castle against assault.

And so everyone dined.

The scenario I was drawing on here is in the Prince Valiant Episode Book - The Littlest Prince - but our framing was rather different, as the castle's army had returned rather than having been routed. So I asked Sir Morgath's player to roll his Fellowship + Presence (7D) against an obstacle of 3 (I think it was), and he succeeded. I told him that he noticed that Sir Satyrion seemed rather sour. Morgath first spoke to Sir Andreas, and (with successful Courtesie) confirmed his suspicion that it was Satyrion who had suggested that they sally forth. (The conjecture was the players, and it certainly fitted with the scenario backstory.)

He then consulted with the other PCs, and decided to speak to Sir Satyrion, to try and learn his motives (eg power-hunger; loyalty to the Arab rather than the Greek cause; etc). This was Glamourie, and the player was rolling 8 dice (Presence 4, Glamourie 2, +2D for greater fame and his prestigious accoutrements). So he was quite successful. First, he learned that Satyrion was jealous of Andreas, and mocked him. When I asked Morgath's player if he likewise mocked Justin and Gerren, he replied "Not Sir Gerren". I asked Sir Justin's player to make a Presence check to see if he overheard the mockery; he did. Justin's player had Justin declare that he was finished dining, with the intention that he would go to the infirmary and tend to the injured; I called for another Presence check to see if he could really hold his pride in check despite Sir Morgath's word. This succeeded too, and he then rolled very well on his healing, further cementing his order's reputation in Cyprus.

A second Glamourie check succeeded, and Satyrion asked Morgath whether he was married. Sir Morgath started to speak of his love for his wife Elizabeth of York - I asked for a Presence check to avoid also speaking of Lorette of Lothian, who he left in Toulouse but still longs for. The check succeeded, and so Lorette didn't come up. Satyrion explained that he wished to marry Flora, Andrea's teenage daughter; although this would not make him the heir, as there was also an infant son, Theo. Satyrion also asked Morgath if his ambitions lay in the East, but Morgath responded that he hoped the crusading urges of Justin and Gerren would soon be satisfied so that they could all return to the West, and Morgath could return to his duties in York. (Morgath's player explained that Morgath isn't really a liar, and that his Glamourie is mostly just for helping resist seduction.)

I explained that this seemed to quieten Satyrion, and asked whether Morgath wanted to pump him for any more information. A third successful Glamourie check was made, and Morgath suggested that he might be able to help Satyrion with his plans with Flora. Satyrion, his guard quite lowered, asked whether Morgath was intimating that he might help with a plot against Andreas and Theo; Morgath replied that he was just referring to wooing! And so Satyrion excused himself.

I asked Sir Gerren's player what Gerren was doing. The reply was, checking the castle's defences. I said that a roll of Battle + Presence (9 dice) could strengthen the defences on 5 or more successes (ie give another die penalty to attacking forces), while two or fewer successes would mean something had gone wrong. Naturally the roll was two successes! As Sir Gerren was at a tower battlement, backlit by torches, an arrow struck him for 1 point of Brawn lost. I then said that he could see someone - a spy - who had infiltrated via the postern and was trying to open the main gate. I asked Gerren's player whether he was prepared to leap from the tower to stop the spy, as Tintin would. He was. I can't recall the difficulty I set - 4 or 5, I think - with every success short of that on Brawn + Agility being a point of Brawn lost in the landing. With Brawn 4 and no Agility skill, 2 points were lost, leaving Sir Gerren with only 1 Brawn to brawl with the spy. Their first round of brawling did not let the spy get to the gate, but nor did Sir Gerren disable him. Sir Gerren called for help, and with a success on his Presence + Oratory check guards came running and the gate remained closed. But about this same time, Flora announced in distress that Theo was missing from the nursery! Sir Morgath spoke to the servants in the castle, including the basement (Fellowship + Presence, with good successes) and they had seen nothing; Sir Gerren looked out from the battlement, and might have seen the riders leaving where Satyrion had spirited Theo out the postern, but failed his Presence check.

Down in the courtyard, Sir Gerren could see that Sir Andreas was getting ready to ride forth searching for his son. He tried to persuade him to hold off, that this was too big a risk to the defence of the castle; but failed: Sir Andreas reminded him that when his son (Sir Justin) had been in trouble, he had risked everything to rescue him (ie in the battle earlier that day). But Sir Andreas agreed that Sir Gerren, Sir Justin and their troops could ride with him. An oratory check mustered the men, although it was one success short of the difficulty I'd set and so they were at -1 for fighting due to the rapidity of the mustering (loose saddles, poorly donned armour, etc). And so Andreas rode out with 2 of his house-knights, 3 sergeants and 6 men-at-arms, as well as Sir Justin, Sir Gerren, their scout Rhan, and their 12 men-at-arms. Sir Morgath's player insisted quite forcefully that his scout retain, Algol the Bloodthirsty, was remaining with him in the castle. It was only once the posse had ridden out, to the echoes of me the GM saying "no backsies!", that the players fully computed that their two commanders with Battle 6 each had left the castle under the command of the teenager Flora and Sir Morgath with his Battle 1.

