[MENTION=59082]Mercurius[/MENTION], there are some point where I think you have mis-spoken, or seem not to understand some RPG techniques.
The PC has agency in the fictional world, but it's fictional ie imaginary agency and so, as I explained to [MENTION=29398]Lanefan[/MENTION] not too far upthread is orthogonal to issues of railroading etc. (A PC might be enlaved by some other being, yet the player have unfettered autonomy, because the player determines the details of what the enslaving being asks of the PC.)
And the player may or may not have agency in the real world, in the playing of the game, depending on his/her capacity to change the state of the shared fiction. But the process of doing that does not resemble throwing rocks or climbing walls in the real world - it resembles playing cops and robbers, or writing a story with some friends, or improv acting, though is not identical to any of those.
As I said, if a kid complains that the fellow players of cops and robbers aren't being fair, no one thinks we address their agency by comparing it to the agency of a real bank robber shooting a real tommy gun. We ask about who is getting to decide whether a shot hits and whether a hit kills, and how that decision-making power is being exercised. These are the same sorts of questions one can ask about a RPG.
The player generally states an intended action - "I pick up a rock" - and then the table (perhaps just the GM, or maybe the GM as first among equals) accepts that this has happened, or alternatively some sort of mechanical process is called for (most often rolling a dice, but not always - eg if the action is "I buy a shovel", then in BW a dice roll may be required (a Resources check), but in D&D the mechancial process is to reduce the number in the treasure box on the PC sheet by the amount indicated by the GM).
But all action declaration in a RPG depends upon the fictional positioning of the PC - you can't say "I pick up a rock" and hope to have that action declaration succeed if your PC is not in the vicinity of some rocks; you can't say "I buy a shovel" and hope to have that action declaration succeed if your PC is not in the vicinity of some friendly purveyors of shovels; etc.
In a game run on the basis of GM pre-authorship with secret elements of the gameworld, it is possible for a player to think that the fictional positioning is appropriate for declaring some action, but in fact it is not. So the player declares the action, and perhaps the GM even allows the dice to be rolled (or rolls them behind the screen), but the check will always fail (and thus the fiction not develop in the way that the player wants) because in fact the fictional positioning was not apposite.
Now, there are borderline cases here, because "secrecy" is a matter of degree. To give a simple example, a combat encounter that starts with an invisible opponent among the visible ones will produce a moment in which the fictional positioniong turns out to be different from what a player understood it to be - eg (in some editions of D&D) s/he will declare some movement and then suffer an opportunity attack that s/he wasn't expecting. The 4e DMG's skill challenge example has a comparable non-combat example: anyone who tries to bully the duke automatically incurs a failure in the attempt to persuade him.
My own view is that if the secret is (i) within the ascertainable scope of the situation as presented to the players, and (ii) is salient within the context of gameplay, and (iii) is not overwhelming in its impact on the situation, then it's fair game. The two examples I've given satisfy (i) - you can find the invisible foe through various means including Perception checks; you can learn the duke's personality through an Insight check.
My reason for caring about (ii) - salience - is because, in practical gameplay terms, this is a major consideration for knowability. The players need to have at least some general sense of what they are expected to be looking for. I think the combat example satisfies (ii) for a default D&D game - we all know that invisible foes go with the territory. The skill challenge example is, in my view, more contentious in respect of (ii) and would depend upon how the campaign, as actually played, has presented personalities and has treated bullying as a method of persuading them.
My reason for caring about (iiI) is that, if the secret is overwhelming and the players don't learn it, then the game has a feel of "rocks fall, everybody dies". I think that is fair game in classic dungeoneering - ToH is full of it - but I personally don't care for classic dungeoneering as a playstyle, and hence include (iii) as a desideratum. Both the examples I gave satisfy (iii) - a skill challenge isn't lost with a single failure; and one invisilbe foe (who is otherwise part of a fairly desinged encounter) isn't going to lead, in iteslf, to a TPK.
As well as secret backstory, there can simply be unestablilshed backstory. Is there a rock nearby? Is the mace inthe tower? Will the captain of the starship respond sympathetically to a mayday call, even if that means going against strict security procedures? (That last one is from the Traveller thread I referenced a bit earlier in this thread.)
For this, I favour three possible approaches.
