GM fiat - an illustration

[pemerton]

I find myself unable to decide if there is a meaningful difference between "DM Fiat" and "DM makes a decision" as the two ideas seem - speaking with natural language - synonymous in my head.

Then you'll have no trouble following my usage in the OP and in my follow-up post.

 

[soviet]

Oh wait, you misunderstood me!

When stuff like this happens in my game, I go all Mentat, have a mild seizure, and imagine every single possible outcome that could happen.

Then I pick six to potentially happen in the game and then roll!

So I have infinite possibilities, too! And then six can potentially happen, instead of just the one!!

I wrote six possible responses to this post

 

[Crimson Longinus]

Yes, of course a GM can make good faith decisions for NPCs. I've said a few times that I'm not really worried about bad faith GMing.

My point is that I like for there to be limits on this decision making. That there is a time and place for it, and then a time and place for mechanics. Absent any mechanics, the more that a given situation of play relies on multiple points of GM fiat... like the Alarm spell from the OP, but also many other instances... the more concern I have as a player and a GM.

For example, I think things like the Alarm spell work best when they interact with Random Encounter rolls and the like.

First, I think almost all of us recognise that some limits are beneficial, were they to take shape of following logic, established fiction or genre conventions or hard and fast rules.

And I actually partially get your sentiment. Some situations in some games are such that, it becomes too much of "GM makes something up" and some additional structure would be beneficial. I for example have long advocated for more guidance, examples and structure for 5e skill section (sadly 5.5 made it even worse than it was, which certainly was an impressive feat.)

I just really, really do not think the Alarm spell is such a situation and I feel that the TB method that is seen as preferable to some strips elements that I feel to be important. But like I've been saying before, it also is a matter of preference at what level of detail you want to handle certain things in an RPG. In practice we end up eliding various factors that matter when utilising various rules, and I don't think there is one objective answer to how much is too much.

The above... the GM has a good idea of what the party can do. They have levels, for crying out loud... a rough measurement of their capability. The game uses challenge ratings and encounter difficulty and the like.

Yes, but I think that is rather different thing than specific responses and counters to specific tactics.

What are they? To make the game fun? To portray a believable world?

Seriously... I'd love to hear what your actual guiding principles are about 5e and where they came from than this back and forth situation we're in.

I think several people have been trying to explain them to you. I prefer to run 5e in pretty high myth way, prep situations not plots, and have clearish mental guidelines to fiction rules connections. These all are sort of "limits to GMs whims" that you as well seem to desire.

Like before the session begins I want to have clear picture of what the situation is, what are salient parts, what are the motivations of NPCs, and what their capabilities are. I also have more general ideas about the setting, and what rules are used to represent what, which help me to extrapolate consistently. Like as one example that I've mentioned before, I like more powerful NPCs to (roughly) use similar class mechanics than PCs, and the classes sort of have diegetic existence. I also have rough ideas of what levels mean and how common NPCs of various levels are. So if I need to extrapolate capability of a just invented NPC, I have guidelines for it, it cannot be just anything. Similarly this is information the players too can use. Like in the last session they were dealing with a wizard antagonist, and when sussing out what the NPC could potentially do, the knowledge whether certain spell was a wizard spell was relevant to them.

Now where do these practices come from? Not from 5e DMG certainly! From decades of experience with running RPGs and general accumulation of osmosis of good practices. And I don't mean to say one must run 5e this way, but it works for me, and I think it actually addresses some of the issues you are having even though not in the exact manner you would prefer.

Again, I find this hard to accept. Do you routinely sic a lone kobold on your 15th level party? Do you send multiple dragons after your first level party?

Not multiple. It was lone adult dragon against a second level party. The PCs wisely fled. And of course the PCs often encounter NPCs that are much weaker than them, though many of them have good sense not to attack them. But yeah, in game like D&D some considerations for "level appropriateness" need to be done, but actually far less than the guidelines might suggest. The PCs in my game have fought things that by the guidelines should have been a clear death sentence and won. Furthermore, in my game this "level calibration" happens mostly via the player choice. There are different sorts of dangers and more safe and more lethal places in the world, and the players make choices about what to engage with.

