NPC Deception/Persuasion and player agency

[Crimson Longinus]

From Gygax's DMG (pp 80-1):

Someone once sharply criticized the concept of the saving throw as ridiculous. Could a man chained to a rock, they asked, save himself from the blast of a red dragon's breath? Why not?, I replied. If you accept fire-breathing dragons, why doubt the chance to reduce the damage sustained from such a creature's attack? Imagine that the figure, at the last moment, of course, manages to drop beneath the licking flames, or finds a crevice in which to shield his or her body, or succeeds in finding a way to be free of the fetters. Why not? The mechanics of combat or the details of the injury caused by some horrible weapon are not the key to heroic fantasy and adventure games. It is the character, how he or she becomes involved in the combat, how he or she somehow escapes - or fails to escape - the mortal threat which is important to the enjoyment and longevity of the game.​

Clearly Gygax doesn't agree that fortune in the middle - that is, establishing a result via a roll (that appropriately reflects the fiction) and then narrating the details in response (within whatever appropriate parameters are set by the fiction and the resolution process) - is not roleplaying.

And as far as resolving haggling via a simple die roll, this is a feature of Classic Traveller (1977) - as a component of the speculative trade rules found in Book 2 - and of Rolemaster (198x - I'm not sure what is the earliest date of publication of the commerce rules found in Character Law and Campaign Law).

To me, it seems clear that the various approaches that @andreszarta has identified are hardly foreign to the hobby, or recent innovations. They are as old as the hobby itself.

So? What that has to do with anything?

 

[CreamCloud0]

So do you in the real life (outside the internet) get stuck in endless arguments that go nowhere and never end and you just keep doing it and doing it because there are no dice or GM fiat to settle it? Like if you and your partner cannot agree on what to make for dinner do you just continue arguing about it until you starve to death?

well, ironically for your point i'd say this very thread probably qualifies quite a bit for that description, (i don't see why not being on the internet really factors into it,) you're not backing down, we're not backing down, the conversation is going in endless circles...

 

[Crimson Longinus]

well, ironically for your point i'd say this very thread probably qualifies quite a bit for that description, you're not backing down, we're not backing down, the conversation is going in endless circles...

Indeed. But that is due the nature of the internet discussion. We can check this thread from time to time, drop a post, and continue to do other things. Continuing the debate doesn't actually make our lives to "get stuck." We could imagine an analogous situation in a game where a two characters had an ongoing disagreement they would bring up now and then, but nevertheless would continue to do other stuff.

Also, if we were to roll the dice for this, would people who disagree with me really feel matter is settled if I rolled higher than them? Would they now be able to internalise this new position as their own? I don't think so. So same with the games. If the situation described doesn't make player the feel that the argument could convince their character, that their character would experience a certain emotion etc, then the dice saying it is so doesn't make them feel that either. It will feel false, it will feel wrong. I hate that.

 

[Aldarc]

Is it? I mean, personally I agree it all gets a bit much, but AFAIK (and I'm not a big expert, but I have played a good bit of 5e) there is no tight association of skills with abilities. Any time you specify a skill check, you MUST perforce identify which ability modifier will apply to that specific check, and this is a GM decision. So, it would follow that one must use the terminology Charisma (Persuasion) or something similar, granting that if the ability score is not explicitly stated you'd assume the most natural one.

Contrast this with 4e, in which the skills are wedded explicitly to specific ability scores. Every Athletics check uses the character's STR modifier, but in 5e it could use CON, DEX, INT, pretty much anything. In the end I don't think it adds a ton of value, but that's still how it works by RAW.

Yes, I still think that it's mostly splitting hairs about parole. If you want to dig up a post that far back just so you can engage in the futile exercise of splitting hairs about this sort of thing, then knock yourself out, but it's not something that I have any interested in engaging further with.

 

[pemerton]

Outside of the internet people will not have endless unproductive discussions. Someone will just leave.

So do you in the real life (outside the internet) get stuck in endless arguments that go nowhere and never end and you just keep doing it and doing it because there are no dice or GM fiat to settle it? Like if you and your partner cannot agree on what to make for dinner do you just continue arguing about it until you starve to death?

In the context of the play of the game, it may also be true that someone will just yield. But does that reflect the authenticity of character? Or rather does it reflect the social dynamics of the play of the game? I agree with @andreszarta that it typically will be the latter:

I believe, and maybe I'm wrong, but I believe that if indeed this is the way you guys play, then what is happening is that you are selectively choosing when to give X and when to give Y.

When the PC works hard for it, or you are bored and you don't see a way out, you give up on what the NPC wants and justify in its internal logic, retroactively, why they gave up.

When you are not ready to budge and the outcome matters to you as much as to your NPCs, you are willing to compromise on "fairness" and be a bit more liberal in your interpretation of what's internally logical for that character.

Or something along those lines.

If the situation described doesn't make player the feel that the argument could convince their character, that their character would experience a certain emotion etc, then the dice saying it is so doesn't make them feel that either. It will feel false, it will feel wrong. I hate that.