(1) If it's obviously inappropriate given genre and established backstory (eg Luke Crane's example of beam weapons in the Duke's toilet) then the answer is no. I would also extend this to low/no stakes distractions - "we search the bodies for loot". In my 4e game I'll often just say "They've got nothing valuable", rather than waste time on what is essentially a distraction - loot is on a "timer" (treasure parcels), and if I've got interesting ideas for how those parcels can come into play, I'm not intersted in spending time on paragon-tier PCs looting hobgoblins. (Contrast our BW game, where Resources is a vital stat constantly under pressure, and the search for loot might generate a high-stakes Scavenging check.)
(2) If the player is looking for some enrichment of the framing to help support some other action declaration (eg S/he wants his/her PC to climb the compound wall - are there any trees nearby? S/he wants his her PC to throw a pebble at the bedroom window to wake the sleeper inside - is there a pebble lying on the ground nearby?) then I would say yes.
(3) If the question itself is high stakes - I need to catch the spilled blood of the mage for my master - is there a ewer in the room?; I've returned to my abandoned tower - is my mace still here?; I'm reading the scroll setting out the cult's theology - are there any hidden marks or writing? - then it's time to roll the dice.
The only sense in which this gives the player the power to determine whether or not a rock exists is that (i) the player has the power to declare an action ("I look for a rock, so I can pick it up and throw it"), and (ii) the GM uses one of the above techniques to determine whether or not a rock is there to be picked up.
In other words, "Say 'yes' or roll the dice" is a GM-side technique, not a player side technique. (Contrast Fate Points in OGL Conan, which are a player side resource that permits the player to simply declare that a rock exists. I've never run or played in a game that uses such a mechanic myself; Cortex+ Heroic is the closest I've come, and it's not the same as that.)
It is generally accepted that, if a PC races and NPC, the outcome should be determined by the mechanics (maybe opposed Speed checks; mabye the one with the highest Quickness score wins; depending on what the particular game dictates). It's generally accepted that the GM can't (reasonably) just fiat that his/her NPC wins the race.
So consider a PC who hopes to find the map in the study; or the mace in the tower; or the pebble on the road. That can also be settled via action resolution (as I've described above in this post) - how is the GM deciding, by fiat, that there's no map or mace or pebble there, any different from fiating that the NPC wins the race?
As a GM, I've got not doubt that I play the "gameworld" as something with which the players (via their PCs) interact. I just don't assert fiat powers to make my guy always win!
Consider the example of the mace.
If the GM is omnipotent, then when the player says "I search the tower for the mace" then either;
(1) I say no (because that's what my notes say; that's what I feel like; I don't want my player's character to have a mace; whatever). That's personally not how I want to run a game - I regard it as railroading.
(2) I say yes. Where's the fun in that? Now the game is essentially about GM largesse, and every time something is up for grabs in the ficiton I'm the one who has to decide it. That's of zero interest to me also.
(3) I roll some dice. That's what I had to do in my Traveller game the other day, when the PCs were trying to find an alien artefact on sale at a market on a world whose inhabitants have mixed human/alien ancestry. I think it's better than (1) or (2), but is a little bit like playing the game with myself rather than my friends. (Unfortunately, Traveller doesn't have an obviously better mechanic - though in our game I have the players roll the reaction rolls for NPCs, so they're more like influence checks, and in retrospect I probably should have improvised something similar on this occasion.)
(4) The player rolls some dice, therefore (i) having a basic investment in the game play, and (ii) being in a position to bring player-side resources to bear. I like this best.
In RPGing, though, it's more prosaic than that. If the GM is ominpotent than decision making is either (1) or (2) immediately above, and I've explained why I don't care for either of them.
I've described upthread how the mace example was resolved. The player had written into his PC's backstory that he had been working on the mace before leaving his tower. The PC arrived at the tower. It had not been established in play whether or not the mace was there. The player declared an action (of searching for it), which was resolved. It failed. Hence the PC didn't find the mace; rather, he found something that revealed to him a horrible truth about his brother.
As it turned out, the mace had been taken by a NPC (the dark elf). But that was determined (by me, the GM) as a consequence of the failed check, not an input into it.
Your (3) - the GM fiats that it's not there - or your "sure, why not" are both approches that I don't like for the reasons I've stated. And I've also explained why I find it more immersive, both as player and GM, to have the player rather than the GM declare the action and roll the dice.
As a GM, I have no interest in being an "illusionist" who decides all outcomes by fiat, treating player action declarations merely as suggestions.
As a player, I want to play my characer as I envision him/her, not the GM's interpretation of my character and his/her situation.