Because I'm playing a game and how the game works matters to me? As a player, I like to understand how we arrive at some new situation. As a GM I want my players to understand how we arrive at some new situation. Their knowledge of the game and how it works is more important to me than the "mystique" of the setting.

Because the less players understand about this stuff, the less agency they have.

Ok. I get that. But also I don't think this is really "actionable" information. Like if an enemy bypassed their defecences via certain method, then it is useful information. They can take precautions against that methods in the future and use similar method against enemies. To me this is more fun information than just dice odds. It gets the players engaged in the fiction. Making it all just a roll removes this aspect. That's what I do not like about it.

According to this logic the Hickman written Dragonlance modules... widely considered a paradigmatic shift in the focus of RPGs that still holds sway to this day... are not railroads.

I think you need to rethink this statement.

I suspect they have more than myth in them, at least the way I mean myth. Like they are not just situations, they're plots. It is the plot that makes them railroads, the GM having to contort the play into a certain path.

 

[Crimson Longinus]

Is this something you've actually resolved in a game?

Not the exact situation, but something very similar several times.

In my experience, it is not easy - do the players, for instance, check whether there are hatches in the roof? A rear entrance from the canal? What if no one mentions the rear entrance when the players are doing their set-up, and then it occurs to the GM that there would be such a thing for the assassin to enter by?

I don't think the players need to individually check every possible method of entrance. Like they can say they check all entrances to the place and the GM tells them what they are. Also, if the players are proceeding with obvious false assumption, then the GM should tell them so that they can course correct. I don't find this to be hard at all. This to me seems like super basic stuff TBH.

This is also an example of what I mean by saying that the resolution method shifts focus away from what is core to the situation (eg Jackson and Bellow and their assassins, as @thefutilist described them upthread) to matters of architecture, and construction, and transportation (do these warehouses have upper entrances with blocks and tackles, like some 19th century warehouses?).

And I can report, from experience, that these are absolutely the sorts of things that I have found comes up in games that use map-and-key style resolution for these sorts of scenarios. The normal method of GM disclaiming of decision-making, in my experience, is to set odds and make rolls. Which, again, is not an example of extrapolating along causal pathways, but rather an alternative to that.

What is salient to the situation needs not to be just one thing. This is what I meant when said you probably understand "the situation" far more narrowly than I do, and I think @thefutilist alluded to in their post about different approaches of handling the situation; that in different games different things can be salient. D&D for example is the sort of game where locations, distances and placements of doors tend to matter, so they are part of resolving the situation.

 

[Manbearcat]

That's Crimsons (and my and Pedantic and Micah's) point though. And Abduls as well really.


Abdul said: However, that degree really rests heavily on a shared understanding of the principles, agenda, and practices of play! THIS is a key insight in the evolution of RPG practice.


In fact there's an interesting practical question that will come up in this sort of play.

The players have wounded Jackson (the imaginary guy from my last post) and then left to go and raid some dungeon or whatever. They're deep on the trail and have made camp (including set an alarm spell) and the GM realises that Jackson would have got one of his assassin friends to track them down and take them out (maybe Henrik from several posts ago).

So:

A) It is legit for the GM to create the assassin and place him in the situation now because that's an extension of what would have happened.

B) You missed your chance, it's not legit to decide Jackson hired Henrik and he's tracked them to the woods because we're in situ. You'll have to come up with a reason why it didn't happen. Maybe it did happen but it took a while for Jackson to contact Henrik so he'll be along later but not in this 'chunk' of situation.

Sorry I haven't been able to answer prior posts (and I'm still not doing it). Rather short on time presently and the amount of work required for me to engage with some of these points would be more than I'm willing to devote right now (and I have plenty in the past so some of this feels like something of a retread for me and therefore of waning interest). But this caught my eye and I want to engage with it:

* "...the GM realizes"

* "It is legit..."

* "It's not legit..."

Importantly here, I wonder aloud (suspecting that I know the answers, but I'd like to hear them confirmed and discuss them) the following:

What exactly are the governing parameters of this realization for the GM in which "Jackson would have got one of his assassin friends to track them down and take them out." I can think of lots of "Jackson-types" that would (a) not have had a hit put out on the group because they are too dangerous and "live-and-let-live" rules the day, (b) not had the means (time, coin, contacts with requisite facility in the matter, other logistical demands) to have a hit put out, (c) been preoccupied by other vendettas or occupation-hazards such that the hit didn't take priority, (d) any number of other things.

