RPGing and imagination: a fundamental point

[Crimson Longinus]

@Manbearcat I don't think the GM fiat thing is binary. The GM decision making is in practice always restricted by something, and how much is a spectrum. There can be clear rules about it, principles, guidelines, suggestions or at least a social contract.

 

[FrogReaver]

1) You're working off of a definition of "Fiat" when it comes to TTRPGs that isn't helpful to understanding how play is differentiated by various forms of systemization. All GM decision-making isn't "Fiat." When it comes to GM decision-making, there are systemitized constraints, systemitized-directives, and systemitized-incentives and whether you can opt-out of any/all of them at GM discretion. A game like Dogs in the Vineyard (for instance) will tell you "do this", "don't do that", "when you do this other thing then this rewarding thing will occur", and "if you don't do this thing, then this punishing thing will occur." The combination of these things curate a GM's decision-space away from subset of choices and approaches and toward a different subset of choices and approaches. Further, these things are not opt-in/out at GM discretion.

Actual GM Fiat does not have these distinguishing characteristics. Instead of system, only the GM curates their decision-space. Further still, overwhelmingly, choices and approaches or subsets of them are almost entirely (or entirely) opt-in/out at GM discretion. Such a game might, contra to DitV, do none or few of those 4 things in the paragraph above and/or they might simultaneously tell the GM "feel free to opt-out or opt-in to any/all of these as play unfolds and as your instincts/discretion take you."

I'd propose that nothing counts as actual DM fiat by this definition.

There is no game that gives the GM carte blanche to at any point declare something like rocks fall you die (as an extreme example). This hopefully illustrates that the GM decision space is always curated. I think maybe there's some discussion to be had around the importance of system constraint vs social contract constraint vs Norm constraint - but in any event the GM is constrained. I'm not sure why GM fiat would just be system constrained vs any constrained in any way.

What purpose does obliterating the distinctions of these things and their impacts (on the cognitive space of the GM, on the cognitive space of the players who are playing a game governed by one vs the other, on the experience play at both the moment-to-moment level and at session-level) serve? I can't fathom any good reason why one would do this?

I'm not trying to obliterate any distinctions - i'm trying to be more accurate in my speech. That accuracy will still allow the conceptual distinction to be discussed, it just will draw more clear lines around independent concepts, and what combinations can exist. There's 2 overarching points here -
1. System isn't the only thing that restricts DMs, there's norms and social contract as well.
2. All systems do restrict DM's. -I'm going to post the 5e SRD on ability checks next.

GM framing of situation (establishing initial conditions) in a closed scene resolution game is not Fiat. The inputs and choices made in framing are systemically constrained and informed (both the mechanics and the components of play that inform the constituent parts of the framing) and the GM doesn't just get to opt-in/out of their system-directed job at their discretion.

This seems like it applies equally to what happens when narrating the outcome of a failed roll.
The inputs and choices made in framing are systemically constrained and informed (both the mechanics and components of play that inform the constituent parts of the framing" and the GM doesn't just get to opt-in/out of their system-directed job at their discretion.

I 100 % know that task resolution and conflict resolution are not only not the same things, but, phylogentic tree-wise, their common ancestor is sufficiently far apart that they're only superficially recognizable as even linked. Further, GM Fiat is not the same thing as systemically constrained and systemically-directed, GM decision-making. Right now, it feels like the work that we're putting in is coming from a position of obliterating the significant differences in these things and the methodology of attempting to prove the lack of differentiating characteristics is a winding conversation that is pulling random anecdotes from all over the place in a "look at this thing"..."ok, what about this other thing"..."ok, what about this" "ok look at this <thing I think is a smoking gun>."

I'm desperately trying to focus like a laser beam on core concepts of closed scene resolution and we're suddenly talking about Blades in the Dark (which doesn't feature it) and what appears to be either "goals at a global level" or "Payoff not directly indexing goals (which Payoff does, in fact, index goals insofar as "Payoff is the reward phase of Downtime where the receipts of prior evinced and recently actualized goals are materially rewarded via system-directed procedure."

I'm just going to say that I'm not doing another one of these.

