D&D General [rant]The conservatism of D&D fans is exhausting.

[Maxperson]

Are...

Are you saying RPG campaigns--that is, the games run by human beings here on Earth--don't occur in the real world?

No.

Again, this isn't about DIStrust. It's that trust has to be earned. The natural state of being is neither trust nor distrust, 0 on the trust-vs-distrust number line. A certain amount of, as I have said, allowance/acceptance is required to get the ball rolling, but that is the opportunity for the DM to earn the trust they will apparently be constantly banking on in the future.

I'm saying the trust is about a fantasy. It's about what we are imagining with the game. It isn't as if you are being asked to trust a stranger to walk next to you while hiking.

If you can't or don't trust the DM, you have no business being in that game. Distrust for no reason disrupts games. Distrust because the DM did something to earn it also disrupts games.

 

[Maxperson]

Then you and I have dramatically different ideas of what "harm" means.

I know a bad GM can be extremely harmful. I've seen it happen. A player emotionally shredded by a bad GM's behavior was, in fact, one of the things that kicked my rear in gear for becoming one myself.

Sufficient emotional pain to cause a person to cry counts as "harm" to me.

I don't know that we do have dramatically different ideas of what harm means. I provided for exceptions to my statement, so what happened to that player might fall under the exception.

We still might have dramatically different ideas of what harm means, but your example doesn't show that.

 

[pemerton]

I recognize that they are different things, and I respect your feelings on this matter. But I conflate them because I have the same (not for me) opinion of them, and because narrative-centric and character journey-centric are not different enough from each other to make a difference to me personally regarding the kind of gaming I prefer. I'm sorry if that feels bad and I will try to separate them more in my discourse, even if I don't care for either style.

Of course posters are free to conflate whatever it is that they want to conflate. But it then seems odd for them to complain about others being insufficiently sensitive to the things they care about!

So games like Burning Wheel and Apocalypse Keys are not character centric storytelling. We're not doing things to inject drama. We're addressing the premise of the characters. There's no narrative to serve. We're playing to find out who these characters are under pressure, but it's not about drama or what makes for a satisfying narrative. It's not about character arcs. It's about following them on their journey.

I think there are some games that get framed as "The players pick their goals, and the DM just puts challenges in front of them until they do that." That's not what the target of play should be. The goal is to have players that react to the challenges right in front of them because the characters care, and have characters that change in that crucible.

I'll talk about Burning Wheel, and Torchbearer 2e.

As I've already posted, the core of BW, in terms of the (asymmetric) roles of player and GM, are set out on pp 9-11 of the (Gold) rulebook:

In the game, players take on the roles of characters inspired by history and works of fantasy fiction. These characters are a list of abilities rated with numbers and a list of player-determined priorities. . . .

One of you takes on the role of the game master. The GM is responsible for challenging the players. He also plays the roles of all of those characters not taken on by other players; he guides the pacing of the events of the story; and he arbitrates rules calls and interpretations so that play progresses smoothly.

Everyone else plays a protagonist in the story. . . . The GM presents the players with problems based on the players’ priorities. The players use their characters’ abilities to overcome these obstacles. To do this, dice are rolled and the results are interpreted using the rules presented in this book. . . .​

And in a sense, that is really it; the rest of the rules are just details, techniques for actually making this work.

So consider the necromancer Thoth (Lifepaths (5)= Born Noble, Arcane Devotee, Court Sorcerer, Rogue Wizard, Death Artist):

Beliefs
I will give the dead new life
Aedhros is a failure, so I will bind him to my will
Cometh the corpse, cometh Thoth!