So I'm wondering (i) what superposition-altering parameters ensure one fictional Jackson emerges rather than another alternative? Then I'm wondering, (ii) what is the agenda that governs this? The (iii) language of "realizes" and "the question of legitimacy" leaves me thinking it is a Sim-Immersionist agenda whereby the apex priority of play is for the GM to attempt to internalize with themselves a "legit simulation of Jackson" and then attempt to render this "legit simulation of Jackson" onto an external, shared imagined space such that the players are somehow privy to the dynamics of the GM's "realization" and the GM's attendant conclusion of "legitimacy."

---

Now, some thoughts on the above that are relevant to the dynamics of play and how alternative models to the (what I suspect are) answers of (i) - (iii) above:

* B/X D&D has answers that systemitize "the great Jackson question" above. I think the best (and hopefully most common) answer would be that the GM populates the follow-on dungeon's Wandering Monster table with Jackson's hired assassin and, if that "hits," the NPC Reaction roll is eschewed (because we know what the NPC's motivation around the PCs is already). Now we have a combat or a guerilla cat-and-mouse game for the rest of the dungeon.

The alternative would be that when the GM generates the dungeon, one of the rooms is automatically stocked with Jackson's assassin. Should the PCs enter this room, then the encounter "goes off" and we handle the typical procedures for encounters. I don't like this one as much as the first, but it is another way to do it.

B/X is not a Sim-Immersionist game. It is squarely a challenge-based game (Gamism in Forge parlance). The machinery around "the great Jackson question" centers around generating an assassin-based threat which invigorates and enriches the challenge-based decision-scape and is embedded within the system architecture of the game. There is no "GM realizes" nor is there "legit/not legit" in play here. There is "there is an imagined world out there in which the great Jackson question comes up as yes, he would hire an assassin...let's do that because (a) the game can handle it and (b) that handling entails a rich, challenge-based decision-scape that is easily rendered onto play."

* Torchbearer (to harken back to the lead post) can handle this as B/X does and for the same reasons expressed in (a) and (b) above but also another reason; (c) the Story Now parameters of play (centering PCs as protagonists with an Enemy such as Jackson at character creation or an Enemy such as Jackson which emerges through play) which demand of the GM to invest the situation-state with PCs-as-protagonist trajectory.

Now Torchbearer can handle this in the following ways (none of which are an exact analogue to B/X's "Wandering Monster table hit", but some procedural DNA is shared):

1) The best way is as a Twist. When a failed Test hits, the GM brings in Jackson's (a PC Enemy) Assassin. Now we have a conflict of the GM's choosing. This would be my preferred way.

2) The Adventure's obstacles/map entails one situation with the assassin.

3) The assassin becomes a part of the Camp Events table.

Like B/X, this isn't about "GM realizes" or "legit/not legit." Again, "there is an imagined world out there in which the great Jackson question comes up as yes, he would hire an assassin...let's do that because (a) the game can handle it and (b) that handling entails a rich, challenge-based decision-scape that is easily rendered onto play, and (c) the Story Now parameters of play (centering PCs as protagonists with an Enemy such as Jackson at character creation or an Enemy such as Jackson which emerges through play) demand the GM to invest the situation-state with PCs-as-protagonist trajectory.

---

I have more to say on the matter that is related to my extreme skepticism over Sim-Immersionism's ability to either (1) generate a gameable decision-scape or (2) consistently center PCs as protagonists. My skepticism around these (2) is because GM-based setting extrapolations inexorably put extreme pressure upon play to be "real/legitimate" based upon the GM's, imo extremely fallible, causality-based extrapolations + sense of continuity and that formulation has a tendency to "backseat PC protagonism" as Sim-Immersionism is foregrounded. Further, because "table-facing play/machinery" is historically anathema to Sim-Immersionists, the GM's ability to articulate/render all of these offscreen matters (which are just GM internalizations) upon play in a way that players can readily recognize and index as they orient to situations and engage in their attendant decision-scape is paramount. The frequency with which I've seen this (live or on various fora) fail is extreme. This is because it embeds so many damn failure-points (from the mental modeling complexities to the generation of a functional and coherent User Interface for players to effectively engage with) onto the play paradigm.