I'll monitor this thread for focused, specific engagement on the core concepts of closed scene resolution. But that is the only thing I'm going to respond to. If I don't see that, then so be it. This isn't a conversation that we need to have (or needs to be had at all).

Seems to me the difference boils down to - we are trying to examine the fundamental concepts and instead of walking through that discussion i'm just being bombarded with 'this is the way it is, thus saith the authoritys!'

There's about 100 apriors to get to your conception. Closely examining all the relevant ones and their relationships i'm sure feels like going aimlessly to one thing from another but it's really fundamental. The concepts, counter concepts and counter-counter concepts being brought up form a web of meaning and disagreeing with even a few strands may make the entire web unstable. It's not to work to an 'aha, i tricked you moment', it's to work toward a shared understanding of an incredibly complex set of ideas.

 

[FrogReaver]

5e SRD rules on ability checks -

An ability check tests a character's or monster's innate talent and training in an effort to overcome a challenge. The GM calls for an ability check when a character or monster attempts an action (other than an attack) that has a chance of failure. When the outcome is uncertain, the dice determine the results.

For every ability check, the GM decides which of the six abilities is relevant to the task at hand and the difficulty of the task, represented by a Difficulty Class. The more difficult a task, the higher its DC. The Typical Difficulty Classes table shows the most common DCs.

Typical Difficulty Classes​

Task Difficulty	DC
Very easy	5
Easy	10
Medium	15
Hard	20
Very hard	25
Nearly impossible	30

To make an ability check, roll a d20 and add the relevant ability modifier. As with other d20 rolls, apply bonuses and penalties, and compare the total to the DC. If the total equals or exceeds the DC, the ability check is a success--the creature overcomes the challenge at hand. Otherwise, it's a failure, which means the character or monster makes no progress toward the objective or makes progress combined with a setback determined by the GM.


We are immediately told what an ability check is for - to test the character's innate talent and training in an effort to overcome a challenge. We are also given our first system constraint - 'GM calls for ability check when the outcome of an action is uncertain'.

We are then given another system constraint - GM decides which of the 6 ability scores is relevant. It doesn't tell us how to determine this, so the first thought might be - GM FIAT, but the obvious implicit answer is that he's to do so by the fictional positioning, coupled with the meaning of each ability score.

Next we come to to the DC. While the SRD doesn't give explicit guidance on what 'task' (it's word) should be easy, medium, hard, etc. One might say GM FIAT to set the DC here as well, but the implicit expectation of the rule is that these should be based on fictional positioning. Different DM's might end up setting a specific DC for a specific task differently, but after doing so other similar tasks that are more or less difficult are then expected to have higher/lowers DC's compared to the first task. So while no explicit guidance is given, there are system constraint around setting DC's.

We then are then given the actual ability check mechanic - roll a d20, add the relevant ability modifier, apply bonuses and penalties, and compare to the DC. Another constraint on the DM!

We are then given a constraint around what to do when success - the creature overcomes the challenge at hand!

We are then given a constraint around failure - either no progress toward the objective or progress combined with a setback determined by the DM. (I'd actually argue they left out a 3rd and more common option, failure to progress toward the objective with a setback - say time cost being a common one). But more importantly in the last bit it says the GM determines the setback! Yes! And no restriction is given on what the setback may be! Yes! The system technically allows here any setback, but in D&D it's the social contract and norms that further restrict the DM here. There's no actual option for rocks fall, you die - despite the system technically allowing that, because the social contract and Norms disallow that. Since ultimately the GM cannot say, rocks fall, you die, - it's not clear to me how this could be GM FIAT.

So, going just by the rules for ability checks, where is this system unconstrainted GM fiat in ability checks?

 

[clearstream]

1) You're working off of a definition of "Fiat" when it comes to TTRPGs that isn't helpful to understanding how play is differentiated by various forms of systemization. All GM decision-making isn't "Fiat." When it comes to GM decision-making, there are systemitized constraints, systemitized-directives, and systemitized-incentives and whether you can opt-out of any/all of them at GM discretion. A game like Dogs in the Vineyard (for instance) will tell you "do this", "don't do that", "when you do this other thing then this rewarding thing will occur", and "if you don't do this thing, then this punishing thing will occur." The combination of these things curate a GM's decision-space away from subset of choices and approaches and toward a different subset of choices and approaches. Further, these things are not opt-in/out at GM discretion.