Instincts
Always sustain Wyrd Light (because afraid of the dark)
Always collect bits and pieces
Always read the Aura

Character Traits
Base Humility
Cynical
Inscrutable
Spooky​

So when I'm GMing for my friend playing Thoth, my job is to frame scenes ("present problems') based on these priorities - that in some fashion put them under pressure, or establish stakes that bear upon them more or less directly. Some of that I can do by looking to my prep and my notes - for instance, there is an established Death Cult in the setting (Hardby and surrounds in the WoG, which we have been playing in for a while now; the Death Cult is adapted from KotB, which in our setting is on the edge of the Abor-Alz), and I have NPC death priests and the like written up, and so in the last session we played of this particular game I had one of those NPCs berate Thoth for the carelessness and brazenness with which he was collecting corpses and taking them to his workshop to try and raise them as undead. That puts pressure on the belief I will give the dead new life, and also the trait base humility (perhaps also inscrutable, cynical or even spooky).

There is no sense here of an "arc" or of "telling a story". There are not obstacles in the way of some goal. Nor is there any adventure in the literal sense. There is Thoth going about his business, but not finding life easy.

The role of goal or intent, in Burning Wheel, is not in relation to how scenes are framed. Rather, it applies to resolution (as per pp 24-25, 30-31, 72):

When declaring an action for a character, you say what you want and how you do it. That’s the intent and the task. . . . Descriptions of the task are vital. Through them we know which mechanics to apply; acknowledging the intent allows us to properly interpret the results of the test. . . .

A task is a measurable, finite and quantifiable act performed by a character: attacking someone with a sword, studying a scroll or resting in an abbey. A task describes how you accomplish your intent. What does your character do? A task should be easily linked to an ability: the Sword skill, the Research skill or the Health attribute. . . .

what happens after the dice have come to rest and the successes are counted? If the successes equal or exceed the obstacle, the character has succeeded in his goal - he achieved his intent and completed the task.

This is important enough to say again: Characters who are successful complete actions in the manner described by the player. A successful roll is sacrosanct in Burning Wheel and neither GM nor other players can change the fact that the act was successful. The GM may only embellish or reinforce a successful ability test. . . .

When the dice are rolled and don’t produce enough successes to meet the obstacle, the character fails. What does this mean? It means the stated intent does not come to pass. . . .

Unless there is something at stake in the story you have created, don’t bother with the dice. Keep moving, keep describing, keep roleplaying. But as soon as a character wants something that he doesn’t have, needs to know something he doesn’t know, covets something that someone else has, roll the dice.

Flip that around and it reveals a fundamental rule in Burning Wheel game play: When there is conflict, roll the dice. There is no social agreement for the resolution of conflict in this game. Roll the dice and let the obstacle system guide the outcome.​

So when I, as GM, frame those scenes that are based on those priorities Thoth's player has determined, the result will be conflict - for instance, the NPC death priest berates Thoth, while Thoth is trying to persuade the priest to allow him (Thoth) to take away yet another body for practising Death Art on. The intent is that the death priest allow Thoth to take away the corpse; the task is a series of arguments and other rhetorical devices, presented and (mechanically) resolved via the Duel of Wits subsystem. If Thoth's player wins, then he achieves his intent; but as it turned out he lost soundly, and so agreed to allow the body to be laid to rest in its ancestral catacombs. (The existence of catacombs beneath the city is another bit of lore established earlier in our play in Hardby.)

There are multiple dimensions of character change in BW. One is that the player is free to rewrite Beliefs and/or Instincts at the point of any lull in the action. Another is that, among the ways to earn Persona (a type of "metacurrency") are Embodiment and Mouldbreaker (from p 64):

Embodiment
When a player captures the mood of the table perfectly and further drives the story onward, one persona point is awarded. Moments like great speeches, desperate decisions or gruesome revenge fall into this category. This is a tough award to get, as a player really must go above and beyond in his roleplaying.

Moldbreaker
If a player comes to a point in the story where his Beliefs, Instincts and traits conflict with a decision he must make - a direction in which he must go -and he plays out the inner turmoil, the conflict within his own guts, in a believable and engaging manner, then he earns a persona point.

