RPGing and imagination: a fundamental point

[Campbell]

Sure, a GM may illicit player intent and make success always mean a player realizes their intent. If done in such a way where that is the binding expectation at the table, we no longer have task resolution. We have conflict resolution. What makes conflict resolution well conflict resolution is that it directly resolves what players are attempting to accomplish in a way that binds everyone to those results.

GM fiat cannot exist in a conflict resolution framework because in a game like Apocalypse World the GM has no authority besides what the game gives them. That means when Plover puts a gun to Diamondtooth's head and yells 'Get down on your hands and knees' the GM/MC has no choice but to allow the Go Aggro mechanics to take effect. Doing otherwise is explicitly and emphatically against the rules of the game. They also are duty bound to make GM Moves and not allow play to meander to conflict neutral play.

Addendum: You can have somewhat binding task resolution rules as seen by what you see in something like Pathfinder Second Edition, but the key difference is that they do not actively resolve the actual stakes. It's still largely up to GM judgement to start and end scenes, decide on fallout, etc.

 

[clearstream]

I don’t have time to dig in too deeply and I’m going to be away the bulk of the day, but we’ve got issues right away. The most significant one is:

* The Social Interaction mechanics in the 5e DMG are not task resolution mechanics. They are classic conflict resolution mechanics.

1) I believe I was the very first person on ENW to bring these up at the release of the game, siting their function, their functionality, their kindred relationship to Pictionary and Apocalypse World.

Virtually no one had any any idea what I was talking about because “no one reads the DMG“ (apparently) as the saying goes.

2) Fast forward two years later in like 2016 and I brought them up again (as I had run the game). Again and still, no one knew what I was talking about.

3) Insofar as I did get conversation around it, it was almost universally “I don’t like artificial minigames. Social stuff should be freeform/roleplay.” That is a Trad players code for “I don’t like (the binding nature and superstructure of) conflict resolution.”

4) However…the issue (insofar as their is one…most Trad D&D players will call this a feature rather than a big), while the superstructure is there and it appears to instruct, constrain, and encode goal + play loop + win con/loss con binding result to facilitate “the minigame of conflict resolution”…its embedded in a rules text that expressly grants GM opt-out discretion any/all of the superstructure, the constraints, and the results of play if the GM deems any of those at odds with the conception of what is good for “the fun” and/or the story/trajectory of play.

That means, in the final analysis, “situation resolves” is governed by GM Fiat. So while the Social Interaction mechanics are indeed conflict resolution, the text has to be holistically indexed to determine whether “situation resolves” is governed by inviolable systemic constraint and binding superstructure or whether it’s governed by GM Fiat. The text (and culture for that matter) enshrining opt-out of system at GM’s discretion supplants inviolable with Fiat. Contra, when a text enshrines “follow the rules” and “neither plan nor subvert outcomes” it supplants Fiat with inviolable. That matters deeply to the process and experience of play for all parties. Further still, if the game delivers on what it promises by just following through on its inviolable systemic constraint and binding superstructure, that backstops everything that comes before it.


Now, any given 5e table can autobiographically encode inviolable and discard opt-out. That is Fiat at inception, but, if followed through upon over the tenure of play, becomes standard practice that informs the process and experience of play for all participants. But that game becomes (for instance) “Manbearcat’s or clearstream’s home game of 5e” which likely bears little resemblance to the rest of the “I don’t like artificial minigames, roleplaying should be freeform, and the GM needs to have discretion to opt-out of outcomes that they deem bad for fun (which is enshrined as the highest virtue in the text) and the story (which is enshrined right behind “the fun”)” 5e space that dominates the culture. That matters.

I'll need a few reads to parse this, but at first glance - we're not debating whether "Situation resolves via referee" can replace "Situation resolves" in the top chart, and otherwise they're the same, are we?

 

[clearstream]

Sure, a GM may illicit player intent and make success always mean a player realizes their intent. If done in such a way where that is the binding expectation at the table, we no longer have task resolution. We have conflict resolution. What makes conflict resolution well conflict resolution is that it directly resolves what players are attempting to accomplish in a way that binds everyone to those results.