In what you call the "traditional" model: the PCs aren't designed with built in hooks or drives or motivations (other than the most basic - we want loot and XP); the GM does not frame scenes that prompt choices driven by those motivations (eg look at [MENTION=29398]Lanefan[/MENTION]'s post, where in years of play the fact that his PC hopes to be a Senator has never been a principal focus of the game); and the GM does not narrate consequences in a way the leads naturally and inevitably into further scenes of that kind. Many posters on these boards, including [MENTION=29398]Lanefan[/MENTION] who is posting in this thread, not only design adventures independently of knowing what PCs the players will be playing, but advocate that as the best way to run a game: that may be standard for the "traditional" model, but is completely antithetical to what Eero Tuovinen is talking about.
Likewise, many posters on these boards (none yet in this thread, I think) assert that it is "contried" or "artificial" for the GM to deliberately frame situations so that they pull on the PC backstories; indeed, many GMs (again, [MENTION=29398]Lanefan[/MENTION] is a professed example) regard player-author backstory as a completley optinal consideration for the GM.
And "traditional" modules are full of fetch quests, "rescue the princess"-type episodes which are simply a series of dungeon challenges with a McGuffin at the end; etc. All this is the complete opposite of the Standard Narrativistic Model.
Look at an AP - with the villains pre-established, all the major scenes sketched out in advance, the climax already written - and then actually imagine playing a game in the way Tuovinen describes it: the players deciding what will be important to their PCs, the GM using those hooks to frame situations and guide the adjudication of consequences, and then ask yourself how could an AP be used in that game? It couldn't. You may be able to take ideas or little vignettes from it (that's how I use modules) but you couldn't "run the AP", because that woudl require knowing all the consequences in advance, and hence all the motivations and choices and the outcomes of those choices in advance; which can't be done.
The player has no agency in the fictional world, any more than you have the power to punch Sherlock Holmes in the nose.
The PC has agency in the fictional world, but it's fictional ie imaginary agency and so, as I explained to [MENTION=29398]Lanefan[/MENTION] not too far upthread is orthogonal to issues of railroading etc. (A PC might be enlaved by some other being, yet the player have unfettered autonomy, because the player determines the details of what the enslaving being asks of the PC.)
And the player may or may not have agency in the real world, in the playing of the game, depending on his/her capacity to change the state of the shared fiction. But the process of doing that does not resemble throwing rocks or climbing walls in the real world - it resembles playing cops and robbers, or writing a story with some friends, or improv acting, though is not identical to any of those.
As I said, if a kid complains that the fellow players of cops and robbers aren't being fair, no one thinks we address their agency by comparing it to the agency of a real bank robber shooting a real tommy gun. We ask about who is getting to decide whether a shot hits and whether a hit kills, and how that decision-making power is being exercised. These are the same sorts of questions one can ask about a RPG.
I don't know of any tabletop RPG where a player makes it true, in the fiction, that his/her PC has picked up a rock by actually doing it, ie actually picking up a rock.
The player generally states an intended action - "I pick up a rock" - and then the table (perhaps just the GM, or maybe the GM as first among equals) accepts that this has happened, or alternatively some sort of mechanical process is called for (most often rolling a dice, but not always - eg if the action is "I buy a shovel", then in BW a dice roll may be required (a Resources check), but in D&D the mechancial process is to reduce the number in the treasure box on the PC sheet by the amount indicated by the GM).
But all action declaration in a RPG depends upon the fictional positioning of the PC - you can't say "I pick up a rock" and hope to have that action declaration succeed if your PC is not in the vicinity of some rocks; you can't say "I buy a shovel" and hope to have that action declaration succeed if your PC is not in the vicinity of some friendly purveyors of shovels; etc.
In a game run on the basis of GM pre-authorship with secret elements of the gameworld, it is possible for a player to think that the fictional positioning is appropriate for declaring some action, but in fact it is not. So the player declares the action, and perhaps the GM even allows the dice to be rolled (or rolls them behind the screen), but the check will always fail (and thus the fiction not develop in the way that the player wants) because in fact the fictional positioning was not apposite.
Now, there are borderline cases here, because "secrecy" is a matter of degree. To give a simple example, a combat encounter that starts with an invisible opponent among the visible ones will produce a moment in which the fictional positioniong turns out to be different from what a player understood it to be - eg (in some editions of D&D) s/he will declare some movement and then suffer an opportunity attack that s/he wasn't expecting. The 4e DMG's skill challenge example has a comparable non-combat example: anyone who tries to bully the duke automatically incurs a failure in the attempt to persuade him.