Other players, as well as the GM, may nominate each other for this award. A majority vote at the table awards the point. To earn the reward, the player must really push his character.​

Now, as I've already posted, I wouldn't normally think of BW as a sandbox game. It doesn't have the right sort of focus on place and journey. Torchbearer 2e, as I've also posted upthread, I think can be a sandbox. But it nevertheless has a lot of overlap with BW. Characters have a Belief, and Instinct and a Goal - each can be changed at the start of the session. And characters of 3rd or higher level have a Creed, which can be changed during a respite (roughly each 8 or so sessions). In addition, the way the game works - its interlocking resource and recovery cycles - means that all characters need loot, to pay for things. The Scholar's Guide, p 218, makes this point about how these priorities interact:

In Torchbearer, we give you four tools you can use to make your expeditions more than just loot hunts and massacres. Beliefs, creeds, goals and instincts all contain the potential to hook in players and push play to another, more intense level.

If, during an adventure, you find an opportunity to present a player with the choice of either playing a belief or acquiring loot, then you’ve offered what we call a meaningful decision. At this juncture, the player must decide what is most important: satisfying that belief or scoring some loot.​

Level advancement is dependent upon accruing Fate and Persona; and these are earned in the following ways, among others (pp 84-5):

When a character stands up and takes action in a manner driven by their belief, they earn one fate point. . . .

Making a test toward achieving a goal but not accomplishing it, earns one fate point. . . .

If a character acts against their belief in a dramatic fashion - if they make a decision in the game that’s counter to what they believe - and they let everyone know about their inner struggle through their performance of their character, they earn one persona point. . . .

A player earns a persona point for an internal crisis:

*If their character stands up for their creed in a moment of crisis.

*If their character’s creed is violated, voided or broken and they demonstrate the inner turmoil born from this revelation.​

Adventurers do not need to succeed, but they must take clear, unequivocal action motivated by their creed to qualify for this reward.​

(Earning loot does not, in itself, play any role in PC advancement.)

Here are the Elven Dreamwalker Fea-bella's belief and creed:

Creed: These are dark times – the free peoples must stand together!
Belief: Only the rich get anywhere in this world – I must become rich!​

And here are the Dwarven Outcast Golin's:

Creed: Elves are lost in dreams; they need grounding in reality.
Belief: Elves are unstable!​

So the interactions and confrontations with Celedhring (Fea-bella's uncle, it turned out, now a Barrow Wight who, according to the pronouncement on the doors that lead to his tomb, "lies . . in communion with the Outer Dark"; with Megloss, the rival Dreamwalker; and with Lareth the Beautiful, the half-brother Fea-bella didn't know existed until she found reference to him in the books written by his human father, the seer Pallando, Beholder of Fates; are not just adventurous challenges. They invite responses from the players that express, embrace or struggle with these beliefs and creeds.

A question that I think some RPGers have - perhaps some in this thread, even, though I don't know - is what will make this sort of game "go"? As in, if the GM doesn't have an "arc" of adventure mapped out (like the DL modules, or many APs); or doesn't have some outcomes in mind, how does the game happen? This is where the technical details of resolution matter - by introducing consequences on a failure, the GM "reframes" the situation so as to step up pressure, or introduce new pressure, and the players respond to that with more action declarations, and so events unfold. These games aren't feasible if the action resolution methods are treated just as "advice" that the GM and player may follow, or not, as they feel like.

Related to this is that rates of failure on tests are high in these games - I don't keep a log, but I would say 50%+.

Torchbearer 2e, unlike BW, also uses events tables to help make the game go: that's part of what makes it more sandbox-y, and less intensely player-driven, than BW. But for TB2e as well as BW, the statement from the BW rules quoted above is true: There is no social agreement for the resolution of conflict in this game. The dice must be rolled. That's what drives play; not some prior conception on anyone's part of what will or "should" happen.

 

[pemerton]

You seem to be conflating sandbox with realism. You can have very realistic sandbox games, and very unrealistic sandbox games. Sandbox is just a style of play and realism can and does vary greatly from one sandbox to the next.