GM fiat cannot exist in a conflict resolution framework because in a game like Apocalypse World the GM has no authority besides what the game gives them. That means when Plover puts a gun to Diamondtooth's head and yells 'Get down on your hands and knees' the GM/MC has no choice but to allow the Go Aggro mechanics to take effect. Doing otherwise is explicitly and emphatically against the rules of the game. They also are duty bound to make GM Moves and not allow play to meander to conflict neutral play.

Addendum: You can have somewhat binding task resolution rules as seen by what you see in something like Pathfinder Second Edition, but the key difference is that they do not actively resolve the actual stakes. It's still largely up to GM judgement to start and end scenes, decide on fallout, etc.

We can distinguish between resolution systems on the basis of who proposes performances and who proposes outcomes, if we like. But that is not what differentiates task resolution from conflict resolution: it's that the resolution binds participants to an outcome. Right?

 

[pemerton]

Yes.

If the forge is hot, and the hammer clanging, then I'm not sure why a sword is not being produced?

I'm also not seeing any connection, here, between task and conflict resolution. I mean, suppose the forging of the sword is being resolved as part of a skill challenge, or similar sort of extended conflict resolution process. Maybe the forge is hot, and the hammer clanging, but part-way through the process the smith is interrupted in some fashion - that could easily be an event in conflict resolution.

It's striking how the norms of modes of play push more toward assertion or negotiation.

To be honest, I'm not seeing this either.

In the example I just gave, suppose that the player fails a check which - per the rules of the game - shifts authority to the GM to narrate some consequence. And the GM narrates that interruption. In your terminology, that looks like the GM asserting. But (i) the RPG in question could easily be HeroWars, or Torchbearer, or Burning Wheel, or some other game that uses conflict resolution, and (ii) it would be a perfect example of what Baker means by rules easing and constraining negotiation.

 

[pemerton]

"Suggestions" is really your take on it, not how folk who embrace such modes necessarily see themselves. In the example here, GM is reminding player of established fiction (table on fire).

I don't understand why one sort of verb of communication ("reminding") is permitted here, but another ("suggesting") is not.

What is the permitted verb to describe a player proposing that the fiction include a certain action taken by their PC, which then has to be retracted or modified because the GM "reminds" them of some bit of the fiction.

 

[pemerton]

that is just framing. What’s at stake - the safe is keeping me from seeing what’s inside. Why do I try to crack the safe - To see what is inside. I’ve just defined the safe example as a conflict resolution without needing to resolve an external conflict.

This makes no sense to me.

You're saying that it is possible to have, as an instance of conflict resolution, that I crack the safe. OK. That doesn't show that there is no difference between task and conflict resolution.

Now, suppose that you succeed on your roll to crack the safe - is the GM nevertheless at liberty to say "Well, the tumblers all fall into place, but you discover the hinges are welded shut, and so the door won't open"? If the GM enjoys that liberty (whether a liberty to make that up, or a liberty to introduce it into the shared fiction by reading from their notes), then we know that the resolution in question is task resolution.

If they are not, we know it's conflict resolution.

We can look at it through the lens of failure also. If you fail on your roll to crack the safe, is the GM nevertheless at liberty to say "As you walk away, you see some photos sitting on the desk" and those turn out to be recent photos taken of the inside of the safe? If yes, task resolution; if no, conflict resolution.

The difference between the two is not macro/micrco (as Baker himself explains in the blog I linked to), nor external/internal (whatever exactly that encompasses). It's the relationship between (i) the player establishing what is at stake, (ii) the making of the roll/check, and (iii) what permissions the GM enjoys, based on the result of the check, to introduce new fiction that breaks the relationship between the result of the check and gaining or losing what is at stake.

 

[pemerton]

Take a look at the Paragon system to see how they differ. With conflict-resolution, we're bound by the outcome, but not the fiction getting there.

<snip>

Benefit: resolving outcomes makes that binding on everyone. Disadvantage: some folk aren't comfortable with leaving how we got there in doubt - possibly leading to narrations we find jarring (maybe they dislike me saying I cracked the safe even though I failed my roll).

Paragon is very close to fortune-at-the-beginning. The post-roll narration is expected to include "how we got there".