My own view is that if the secret is (i) within the ascertainable scope of the situation as presented to the players, and (ii) is salient within the context of gameplay, and (iii) is not overwhelming in its impact on the situation, then it's fair game. The two examples I've given satisfy (i) - you can find the invisible foe through various means including Perception checks; you can learn the duke's personality through an Insight check.
My reason for caring about (ii) - salience - is because, in practical gameplay terms, this is a major consideration for knowability. The players need to have at least some general sense of what they are expected to be looking for. I think the combat example satisfies (ii) for a default D&D game - we all know that invisible foes go with the territory. The skill challenge example is, in my view, more contentious in respect of (ii) and would depend upon how the campaign, as actually played, has presented personalities and has treated bullying as a method of persuading them.
My reason for caring about (iiI) is that, if the secret is overwhelming and the players don't learn it, then the game has a feel of "rocks fall, everybody dies". I think that is fair game in classic dungeoneering - ToH is full of it - but I personally don't care for classic dungeoneering as a playstyle, and hence include (iii) as a desideratum. Both the examples I gave satisfy (iii) - a skill challenge isn't lost with a single failure; and one invisilbe foe (who is otherwise part of a fairly desinged encounter) isn't going to lead, in iteslf, to a TPK.
As well as secret backstory, there can simply be unestablilshed backstory. Is there a rock nearby? Is the mace inthe tower? Will the captain of the starship respond sympathetically to a mayday call, even if that means going against strict security procedures? (That last one is from the Traveller thread I referenced a bit earlier in this thread.)
For this, I favour three possible approaches.
(1) If it's obviously inappropriate given genre and established backstory (eg Luke Crane's example of beam weapons in the Duke's toilet) then the answer is no. I would also extend this to low/no stakes distractions - "we search the bodies for loot". In my 4e game I'll often just say "They've got nothing valuable", rather than waste time on what is essentially a distraction - loot is on a "timer" (treasure parcels), and if I've got interesting ideas for how those parcels can come into play, I'm not intersted in spending time on paragon-tier PCs looting hobgoblins. (Contrast our BW game, where Resources is a vital stat constantly under pressure, and the search for loot might generate a high-stakes Scavenging check.)
(2) If the player is looking for some enrichment of the framing to help support some other action declaration (eg S/he wants his/her PC to climb the compound wall - are there any trees nearby? S/he wants his her PC to throw a pebble at the bedroom window to wake the sleeper inside - is there a pebble lying on the ground nearby?) then I would say yes.
(3) If the question itself is high stakes - I need to catch the spilled blood of the mage for my master - is there a ewer in the room?; I've returned to my abandoned tower - is my mace still here?; I'm reading the scroll setting out the cult's theology - are there any hidden marks or writing? - then it's time to roll the dice.
The only sense in which this gives the player the power to determine whether or not a rock exists is that (i) the player has the power to declare an action ("I look for a rock, so I can pick it up and throw it"), and (ii) the GM uses one of the above techniques to determine whether or not a rock is there to be picked up.
In other words, "Say 'yes' or roll the dice" is a GM-side technique, not a player side technique. (Contrast Fate Points in OGL Conan, which are a player side resource that permits the player to simply declare that a rock exists. I've never run or played in a game that uses such a mechanic myself; Cortex+ Heroic is the closest I've come, and it's not the same as that.)
I discussed this upthread.
It is generally accepted that, if a PC races and NPC, the outcome should be determined by the mechanics (maybe opposed Speed checks; mabye the one with the highest Quickness score wins; depending on what the particular game dictates). It's generally accepted that the GM can't (reasonably) just fiat that his/her NPC wins the race.
So consider a PC who hopes to find the map in the study; or the mace in the tower; or the pebble on the road. That can also be settled via action resolution (as I've described above in this post) - how is the GM deciding, by fiat, that there's no map or mace or pebble there, any different from fiating that the NPC wins the race?
As a GM, I've got not doubt that I play the "gameworld" as something with which the players (via their PCs) interact. I just don't assert fiat powers to make my guy always win!
No. It's the omnipotence I personally don't enjoy (either as player or GM).
Consider the example of the mace.