If the setting is unrealistic, then I don't understand how the heuristic extrapolate what is the most realistic/plausible is supposed to be applied.

 

[pemerton]

It's not like there's a dictionary of terms for people to use. It doesn't help that if someone doesn't use the "correct" term we aren't told what term should be used we're just repeatedly told we're wrong.

So if people are using the term narrative or character driven incorrectly maybe you should clearly explain what terms you feel are correct.

Correct for what?

If you mean, correct for describing how Burning Wheel plays, well I've posted the relevant rules, plus examples of play, multiple times in this thread. Including, again, just upthread of this post.

The core is that the player determines priorities for their PC, and the GM then frames scenes that speak to those priorities. The resolution system then builds on this, by way of (i) "say 'yes" or roll the dice", ie the dice are rolled when something is at stake, as determined by the player's priorities for their PC which the GM is addressing via the scene they've framed, and (ii) intent + task, with negation of intent being the key requirement for a consequence of failure.

If you want jargon to describe this, scene-framing story now is probably as good as any.

This broad approach can work for RPGs other than BW, though there will be differences that may affect the actual play experience quite a bit. As I've posted in this thread, Torchbearer is somewhat similar but certainly not identical to BW.

And here's a thread about using an approach similar to this for 4e D&D: D&D 4E - Pemertonian Scene-Framing; A Good Approach to D&D 4e

 

[Maxperson]

If the setting is unrealistic, then I don't understand how the heuristic extrapolate what is the most realistic/plausible is supposed to be applied.

There are a few reasons. First, realism is a spectrum, not all or nothing. Second, there are different dials. You can have one thing be very unrealistic, while having ten other things be fairly realistic. Third, fantasy realism is a thing. Fantasy realism says that since the lore explains that magic exists in the world, for the purposes of this game/setting, magic is realistic. You can use that to help you determine what is realistic and what is not.

 

[Thomas Shey]

But that’s not the example. There’s no misunderstanding here. @Lanefan specifically stated why he described the floors that way.

I'm just saying that your premise seems, at least in part, based on the idea that make-it-as-you-go GM in a sandbox is most likely steering the players toward what he wants, and that doesn't seem to match my experience when I was doing that. I rarely even had anything I wanted out of them. By the time I had any plans, it was usually because they showed they were interested in something and we might as well explore it.

In other words, though I ran it as more or less an old-school sandbox, to the degree there was any intention, it was usually PC-originated even though I didn't have anything pre-done other than some relatively large-scale rest state things. I mean, I might have decided that at some particular point in the future City-State X was going to invade City-State Y, but just as likely exactly when that happened was going to be an output of some regular die rolls, and depending on what the PCs were doing it might have little or no impact on them.

 

[pemerton]

I think a major part of the definition of a sandbox is not just the possibility, but the expectation, that the PCs will start in a place, and be presented with various options as to possible paths to follow, with various degrees of clarity.

The most basic sandbox imaginable would be a Zork-like text adventure with a starting point of "You start in a room. There are doors to the north, east, and west. What do you do?"

I think another feature, which is brought out in your example, is that the default starting point for a certain sort of sandbox is low stakes player decision-making. Play begins with the players in a position where they need to acquire information, and they do that by making low stakes action declarations that will prompt the GM to reveal that information.

In Gygax's PHB advice on Successful Adventures, he draw an express contrast between player goals of obtaining information, and - once that information is obtained - the different, and higher stakes goal, of acquiring a particular loot, perhaps by targeting a particular dungeon denizen.

Here's a scenario that also starts in a room, with decisions to be made - but they're not low-stakes! Burning Wheel The Sword Demo Adventure PDF

When players sit down with me to play this demo, I give them the following preamble:

You’ve journeyed long through this crumbling, ancient citadel, down through ruined chambers into muck-filled tubes. You arrive, at last, in the wreckage of this collapsed temple. Laying on the shattered altar, in the chamber before you, is that which you seek: The sword!