I'd like to get a bit clearer about what I'm addressing. A claim was put forward that "all closed scene resolution is conflict resolution." That's a strong claim and my first question is simply: is it right? I have been told that the distinction between task-resolution and conflict-resolution comes down to being

precisely in the sort of relationship that obtains between succeeding on the check, the GM's authority over the fiction, and what happens next.

Accepting that for the sake of argument, means that in order to decide if "all closed scene resolution is conflict-resolution" I should test whether any scene resolution is not conflict resolution, e.g. task-resolution according to that definition.

Actual GM Fiat does not have these distinguishing characteristics. Instead of system, only the GM curates their decision-space. Further still, overwhelmingly, choices and approaches or subsets of them are almost entirely (or entirely) opt-in/out at GM discretion. Such a game might, contra to DitV, do none or few of those 4 things in the paragraph above and/or they might simultaneously tell the GM "feel free to opt-out or opt-in to any/all of these as play unfolds and as your instincts/discretion take you."

I used "GM fiat" to mean "GM's authority over the fiction, and what happens next." You've now postulated the additional stricture that GM-fiat is - if I understand correctly - not impinged by rules.

I would take a philosophically skeptical position toward knowing anything about what a GM who is both deaf to player intentions and blind to rules. We've denied ourselves knowledge on that score. To make that a little clearer, what we have said is that no game with rules presents a case of GM fiat, because GM fiat is the case where rules don't apply.

I suspect what you describe by "GM fiat" is Baker's case where GM inconsistently follows rules: is that right? If so, unless we can describe the nature of their inconsistency - is it random? is it perverse? - I would again be philosophically skeptical toward knowing how a game they run might go.

What purpose does obliterating the distinctions of these things and their impacts (on the cognitive space of the GM, on the cognitive space of the players who are playing a game governed by one vs the other, on the experience play at both the moment-to-moment level and at session-level) serve? I can't fathom any good reason why one would do this?

GM framing of situation (establishing initial conditions) in a closed scene resolution game is not Fiat. The inputs and choices made in framing are systemically constrained and informed (both the mechanics and the components of play that inform the constituent parts of the framing) and the GM doesn't just get to opt-in/out of their system-directed job at their discretion.

I was pondering this tonight, and I can accept that it is disturbing to have distinctions undermined. Again, my sole purpose was to answer the basic question - is it right to say that all closed scene resolution is conflict resolution? I picture that you come to this discussion with that settled in your mind. You can hopefully see that it is not only not settled in other posters' minds, but also to them feels like an essential piece of foundation to lay. One of the first steps in any investigation is establishing shared predicates.

I considered taking a - for the sake of argument let's say that - approach, but I felt that would lead to problems down the line, when notions came up that rested on predicates that I might not accept. It seems better to get firm on those first.

2) I'm not going to watch a Blades in the Dark video. I probably watched it in 2017 or whatever and by this point I've probably GMed more Blades in the Dark than almost anybody in the world not named Jon Harper (and I might even be able to give him a run for his money in terms of total hours GMed in the game).

Blades in the Dark isn't a closed scene resolution game. Blades in the Dark absolutely features various conflict resolution tech (Clocks in particular), but its not a closed scene resolution game like Dogs or several other games. It is a snowballing resolution game which features an abundance of conflict resolution and other tech.

In hindsight this was a poor choice of example, sorry. I was not saying that BitD presented here an example of closed scene resolution, rather I was thinking about clear examples of GMs ending a scene and recalled that one. It stuck in my mind because Harper says something like - "I'm going to end the scene here." I don't know if you saw my post a few threads back, but I committed to providing concrete examples where I could. I felt I had one here, but I see it lead to concerns about looking for a smoking gun or whatever, which couldn't have been further from my mind.