So, imo, while the Sim-Immersionist approach can absolutely generate a preferred dynamic for certain folks who are predisposed to needing this paradigm to be immersed (and who place their particular brand of Immersionism as the paramount priority of play), I see it failing to reliably generate a gameable decision-scape for challenge-based play (for Gamists) and I see its particular formulation around "legitimacy" puts extreme pressure on PC protagonism (Narrativism where the premise of PC dramatic needs and relationships are centered).

 

[Crimson Longinus]

I have more to say on the matter that is related to my extreme skepticism over Sim-Immersionism's ability to either (1) generate a gameable decision-scape or (2) consistently center PCs as protagonists. My skepticism around these (2) is because GM-based setting extrapolations inexorably put extreme pressure upon play to be "real/legitimate" based upon the GM's, imo extremely fallible, causality-based extrapolations + sense of continuity and that formulation has a tendency to "backseat PC protagonism" as Sim-Immersionism is foregrounded. Further, because "table-facing play/machinery" is historically anathema to Sim-Immersionists, the GM's ability to articulate/render all of these offscreen matters (which are just GM internalizations) upon play in a way that players can readily recognize and index as they orient to situations and engage in their attendant decision-scape is paramount. The frequency with which I've seen this (live or on various fora) fail is extreme. This is because it embeds so many damn failure-points (from the mental modeling complexities to the generation of a functional and coherent User Interface for players to effectively engage with) onto the play paradigm.

So, imo, while the Sim-Immersionist approach can absolutely generate a preferred dynamic for certain folks who are predisposed to needing this paradigm to be immersed (and who place their particular brand of Immersionism as the paramount priority of play), I see it failing to reliably generate a gameable decision-scape for challenge-based play (for Gamists) and I see its particular formulation around "legitimacy" puts extreme pressure on PC protagonism (Narrativism where the premise of PC dramatic needs and relationships are centered).

What I would say, (and I don't know if this makes my play not proper sim-immersionist) is that I approacch "setup phase" and "play phase" differently. Like when I setup the initial situation I consider how it might engage the characters, I consider what sort of things would be cool and interesting, I consider what sort of themes the situation might evoke and all sort of other not-sim considerations. But once the ball actually gets rolling, once we start to play, then my resolution relies mostly on sim logic. I don't invent new significant elements, I don't "move things" behind the curtains etc. We have the initial situation, we introduce the PCs into the mix and see what happens. And sure, in resolution when several things might be plausible, the GM must make calls, and those might be influenced by non-sim concerns such as what's cool, what's thematic, what's relevant to the characters etc. That's fine by me.

 

[soviet]

Sorry I haven't been able to answer prior posts (and I'm still not doing it). Rather short on time presently and the amount of work required for me to engage with some of these points would be more than I'm willing to devote right now (and I have plenty in the past so some of this feels like something of a retread for me and therefore of waning interest). But this caught my eye and I want to engage with it:

* "...the GM realizes"

* "It is legit..."

* "It's not legit..."

Importantly here, I wonder aloud (suspecting that I know the answers, but I'd like to hear them confirmed and discuss them) the following:

What exactly are the governing parameters of this realization for the GM in which "Jackson would have got one of his assassin friends to track them down and take them out." I can think of lots of "Jackson-types" that would (a) not have had a hit put out on the group because they are too dangerous and "live-and-let-live" rules the day, (b) not had the means (time, coin, contacts with requisite facility in the matter, other logistical demands) to have a hit put out, (c) been preoccupied by other vendettas or occupation-hazards such that the hit didn't take priority, (d) any number of other things.

So I'm wondering (i) what superposition-altering parameters ensure one fictional Jackson emerges rather than another alternative? Then I'm wondering, (ii) what is the agenda that governs this? The (iii) language of "realizes" and "the question of legitimacy" leaves me thinking it is a Sim-Immersionist agenda whereby the apex priority of play is for the GM to attempt to internalize with themselves a "legit simulation of Jackson" and then attempt to render this "legit simulation of Jackson" onto an external, shared imagined space such that the players are somehow privy to the dynamics of the GM's "realization" and the GM's attendant conclusion of "legitimacy."