There is some hidden information here. Most of it pertains to Sir Satyrion and the details of his villainy: that he suggested that Sir Andreas sally forth; that he wishes to replace Sir Andreas by marrying Flora; that he intends to kidnap (and kill?) Theo. There is also the infiltration by the spy.

I would post most of this under the description of actively revealing the backstory in play. Prince Valiant uses checks - perception and social-type checks, as the example illustrates - to mediate this active revelation, but this is much more about the colour of the revelation, and whether the revelations occur with the players (via their PCs) having the momentum, or whether the momentum is against them.

Its of the essence of the RPG medium in its typical form - ie most of the participants take the role of particular characters in an unfolding situation, while one participant manages backstory and framing - that it wouldn't make sense just to lay out in advance everything in the fiction that matters to play. Where the play is about intrigue and derring-do among knights - as Prince Valiant is - it therefore doesn't make sense to lay out everything about Sir Satyrion. Hence the "active revelation" approach: if the Presence check to notice his sourness had failed, I - as GM - would have found another way to reveal his villainy, but in a different fictional context and with the momentum running in a different direction.

The approach I've just described relies heavily on the GM to manage (manipulate) pacing and stakes. This can be contrasted with (say) the role of the GM in classic D&D (Moldvay/Gygax) play. In the example of play from Prince Valiant, the fact that the players became aware of Sir Satyrion early doesn't mean that they "win". Sir Satyrion excuses himself; the player of Sir Morgath lets him go; action then shifts to another PC (Sir Gerren); and as GM I judge that the upshot of his actions (which include calling the guards to him) makes it permissible for me to decide that Sir Satyrion has kidnapped Theo. This is not penalising a misplay, but simply me, as GM, making decisions about what might follow from the current state of the fiction that would be fun and interesting and prompt the players to declare actions for their PCs; the only real "misplay" by the players is the subsequent one, of leaving weak commanders in charge of defending the castle. (And that misplay soon yielded its consequence, as the castle was assaulted, and fell, and Sir Morgath had to flee with Elizbeth his wife, Agol his retainer, and Flora.)

Now I don't think your play in which we want to see how the PC reacts when they learn Michelle was murdered is identical to my Prince Valiant play. But I feel it is closer to that, than to solving a traditional CoC mystery or to mapping a dungeon in classic D&D.

 

[pemerton]

We (our side) have very strict jobs regarding how we view the fiction and what we do with that and those jobs don't cross over to other players. There's no consensus required.

The Narrativists have a different job regarding the fiction but it 'does' have cross over. This can cause conflicting agencies (GM V Player say), so system mediation is required. That's why a dice roll is seen as more fair than fiat, for example.

To me, this doesn't seem right.

First, it seems to leave it mysterious why you use dice in your Sorcerer-esque play.

And second, presumably Apocalypse World falls on the narrativist rather than the "we" side of your distinction; but it has as strict a set of allocations of jobs regarding the fiction as any RPG that I know.

Yeah this is it. A ton of the criticism is about not having an equal/fair say in the outcome

On the premise that the outcome is what we are playing for.

In your example, where the outcome we are playing for is (inter alia) to find out how the player decides that their PC reacts to learning Michelle has been murdered, it wouldn't make sense to let the GM dictate that reaction.

Suppose that two things that we wanted to learn, from our play, were (i) whether Jackson's assassin is so good that they can beat an Alarm spell, and (ii) what that would mean for a confrontation between the assassin and the PCs. Then the 8 hour duration of the spell would contribute context/colour for that - and maybe the GM, following the sort of approach you've described out, reaches the view that yes, Jackson's assassin is that good and so narrates Jackson's assassin beating the spell. This is the expressive/artistic aspect you have posted about upthread.

In this case, though, the GM wouldn't then narrate to the player, "You die as Jackson's assassin plunges her knife into your chest". Because that would undercut the second goal.

But in any event I don't think a great deal of D&D play has the sorts of goals that you are talking about. It is loosely either "gamist" with a fair bit of exploration in the neighbourhood, or exploratory perhaps with intermittent bouts of gamism, especially during combat. The 8 hour duration for the Alarm spell is primarily meant to interact, in some fashion, with a rest economy; not provide colour for what it means to narrate a puissant assassin.

 

[Umbran]

I really don’t think what I wrote was all that perplexing or hard to understand

I suspect he's getting hung up on the... absoluteness as presented.

The iconic example of play that allows for player agency is a "sandbox campaign". But, when we invoke that metaphor... sometimes we forget that the sand is in a box. There are sides - boundaries - for play within the box. If you step out to go play on the swings, you are not playing in the sandbox.

Ultimately, playing RPGs with a group is a collaborative activity. Collaboration requires participants to accept some restrictions on their personal agency.

The real question becomes exactly where the lines of agency are drawn, and how fuzzy they are.