1) Considering "goals" either (a) too much in isolation or (b) too globally or (c) entirely out of context of the particular game in question. Again, we're now careening wildly away from the very specific conversation of goals/stakes in closed scene resolution (which is a form of conflict resolution...and after we have that fully canvassed, we can discuss other forms of conflict resolution and nail down "what these various forms of resolution share and what they do not share"). But, I'll humor this:

"Goals" at the Score level (which, again isn't closed scene resolution) are 100 % player-derived. They want to do this Score vs that Score and their thinking is invested with all of (i) individual PC protagonism (personal motivations and goals) and (ii) Crew protagonism (collective motivations and goals):

* Take out my Rival.

* Protect our Friend/Contact.

* Gain allies or help our current allies.

* Hit our enemy where it hurts/while they're weak.

* Hit our enemy in such a way that achieves misdirection and pits two of our enemies against each other.

* Gain this Claim or open up this other Claim on the Map for later.

* Prevent this Faction/Setting Clock from going off.

* Earn this beefy Payoff.

* Remove this amount of Heat/Wanted Level.

On and on and on.

These Goals inform both the general shape of the menu of prospective Scores that players mull and then inform the specific one they choose from their derived subset.


2) Confusing what Payoff is. Payoff is a part of the rewards/upkeep/maintenance phase of play after "the goal cake has already been baked." Its "the receipts." You get your Rep, your Coin, and we discuss if a district Crime Boss in play and what the fiction is for that and whether they pay them off or whether they suck it up and we start a clock for your comeupance. This is all principally constrained/guided and systemitzed and, again, has nothing to do with player goals (because the goals led us here in the first place).

Unfortunately this is all misdirected by my poor choice of example. I simply wanted to point to as literal a case of "GM ends the scene" as I could recall, and to suggest that there's really nothing stopping a GM from ignoring player intentions while still following mechanics that deal with the necessaries. Given you want to reserve "GM fiat" to mean a GM who ignores or unreliably applies mechanics, let's insert a GM who we're satisfied can't deliver conflict-resolution because they ignore player intentions, but does follow rules. Where any rule would require them to heed player intentions, they simply insert intentions that suit them in their place.

Why picture such a GM? Because the strong claim - all closed scene resolution is conflict resolution - ought to withstand any sort of GMing that isn't conflict resolution. It should be impossible to imagine doing the things necessary to achieve closed scene resolution while interposing GM authority between player goals and what comes next.

Maybe this is not going to work. This conversation is looking like some collection "wandering through the corridors of our mind meets a spray of conceptions/priors" rather than focusing on very specific game tech, nailing down exactly what that thing is/does and then working outward from there to nail down "what other forms of conflict resolution are" and "how this diverges from task resolution."

Let me say this as straight-forward as I can.

I 100 % know that task resolution and conflict resolution are not only not the same things, but, phylogentic tree-wise, their common ancestor is sufficiently far apart that they're only superficially recognizable as even linked. Further, GM Fiat is not the same thing as systemically constrained and systemically-directed, GM decision-making. Right now, it feels like the work that we're putting in is coming from a position of obliterating the significant differences in these things and the methodology of attempting to prove the lack of differentiating characteristics is a winding conversation that is pulling random anecdotes from all over the place in a "look at this thing"..."ok, what about this other thing"..."ok, what about this" "ok look at this <thing I think is a smoking gun>."

Maybe not. The only question I'm addressing at this point is whether all closed scene resolution is conflict resolution. If it helps, my intuition is - probably not. It's probably possible to conjure up a version of closed scene resolution that isn't conflict resolution. And something that isn't riding on that is whether task-resolution is the same as conflict resolution.

I'm desperately trying to focus like a laser beam on core concepts of closed scene resolution and we're suddenly talking about Blades in the Dark (which doesn't feature it) and what appears to be either "goals at a global level" or "Payoff not directly indexing goals (which Payoff does, in fact, index goals insofar as "Payoff is the reward phase of Downtime where the receipts of prior evinced and recently actualized goals are materially rewarded via system-directed procedure."

Can you say anything about why you feel it's vital to explain the core concepts of closed scene resolution at this juncture? The only aspect of them that I feel is in doubt is the assumption that all closed scene resolution is necessarily conflict resolution. But I don't see any particular reason you couldn't just assert the constructive characteristics, and motivating utility and virtues, of cases of closed scene resolution that are conflict resolution.