After the preamble, I lay out the characters and describe each one in brief. After the players have chosen their characters, I instruct them to read their Beliefs. Then I simply ask, “Who gets the sword?”

Mayhem ensues, and suddenly we’ve got game. . . .

I’ve found it useful to start the scene with the whole group in the doorway to the chamber - still in the tunnel, really. The Roden, if he’s in play, should always be in the lead.​

Here are some of the Beliefs:

The Dwarven adventurer: This sword was a treasure of my clan for generations, stolen by foul Roden and abandoned here. I’ll restore it to its rightful place among my people.

The Elven bard: This sword was made by my father. Using its markings, I will demonstrate its origin to my companions so they cannot dispute its ownership.

The human gambler: Master Kogan of the gambling house is going to break my knees if I don’t pay off my debts. I’ve got to get paid in this venture! and also I was the one who figured out where this treasure was; it belongs to me!

The Roden cultist: To enter the Fields of Paradise, I must present my Visionary with this fabled sword.​

These players have options as to which path to follow. They can cooperate, argue, fight, try to betray one another, etc. But I don't think it has the character of a sandbox.

 

[pemerton]

So I want to come back to the way I framed it before...are statements like "the GM adjudicates the world" and "the GM adjudicates the rules" railroading? The key point to me here seems to be that the GM is not adjudicating successes; the players are.

Well, on a success there is no adjudication: the task succeeds, and the intent is realised. The GM can embellish.

On a failure, it is the GM who decides what happens. But (i) it must negate the intent, and (ii) the GM is still working within the broader framework of the game, which includes presenting problems based on player priorities.

Consider an example: your players are in a port city and want to hire a ship for a week-long passage.

Prep-DM wrote his city in advance, and he has a list of ship types, how many are are in port (roll for), and the costs associated with the journey. He consults the tables, rolls, and gives the players the options based on ship type.

Improv-DM made his city 20 minutes ago, so he doesn't know. He rolls up three NPC captains to give the players a choice, then picks ship types and fees that strike him as reasonable in the moment.

Burning Wheel DM doesn't have to take on this authority--the players declare their intent, and roll. They succeed, so then they define the captain and kind of ship they've hired, as well as a reasonable cost.

The first action here is something like, "I look for a ship and captain who will take us where we want to go".

If there is nothing at stake here - eg no one has a Belief that pertains to travel, finding a ship, getting to some distant place, etc - then the GM should say "yes". The search for a ship is just colour, and can be easily enough narrated.

Otherwise, the mechanic for finding a helpful NPC is called Circles. In this case, the player might augment with appropriate "wises" (a general category of knowledge skill) - eg "Ships-wise", "Captains-wise", perhaps even "Docks-wise" or "Trade-wise" (though these, being a bit more tangential, would mean a higher obstacle for the augment). The difficulty of the Circles test depends on factors like how common is this sort of person in the PC's social milieu and how urgently does the PC want the meeting/finding to take place?

If the test succeeds, the NPC is found. If it fails, the GM narrates a failure as usual. A distinct option for a failed Circles test is the "enmity clause" - the PC finds a NPC, but they are hostile in some fashion - eg maybe in this case the captain really doesn't like the cut of this character's jib, and is going to insist on a higher price.

Once a ship and captain are found, there is then the issue of price. In Burning Wheel this is a Resources test. The same rules apply for this as for any other test. There are associated rules for haggling. Whether this is mere colour, or is resolved via a simple test, or perhaps via a Duel of Wits for the haggling, depends on how big a deal the GM or the player wants it to be (if haggling is in the offing, the player can call for a Duel of Wits as much as the GM can).