If the GM is omnipotent, then when the player says "I search the tower for the mace" then either;
(1) I say no (because that's what my notes say; that's what I feel like; I don't want my player's character to have a mace; whatever). That's personally not how I want to run a game - I regard it as railroading.
(2) I say yes. Where's the fun in that? Now the game is essentially about GM largesse, and every time something is up for grabs in the ficiton I'm the one who has to decide it. That's of zero interest to me also.
(3) I roll some dice. That's what I had to do in my Traveller game the other day, when the PCs were trying to find an alien artefact on sale at a market on a world whose inhabitants have mixed human/alien ancestry. I think it's better than (1) or (2), but is a little bit like playing the game with myself rather than my friends. (Unfortunately, Traveller doesn't have an obviously better mechanic - though in our game I have the players roll the reaction rolls for NPCs, so they're more like influence checks, and in retrospect I probably should have improvised something similar on this occasion.)
(4) The player rolls some dice, therefore (i) having a basic investment in the game play, and (ii) being in a position to bring player-side resources to bear. I like this best.
Well, in politics this is up for grabs - historically the objection to monarchy hasn't just been that some monarchs are tyrants, but that every occasion of monarchy is a possible occasion of tyranny, which leaves those subject to it fundamentally unfree.
In RPGing, though, it's more prosaic than that. If the GM is ominpotent than decision making is either (1) or (2) immediately above, and I've explained why I don't care for either of them.
As I keep repeating, I have never played a RPG that resembles the "collaborative game" you describe. What RPGs do you have in mind in describing it?
I've described upthread how the mace example was resolved. The player had written into his PC's backstory that he had been working on the mace before leaving his tower. The PC arrived at the tower. It had not been established in play whether or not the mace was there. The player declared an action (of searching for it), which was resolved. It failed. Hence the PC didn't find the mace; rather, he found something that revealed to him a horrible truth about his brother.
As it turned out, the mace had been taken by a NPC (the dark elf). But that was determined (by me, the GM) as a consequence of the failed check, not an input into it.
Your (3) - the GM fiats that it's not there - or your "sure, why not" are both approches that I don't like for the reasons I've stated. And I've also explained why I find it more immersive, both as player and GM, to have the player rather than the GM declare the action and roll the dice.
If others like this sort of game, then go for it. I think I've explained why I don't like it.
As a GM, I have no interest in being an "illusionist" who decides all outcomes by fiat, treating player action declarations merely as suggestions.
As a player, I want to play my characer as I envision him/her, not the GM's interpretation of my character and his/her situation.
If you're serious about this, then I strongly encourage you to re-read what Eero Tuovinen has written (which I quoted).
In what you call the "traditional" model: the PCs aren't designed with built in hooks or drives or motivations (other than the most basic - we want loot and XP); the GM does not frame scenes that prompt choices driven by those motivations (eg look at [MENTION=29398]Lanefan[/MENTION]'s post, where in years of play the fact that his PC hopes to be a Senator has never been a principal focus of the game); and the GM does not narrate consequences in a way the leads naturally and inevitably into further scenes of that kind. Many posters on these boards, including [MENTION=29398]Lanefan[/MENTION] who is posting in this thread, not only design adventures independently of knowing what PCs the players will be playing, but advocate that as the best way to run a game: that may be standard for the "traditional" model, but is completely antithetical to what Eero Tuovinen is talking about.
Likewise, many posters on these boards (none yet in this thread, I think) assert that it is "contried" or "artificial" for the GM to deliberately frame situations so that they pull on the PC backstories; indeed, many GMs (again, [MENTION=29398]Lanefan[/MENTION] is a professed example) regard player-author backstory as a completley optinal consideration for the GM.
And "traditional" modules are full of fetch quests, "rescue the princess"-type episodes which are simply a series of dungeon challenges with a McGuffin at the end; etc. All this is the complete opposite of the Standard Narrativistic Model.
Look at an AP - with the villains pre-established, all the major scenes sketched out in advance, the climax already written - and then actually imagine playing a game in the way Tuovinen describes it: the players deciding what will be important to their PCs, the GM using those hooks to frame situations and guide the adjudication of consequences, and then ask yourself how could an AP be used in that game? It couldn't. You may be able to take ideas or little vignettes from it (that's how I use modules) but you couldn't "run the AP", because that woudl require knowing all the consequences in advance, and hence all the motivations and choices and the outcomes of those choices in advance; which can't be done.