EDIT To avoid being disingenuous, I should add that conceding the possibility of closed scene resolution with task-resolution would open up a path for folk to lay out design patterns, best practices and perceived benefits of that. Those wouldn't be expected to be the same as with conflict-resolution, but one might dislike the possibility nonetheless.

 

[Campbell]

The entire point of closed scene resolution is that at the start of the scene the group establishes what victory in the conflict means. Within the context of a game within which players are allowed to set the aims of their characters how then could that not be conflict resolution? I mean if the GM is disregarding that predetermined victory state than they are no longer engaging in closed scene resolution.

 

[clearstream]

The entire point of closed scene resolution is that at the start of the scene the group establishes what victory in the conflict means. Within the context of a game within which players are allowed to set the aims of their characters how then could that not be conflict resolution? I mean if the GM is disregarding that predetermined victory state than they are no longer engaging in closed scene resolution.

Say the GM establishes what victory in the conflict means. In what way exactly is the scene now not closed?

Speaking in the context of a game that isn't (due to GM setting the aims) conflict resolution of course.

 

[Manbearcat]

Can you say anything about why you feel it's vital to explain the core concepts of closed scene resolution at this juncture? The only aspect of them that I feel is in doubt is the assumption that all closed scene resolution is necessarily conflict resolution. But I don't see any particular reason you couldn't just assert the constructive characteristics, and motivating utility and virtues, of cases of closed scene resolution that are conflict resolution.

EDIT To avoid being disingenuous, I should add that conceding the possibility of closed scene resolution with task-resolution would open up a path for folk to lay out design patterns, best practices and perceived benefits of that. Those wouldn't be expected to be the same as with conflict-resolution, but one might dislike the possibility nonetheless.

My intent on focusing the conversation on closed scene conflict resolution was thus:

* Remove the 5e angle of the conversation entirely because, whether intended or not, you guys are framing this conversation as a full-throated defense of 5e and its making this utterly impossible. For my part (I can only speak for me), this conversation had absolutely zero to do with D&D or 5e D&D. Task Resolution and Conflict Resolution exist en masse outside of the D&D-osphere.

* Examine the core concepts of closed scene resolution because its a novel form of conflict resolution that basically includes ALL_THE_THINGS (TM) and then, once that is absorbed, eliminate constituent parts to arrive at alternative versions of conflict resolution (and see why they still work as conflict resolution).

But that focused conversation obviously can't happen and this has turned into a scattershot "but what about this" "but what about this" "but what about this" of auxiliary or tenuously related (at best) things when we haven't even nailed down the core concepts and, whether it is or not, it looks an attempt to both (a) find a magic bullet and (b) obliterate distinctions.

On (b), the obliterations of distinctions aren't "disturbing" to me. Its not an emotional response. The obliteration of distinctions are a problem for me when they're wrong and that wrongness generates a net harm to understanding concepts and processes.

---

My fallback was going to be to discuss the Gather Information skill in 3.x…the most task resolution aspiring form of D&D…and contrast it with other 3.x skills. Then discuss what inherent features of 3.x make Gather Information about the poorest form of conflict resolution there is.

My concerns there are:

* Too close to the 5e problems so it would again turn into a conversation about/full-throated defense of 5e.

* We would have to get into the weeds of 3.x and I don’t know how prepared folks are to do that (nor whether I want to).

---

However, this is officially the white flag.

I'm out. This conversation is impossible. So please don't tag me again on this.

Folks have a Happy Holidays.

GGs.

 

[Lanefan]

For instance, suppose that we have an action resolution system for power-lifting (maybe some combination of dice and resource expenditure associated with a STR stat). And what is at stake is Can I power-life 300 lb? Now our action resolution system gives us conflict resolution, without factoring in any element of the fiction beyond performance of the task.

But that is not a very exciting RPG.

Here's the reason why it works out like this:

RPGs are, at least in mainstream cases, based around players declaring actions for their PCs. That is one important thing that distinguishes a RPG from mere shared storytelling: the players "interact" with the fiction by declaring actions for particular characters in that fiction. They perform tasks.