Here's an actual play example of this scenario; although the game being played is Torchbearer 2e rather than Burning Wheel, the basic dynamic of play is pretty similar. The PCs wanted to get from Nulb (in the Troll Fens, in this game) downriver to Wintershiven, and so tried to persuade a pirate ship to carry them there. You'll see that the presence of the pirates in Nulb had already been established - in the previous session all the accommodations in Nulb had been occupied by pirates (meaning that the PCs had to sleep on the streets), and I've included some earlier events of this session which show more about how that fiction was established:

Telemere then decided there must be some other secret room or cache, and tested his Beginner's Luck Stonemason to find it. Golin - whose Creed is that Elves are lost in dreams, and need grounding in reality - declined to help him; but Fea-bella - whose Creed is that These are dark times, so all Elves need help! - decided to help, even though she thought it was hopeless.

The test failed, and so Telemere found nothing. Rather, Golin - watching from atop the shrine - saw a vessel sailing down the river towards them. He recognised it as a pirate river galley, especially when it ran up the Jolly Roger! He also thought he heard some words drifting through the still air, something like "Is that Golin the Beardless?"

The PCs decided to flee. With the players' approval, I decided to give the Pirate a disposition of 2d6 (it being a bit arbitrary), and got 9. The players got a disposition for their PCs of 9 also. They won the conflict, but with their disposition reduced to 5, so I explained that they escaped the pirates, but were lost somewhere west of the river and would need to start a new journey. The players accepted this compromise.

Roles were allocated: Fea-bella as guide (because she had the map), Telemere as Scout, and Golin as Forager (on Instinct) and Cook. A roll for weather confirmed that it was still Clear and Cool; but Trouble on the Road indicated that the PCs would get Lost, adding 2 to the Toll and requiring another Pathfinder test.

The first Pathfinder test failed; I don't have any notes indicating a twist, so I think this made Fea-bella Exhausted, and the others Hungry and Thirsty. Golin successfully found some forage and cooked it, alleviating a point of Toll. The second Pathfinder test failed too; here I opted for a twist, and called on Telemere, as the Lookout, to make a Scout test. He failed, and so the PCs stumbled upon a group of 3 pirates (on the shore, away from the rest of their crewmates), who initiated a Capture conflict.

I rolled fairly well in this conflict, and so the PCs were captured, but with a half-compromise owed. I can't remember now exactly how we resolved this (EDIT: I think it was here that we established that the rescued prisoners escaped, so that only the PCs were caught), but the action rolled into a Convince Crowd conflict, as the PCs - led by Golin, the only Orator - tried to persuade the pirates that they were friends of Tolub and so should be taken to him free rather than as captives. The pirates had only 2 hp for this conflict, while the PCs had 3, and the PCs won with only 1 hp lost. What the pirates really wanted was Fea-bella's silver bucket, but the compromise was that instead they were given some draughts of magic water from that "sacred bucket".

The PCs were therefore taken back to the pirates' vessel, which duly sailed them back to Nulb. They paid of some of their toll by spending the ruby to procure victuals from the pirates. Golin's galoshes helped him, given that some of their journeying had been on the river bank; and they took four conditions between them, leaving Golin Exhausted.

Which produced groans, given this was where the PCs had been trying to get away from at the start of the session. But the town event roll was a 6:

Bandits. The village is beset with bandits (3d6 in number). Remain in the adventure phase until the bandits are driven off or until you arrive at a different settlement.​

In this case, of course it was pirates - 11 of them - who were not just occupying the accommodation, but had literally taken over the township. The PCs didn't want to fight pirates. They tried to persuade the pirates to let them sail back south, with the Jolly Roger lowered and the pirates themselves hiding below deck so as to avoid notice and capture by Wintershiven authorities; but this attempt - resolved as a simple Orator vs the pirate captain Orgoth Bloodeye's Orator 3 - failed, and so Orgoth shifted to negotiation instead: he would do what the PCs asked in exchange for the silver bucket, in advance. This was resolved as a full Negotiation conflict, and Orgoth (aided by his second-in-command) succeeded with no compromise owed. This was my best rolling of the session - with 9 successes out of 9 dice on Orgoth's two rolls - and caused much gnashing of teeth and wails of lamentation from the players, who had to hand over their remaining loot.