 

[pemerton]

It's certainly real deduction though. Like, Clue's gameplay is certainly weird, and I think you can get better deductive structures in other games, <snippage> but you are absolutely using systems to learn a specific unknown thing, and getting there faster/slower and being right are measurable things.

But straight away this tells us a difference from actually solving a mystery.

We could change the flavour of the game - say, make the cards be dossiers, embassies, and foreign agents - and now it becomes a game of working out how the military secrets were stolen (It was Sergei getting blackmail dirt on the military attaché in the US embassy in Istanbul.) But actually solving a mystery is not about collecting information within a finite problem-space to then draw a logical inference; and solving a murder doesn't necessarily involve the same sorts of inquiries as engaging in counter-espionage (not only are the forensics apt to be different; the interviews/interrogations are likely to be different also).

To me, here is the most obvious difference between the information space involved in actually solving a mystery, and the play of a mystery-solving RPG: in real life, the perpetrator needs to be identified, and could - in principle - be just about anyone. But in a RPG, the players rely on the GM to put salient NPCs in front of them. This can be "lampshaded" in the same way that Agatha Christie-esque mysteries do (the last time I ran a murder mystery, it was on a Traveller-esque starship that was in jump space, so no one was able to board or depart the vessel). But the CoC modules that I mentioned upthread don't use those particular techniques - they just rely on the players' understanding that the GM will present to them, in some or other fashion, the necessary NPCs.

This is not a criticism of RPG mystery solving. But I think it may be another reason why @hawkeyefan is saying that, even when they are based around pre-authored backstoyr, these are not real mysteries. They are confined puzzles - not logic puzzles like Clue(do); but something in the same sort of neighbourhood.

 

[pemerton]

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?

For this to carry weight, I take it that we have to say they are more interested in seeing what's going on to the west, yet pay to the fidelity of their PC and hence accede to the Duke's request.

And then your question becomes an empirical one. I don't know the answer. But when I'm GMing, I would regard it as a bit of a failure if the players felt compelled, by the logic of the fiction (including the fiction of their PCs), to do the thing that is less interesting to them. I would work on trying to make it more interesting.

 

[Micah Sweet]

What if we look at it another way.

A mystery is something unknown, right?

So isn’t the only way to do a real mystery for nobody to know the answer? So that it’s actually a mystery to all parties in the game?

Not in my view. The DM is running the entire world and everything in it except the PCs, so they have to represent the closest thing to objective truth in the setting that we're going to get. I know you have a different view though, do I doubt this will get us anywhere.

 

[Bedrockgames]

I suspect he's getting hung up on the... absoluteness as presented.

The iconic example of play that allows for player agency is a "sandbox campaign". But, when we invoke that metaphor... sometimes we forget that the sand is in a box. There are sides - boundaries - for play within the box. If you step out to go play on the swings, you are not playing in the sandbox.

Ultimately, playing RPGs with a group is a collaborative activity. Collaboration requires participants to accept some restrictions on their personal agency.

The real question becomes exactly where the lines of agency are drawn, and how fuzzy they are.

I didn't just mean it in terms of sandbox though. It could be in terms of an adventure. My point was more about the definition of agency and I just used the duke as a stark example of having the most agency in terms of what your character can do (because you are literally rejecting the adventure premise). But I am not saying that is automatically better: in some campaigns not engaging with the adventure could be a jerk move (to your point about RPGs being a collaborative activity where some restriction on agency may be required).

On sandbox play, I've never liked the metaphor precisely because it is bounded on four sides and sandbox play is more about having a world that isn't bounded in this way. Though I do think they still have limitations. But I think that is separate topic

 

[pemerton]

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?

I can't speak for @AbdulAlhazred, but What is the square root of 169? is not a mystery. It's a logic puzzle, that might be easier or harder to solve; but the answer is contained in the question. Likewise, say, a chess puzzle of the form "White to mate in 3 moves".

I think of mysteries in terms of empirical questions whose answer is not known, is not obvious, and requires the collection of actual empirical information to solve.

Game play provides examples of "puzzle solving" that are neither logical puzzles nor empirical mysteries. Charades and Pictionary and rebuses are all examples: the players are not collecting empirical information and making inferences from that; they are being given clues and then trying to engage in interpretive/analogical reasoning (what is this a picture of? what film has a word in its title that rhymes with smile; etc).

Empirical information can help in these games, but it is not empirical information provided within the game, but rather empirical information about the participants that the players bring into the game (eg what is pemerton's approach to drawing things - eg what caricatures does he default to? what films has pemerton seen, that he is likely to then portray in charades? etc).

Classic D&D dungeon exploration is closer to these sorts of parlour games than it is to either Clue(do) or actually solving a mystery. So are CoC-esque mysteries, I think: the reasoning doesn't actually involve collecting empirical details and drawing empirical inferences from them, but rather collecting bits of information that one, as a player, knows matter to the game, and then working out what picture or pattern of events these point to.