But RPGs also, again at least in mainstream cases, aim to be exciting or engaging in some fashion. They presume that players will declare actions for their PCs with an eye to those actions achieving, or at least furthering, some goal. Not just for the sake of seeing whether or not the action can be performed.

What about when the goal is simply to perform the task and see what happens next? For example opening a safe just to see if there's anything in there worth stealing, or climbing a curtain wall around a manor house just to see what other obstacles might lie between the wall and the building before attempting a break-in. The task - i.e. the declared action - is specific (open the safe, climb the wall, etc.) but the goal is both malleable and open-ended and the range of possible outcomes is large.

And yes, sometimes the outcome might not be very exciting. So be it. Maybe the safe turns out to be empty. Maybe there's a piranha-filled moat between the wall and the manor house meaning the break-in is a no-go. Not everything is going to come up roses every time.

 

[pemerton]

This is where I run into a problem with our interpretation of canonical (what I've called "rawest") task-resolution. Players aren't permitted intentions, or if they have them we don't care what they are. Baker explains that

In conventional rpgs, success=winning and failure=losing only provided the GM constantly maintains that relationship - by (eg) making the safe contain the relevant piece of information after you've cracked it. It's possible and common for a GM to break the relationship instead, turning a string of successes into a loss, or a failure at a key moment into a win anyway.​

But what this means - as found in conversation with @Campbell - is that cases of the former (GM maintains the relationship) are not task-resolution. It would seem that canonical task-resolution cannot possibly include that the "goals/stakes are transparently understood by the participants." Maybe that's a good place to stop and check my intuitions and analysis with others thus far.

Imagine a fairly typical CoC or other investigation-type RPG, being played in a fairly typical way.

The players are trying to find out who did it, where the kidnap victim is, etc. They do this by declaring actions like We break open the safe, We rifle through the mail, We "interview" the maitre d' at the club, etc. All standard stuff.

The players have intentions and goals in doing this - to locate the victim, find the dirt, etc. They are also trying to build up a picture of the fiction that - at least in principle - the GM has already established. This is the puzzle-solving aspect that @Campbell mentioned upthread.

This is classic task resolution. The Alexandrian has various bits of advice on how to make this sort of play work: node-based scenario design, and the "three clue" rule. One point of these techniques is that they reduce the need for the GM to break the failure=lose connection (eg no need to put the clue into the wastepaper bin), by creating a systematic framework for retries.

This sort of play relies very heavily on GM-authored secret/hidden backstory; on the players declaring relatively low-stakes actions that will elicit that backstory ("low-stakes" in the sense that, if they succeed, backstory is elicited, and if they fail, the status quo is not perturbed in a way that would invalidate previously-acquired information); on the players not bringing stakes into play that have not already been put there by the GM (or, perhaps, the game system itself - eg loot in D&D); and on the GM managing all the interlocking parts of the fiction, deciding what is at stake in any situation, what consequences flow from one scene to the next, etc.

Is it really very mysterious how different this is from (say) DitV?

(Notice how Harper, in presenting the first of his two diagrams, says "This is remedial stuff as far as theory is concerned (it's a rehash of Dogs town creation, really)".)

 

[pemerton]

To me is seems that if we have conflict resolution and "say yes or roll the dice" we also need to have no myth.

DitV coined "say 'yes' or roll the dice" and is a paradigm of conflict resolution.

And is not "no myth". The GM is required to prep a town for each scenario, and the game provides extensive advice on how to do so, and on how to use the town in play.

The difference from "secret backstory" is that one of the GM principles in DitV is to actively reveal the town in play.

But if we don't have "say yes or roll the dice" I assume the GM could still veto attempts based on their backstory, even if the game used conflict resolution when the rolls were allowed. For example in case where the players were looking the papers from the wrong place the GM could just say without any roll, "you open the safe but the papers are not there." This is saying "no" as the intent of the player was not just open the safe but find the documents.

If the player is declaring an action to look for documents in the safe, and there is (somehow-or-other) fiction that establishes there are no documents there, then something has gone wrong, because the player thinks something is at stake - "Can I get the documents from the safe" - when it fact the answer (no) is already settled.