---

Now, some thoughts on the above that are relevant to the dynamics of play and how alternative models to the (what I suspect are) answers of (i) - (iii) above:

* B/X D&D has answers that systemitize "the great Jackson question" above. I think the best (and hopefully most common) answer would be that the GM populates the follow-on dungeon's Wandering Monster table with Jackson's hired assassin and, if that "hits," the NPC Reaction roll is eschewed (because we know what the NPC's motivation around the PCs is already). Now we have a combat or a guerilla cat-and-mouse game for the rest of the dungeon.

The alternative would be that when the GM generates the dungeon, one of the rooms is automatically stocked with Jackson's assassin. Should the PCs enter this room, then the encounter "goes off" and we handle the typical procedures for encounters. I don't like this one as much as the first, but it is another way to do it.

B/X is not a Sim-Immersionist game. It is squarely a challenge-based game (Gamism in Forge parlance). The machinery around "the great Jackson question" centers around generating an assassin-based threat which invigorates and enriches the challenge-based decision-scape and is embedded within the system architecture of the game. There is no "GM realizes" nor is there "legit/not legit" in play here. There is "there is an imagined world out there in which the great Jackson question comes up as yes, he would hire an assassin...let's do that because (a) the game can handle it and (b) that handling entails a rich, challenge-based decision-scape that is easily rendered onto play."

* Torchbearer (to harken back to the lead post) can handle this as B/X does and for the same reasons expressed in (a) and (b) above but also another reason; (c) the Story Now parameters of play (centering PCs as protagonists with an Enemy such as Jackson at character creation or an Enemy such as Jackson which emerges through play) which demand of the GM to invest the situation-state with PCs-as-protagonist trajectory.

Now Torchbearer can handle this in the following ways (none of which are an exact analogue to B/X's "Wandering Monster table hit", but some procedural DNA is shared):

1) The best way is as a Twist. When a failed Test hits, the GM brings in Jackson's (a PC Enemy) Assassin. Now we have a conflict of the GM's choosing. This would be my preferred way.

2) The Adventure's obstacles/map entails one situation with the assassin.

3) The assassin becomes a part of the Camp Events table.

Like B/X, this isn't about "GM realizes" or "legit/not legit." Again, "there is an imagined world out there in which the great Jackson question comes up as yes, he would hire an assassin...let's do that because (a) the game can handle it and (b) that handling entails a rich, challenge-based decision-scape that is easily rendered onto play, and (c) the Story Now parameters of play (centering PCs as protagonists with an Enemy such as Jackson at character creation or an Enemy such as Jackson which emerges through play) demand the GM to invest the situation-state with PCs-as-protagonist trajectory.

---

I have more to say on the matter that is related to my extreme skepticism over Sim-Immersionism's ability to either (1) generate a gameable decision-scape or (2) consistently center PCs as protagonists. My skepticism around these (2) is because GM-based setting extrapolations inexorably put extreme pressure upon play to be "real/legitimate" based upon the GM's, imo extremely fallible, causality-based extrapolations + sense of continuity and that formulation has a tendency to "backseat PC protagonism" as Sim-Immersionism is foregrounded. Further, because "table-facing play/machinery" is historically anathema to Sim-Immersionists, the GM's ability to articulate/render all of these offscreen matters (which are just GM internalizations) upon play in a way that players can readily recognize and index as they orient to situations and engage in their attendant decision-scape is paramount. The frequency with which I've seen this (live or on various fora) fail is extreme. This is because it embeds so many damn failure-points (from the mental modeling complexities to the generation of a functional and coherent User Interface for players to effectively engage with) onto the play paradigm.

So, imo, while the Sim-Immersionist approach can absolutely generate a preferred dynamic for certain folks who are predisposed to needing this paradigm to be immersed (and who place their particular brand of Immersionism as the paramount priority of play), I see it failing to reliably generate a gameable decision-scape for challenge-based play (for Gamists) and I see its particular formulation around "legitimacy" puts extreme pressure on PC protagonism (Narrativism where the premise of PC dramatic needs and relationships are centered).