Fea-bella was in charge of Sailing the vessel, but her test failed, and a severe rainstorm came up (which meant that I imposed 2 Toll on each PC, which otherwise I was not going to do). The plus side was that the rain gave Telemere +1D to his Scout test to ensure the pirates weren't spotted as the ship approached Wintershiven. The town event roll for the Walled Town was a 9:

Fire. One facility has burned down. Roll 2d6 to determine the facility . . .​

A roll of 7 indicated that it was the docks that had burned down, which cemented the need to leave the ship to the pirates while the PCs entered town.

To pay off their tool, Fea-bella and Telemere each relied on their cloak as raiment and took Angry as a condition; while Golin crossed off his scarf (sodden from the rain) and his rope and candles (which he had burned up to try and keep warm on deck).

I hope that this gives you a sense of how the introduction of consequences draws on the extent fiction, but is not impartial. But nor is it directed towards anything like an "arc". It's about presenting a situation to the players (and thus their PCs) that prompts them to action based on their priorities for their PCs.

Based on my understanding of your views, you'd find the first to not be railroading (as in Dungeon crawling), the second to be railroading (because based only on the GM's whim) and the third to not be railroading (players in control).

The first may or may not be part of railroading - again, without more context it's hard to tell. Eg how do the players come to know what the ship-hiring options are? how is the reaction of the captain to their offer determined? Suppose that the GM has the captain say "Yes, I can take you to <place>, but only via <other place where the GM's living world will make some other stuff salient>." Now the play - time spent at the table, fiction being jointly created, etc - is going to be about this GM-driven stuff. That looks a bit railroad-y to me.

Whether the second is railroading also depends a bit on what is going on. If this is really just a bit of colour narration on the way to framing the scene where the PCs arrive at <place>, then I don't see the railroad. It's just embellishment. Conversely, if the whole session is spent engaging with this improvised content the GM has created, that sounds pretty railroad-y.

Do I understand Burning Wheel correctly? Does that capture your views?

I hope the above helps answer this.

I had an example about hiring a ship earlier. Consider a variant of that. The players are in a northern city and the ice has closed passage for the winter. In my GM notes I have written "Ice blocks all travel". However, one of the characters must return a yeti horn to the south, because it can make a potion that will save their sister's life.

I think this is a compelling character moment and I want to see it happen. Therefore I say, "it turns out this year there is a gap in the ice, and you can find a captain willing to risk it for a large fee".

In this case, I'm not being impartial, I'm making the world change in response to the PCs stated goals.

So, in Burning Wheel, the key difference here from the previous example is how is the weather determined? Maybe it's just colour. Maybe a PC has the Weather Sense trait, which is described thus: "Unless the GM plays with strict house weather rules - I don't - the Weather sense in essence gives the player ability to dictate the weather. Why would a player pick this trait unless he were concerned about having the right weather for something or other?" But the "dictation" is via the standard rules - a test on Perception - and so if the test fails, then the GM narrates a consequence in the usual way.

But if the player wants to get south (say, the belief I will return the yeti horn to the south), and either has failed their test for Weather Sense or else doesn't have some applicable test to make for the weather, then the GM is entitled to frame a scene that puts pressure on that belief. Whether the ice-closed passage counts as such a scene will depend on the details. Maybe that PC, or another PC, has the Hauling skill and a Belief like Heave-ho, away we go! and the players decide to have their PCs portage a ship over the ice until they find water!

 

[EzekielRaiden]

The reasons are irrelevant.

When a person strIght up says, "I don't understand why you would do any of that when the point of play is doing something else", the reasons are not irrelevant. They are literally the topic of conversation as a result of that question.

You might as well have told Hussar "stop talking, no one wants to talk about that." (He was the one who originally asked the question, not me.)