But in a system like Burning Wheel or Torchbearer, the "disadvantage" that you describe is quite absent, in that the dice can't be rolled until the task is declared - because until the task is declared, we don't know what skill/ability to test, nor what the obstacle is.

Apocalypse World also requires the task to be declared, so that we can know how to resolve it.

Also, disliking me saying I cracked the safe even though I failed my roll isn't about jarring narration. The roll is not part of the narration; it's a thing that happens in the real world, not the fiction. This is simply someone who doesn't like the player's roll to have implications for parts of the fiction that the player doesn't "own".

It's worth highlighting that the reason you might not (in conflict resolution) could be you fail your task but get your objective (you don't open the safe, but turning away you see the dirt in the wastepaper basket.)

What RPG do you have in mind, which permits a successful check to be narrated as failure of task but success of intent?

 

[pemerton]

Its interesting to me that RPG’s started off in the special conflict resolution case of task resolution and have generalized out from there.

One of the earliest bits of non-combat resolution in RPGing is listening at doors and looking for secret doors.

And neither is a special case of conflict resolution, or any sort of conflict resolution. It's clear from the classic D&D rules that the roll to find a secret door might come up 1, yet nothing be found (if there is no secret door noted on the GM's map). Or that the roll to listen at a door might succeed, yet nothing be heard (if the room is empty, according to the GM's key).

So I don't know on what basis you are saying that RPGs started off in a special case of conflict resolution. They didn't (for non-combat), because it doesn't make sense in map-and-key resolution.

The earliest non-combat examples of conflict resolution that I know of are found in Traveller (1977), in the rules for Streetwise skill and Vacc Suit skill.

 

[pemerton]

Its not the perfect fit (as even D&D combat has some elements of "situation resolves" that the GM can subtly opt-out of due to extra-combat, holistic authority granted to the GM), but the easiest way to look at it is to contrast combat resolution and noncombat resolution in all non-4e D&D.

Right. That's why, in my post just upthread, I've talked about non-combat resolution.

Note that Vincent Baker, in the same blog post, asks this question:

Something I haven't examined: in a conventional rpg, does task resolution + consequence mechanics = conflict resolution? "Roll to hit" is task resolution, but is "Roll to hit, roll damage" conflict resolution?​

What he has in mind is that, if success on the attack roll automatically entitles the player to a damage roll, then the GM doesn't have the capacity to insert some new fictional thing (such as the safe being empty) between success on the task and getting the intended outcome.

 

[pemerton]

I think about what the top chart looks like if someone keeps picking goals that won't resolve the situation?

What does this mean?

You are supposing that the player has picked a goal (= set stakes) and yet achieving those goals won't resolve the situation. So what is the situation? It must be something the GM has in mind, but that hasn't been shared with the player. Which would be the GM fiat of the second diagram.

We can distinguish between resolution systems on the basis of who proposes performances and who proposes outcomes, if we like. But that is not what differentiates task resolution from conflict resolution: it's that the resolution binds participants to an outcome. Right?

Conflict resolution: succeeding at the check, which (in all the RPGs I can think of) means succeeding at the task, also secures the goal/intent/stakes.

Task resolution: succeeding at the check means succeeding at the task, but the GM then has to decide whether or not succeeding at the task secures the goal/intent/stakes.

Examples of task resolution are rife in classic D&D - as per my post not far upthread - because the connection between task and stakes is determined by the GM's secret map and key, which is independent of the player's action declaration.

To the extent that RPGing generalises from that map-and-key model (see, eg nearly every published adventure ever, the whole framework for CoC, etc) then it is using task resolution.

To the extent that RPGing uses techniques like "As you turn away from the safe, you notice some paper in the waste paper basket . . ." or "When you search the bodies for loot, you find a note . . ." or "Instruction to the GM: if the PCs kill such-and-such plotting NPC, then have this other NPC come out of the shadows to take control of the plot" (see, eg, a good chunk of all the published adventures ever, the whole framework for CoC if it is not to grind to a halt, etc) then it is using task resolution.

If resolution binds participants to an outcome, but the GM gets to choose that outcome, and/or the outcome is not transparent in the framing of the situation, and/or the GM can introduce that outcome after the check is framed and rolled, then we have task resolution, not conflict resolution.

Again, this would be the GM fiat of the second diagram.