Great post, a lot of interesting stuff to digest. Thank you.

 

[Manbearcat]

As regards Narrativism you'd have to lay out your case in a bit more depth for me to respond. Given I see group consensus as the death of play but unilateral fiat as its beating heart I suspect we're on totally different pages.

So a couple thoughts on this that I hope clarifies my position as it pertains to Narrativism.

* When we talk about unilateral fiat being the beating heart of Narrativism, I have the following questions:

Are we talking about GM having unilateral fiat over situation-framing or is their situation-framing constrained by overt and binding agenda and principles?​

​

Or are we also referring to GM having unilateral fiat over consequences of action resolution (put another way, "the GM is the system which mediates action resolution")?​

​

Are we talking about players having unilateral fiat over their character's dramatic needs/relations only at the outset of play with the collision of system + other participants' similar unilateral fiat over their own characters & responsibilities having their own system-directed say as play unfolds?​

* The answer to those three questions are paramount here. Because I would say the following:

1) GM's cannot have unchecked, unilateral fiat over situation-framing and play be functionally Narrativist. If systemitized and clear constraints aren't in play (which by definition should center premise and questions around protagonism), then the GM is afforded the right to drift play, anchoring it to interests alternative to premise and protagonism. Further, the lack of overtness in the (lacking) constraint puts enormous pressure on play due to its impacts on "the great bag-of-breadcrumbs paradigm" which robustly systemitized Story Now engines solve with intentional design. This is because that lack of overt constraint subverts (in part or in whole) the dynamics of (a) players understanding what is entailed within the bag of breadcrumbs because system tells them plainly and (b) the players being afforded at least equal rights over the bag of breadcrumbs (contravening the traditional model where GMs exclusively have rights over the breadcrumbs and keep some/all of what that bag entails as their own secret to reveal during play).

Now significant latitude in situation-framing within overt constraints? Yes. But that isn't the same as unilateral fiat.

2) The problem of unilateral fiat over consequences is similar to (1) above. Further, I would say that if there isn't a "robust system's say" when it comes to action resolution, then the GM is locked out from "playing to find out." Without being beholden to the dynamism-infusing, play-trajectory perturbing factor of well-systemitized procedures around action resolution and fallout, then the GM already knows and is therefore effectively engaged in a version of Story Before (even if muted in contrast with the alternative model of Hickmanesque metaplot and Golden Rule GM Storytime).

Further, if you combine (1) with (2), things only become amplified. GM having unilateral fiat over situation-framing + GM being the unilateral mediator of action resolution? That is defacto Story Before (and some kind of version of GM Storytime) even if it isn't prepped and prefabricated metaplot that is mainlined onto play.

3) Finally, that "only" in my last question does essential work. If play removes that constraint and players are expected to be afforded unilateral control over the conception and realization of character as play unfolds (with a GM obliging this conception and realization via framing & consequence control and/or with an intentionally limp/impotent system that ensures it), then players similarly aren't engaged in "play to find out." Like the GM in the above, "they already know." This is a form of OC/Neotrad play that is anathema to Story Now play.

---

Hopefully those questions are clear and my elaborations all makes sense (whether you agree with them or not)!

EDIT - @soviet , thanks! Glad those words worked for you! @Crimson Longinus , my sense of your play is that it is very much High Concept/Genre Simulationism (in Forge parlance...about halfway down on that link if you have any interest whatsoever...likely not I suspect!). OC/Neotrad play is a form of that sort of play, but not the only form. My sense is that does not at all describe your play and you are much more typical of the apex priorities being theme/genre simulationism meets some level of process simulationism (which is basically genre sim) with players having some rights to "anchor play" with a session 0.

 

[Hriston]

Great post, a lot of interesting stuff to digest. Thank you.

I agree. @Manbearcat, you are on fire today!

 

[Maxperson]

One day you'll have to point me to this Official RPG Lexicon of yours!

But anyway, I try to avoid jargon, and use language with its natural meaning as much as possible.

All you have to do is 1) Google it and you'll see how it's used, and 2) look at threads in all sorts of forums where pretty much everyone uses it that way.