This assumes that the player is not able to actually accept the established fiction. In principle, it can be equally true of combat - my implacable guy would never be bested by a mere Orc - but most RPGers know that they don't get to prioritise their feeling about their character over the dictates of combat dice.

@kenada, in my view, dealt with this pretty early in the thread:

It seems to me that if an NPC succeeds at check to deceive a PC, then the player of that PC should roleplay as if they believe the lie until circumstances change to expose it.

I’m separating what the player thinks from what the character thinks. Otherwise, it muddles (or limits) resolution of certain situations (like the one posed in the OP). Obviously, I would prefer a game that does a better job of handling these interactions, but that’s not what’s being asked.


If the change in game state (due to a check, etc) would result in the character’s doing something stupid¹, I see no reason not to do the stupid thing. I think that makes play more interesting.

Generalising these points - one part of the skillset of playing a PC in a RPG can include being able to portray the character as the game requires them to be portrayed.

 

[thefutilist]

In practice, I tend to agree with @andreszarta about what it is that typically brings things to a close:
I think this sort of account of what is happening also - often, at least - generalises to PC vs PC rather than PC vs NPC. One of the players compromises or folds not because that is what a true conception of their PC demands, but because of the passage of time, or the fact that they are bored, or that they can see that the other player is more invested, etc.

There is some kind of fundamental difference in how we parse fiction. If not in general exegetic terms then certainly as it applies to rpg's. It's not that I don't have the problems you, Luke Crane and Andre have, it's that I can't even conceive of how you get those problems. So I suspect it comes more from the social layer.

At a guess it's like we diverged from trad games but took two totally different paths. Luke, Andre and you saw issues with the rules and authority and so your approach to system nails them down. My issues were more with a kind of expressive coherence, how are we mutually parsing what's happening.

I think it's impossible to gain any further understanding because forums turn into pissing contests and we'd need to actually compare video to see what was happening. Even then we probably just value really different things.

Anyway it's been an interesting discussion, I've had a lot to think about and it might have made me more open minded towards your and Andre's playstyle.

 

[Oryzarius]

My point is that not only you can never achieve a perfect interpretation of a character—because who could even determine what that is? But more importantly, even when playing a character to the best of your ability, you can’t simultaneously be invested in both getting it right and advocating for their outcomes. This becomes especially problematic when the desired outcome for the other player depends on figuring out the very thing you’re supposed to get right in order to achieve what they want. It’s circular.

This statement is really puzzling to me, because it seems to assume that there's some sort of Platonic Ideal for each NPC in a published adventure. But that's not true. Just as there is no definitive stage production of Hamlet for all of time and space, there is no singular play-throught of Curse of Strahd for all players in all sessions everywhere. A piece of theatre, or a role-playing adventure, exists briefly and specifically... and differently for each of its instances.

 

[deleuzian_kernel]

This statement is really puzzling to me, because it seems to assume that there's some sort of Platonic Ideal for each NPC in a published adventure. But that's not true. Just as there is no definitive stage production of Hamlet for all of time and space, there is no singular play-throught of Curse of Strahd for all players in all sessions everywhere. A piece of theatre, or a role-playing adventure, exists briefly and specifically... and differently for each of its instances.

I hope you can see that I completely agree with you?

My statement comes at the tail of a series of exchanges that started here: NPC Deception/Persuasion and player agency, where it was kind of implied that the person responsible for making choices for the NPCs would only need use common sense and extrapolate from a personality baseline to instantly know the right answer; the authentic answer, the coherent answer. I argued that this is not what happens at all.

There isn't such a Platonic ideal, I agree, which stands to prove that when it comes to the logical extrapolation of in-character decisions from a personality baseline, the interpreter inevitably introduces bias—not in the sense of distorting some pure ideal, but in the sense that every interpretation is shaped by the priorities, understanding, and framing of the person making it. A GM making an in-the-moment call about an NPC’s reaction is not channeling some objective, preordained truth but making a creative decision, whether consciously or not.

This doesn’t mean there’s no coherence, but it does mean that coherence is something actively maintained, not something that passively exists in a vacuum. And that maintenance is itself an act of interpretation.

 

[deleuzian_kernel]

I think I'm gonna close up my intervention on this thread (unless someone wants to keep chasing something I have said that they feel might be an incorrect reading or interpretation) with the following:

The response of Narrativism-oriented design to the problem I picked up back at NPC Deception/Persuasion and player agency, has been that, instead of trying to simultaneously be invested in getting it right while also advocating for a character's preferred outcomes, we recognize that getting it right is inherently an act of creative interpretation and let that fall naturally into place.

We dispense some notions of objective truths and instead we go all-in on advocating for the NPC’s preferred outcomes, allowing the game’s instrumentation to guide our incorporation of consequences in alignment with our creative understanding of the characters evolving motivations and desires. And it totally works, and is coherent and grounded when we do it right!

---

The opposite approach, seems to say: let's go all-in on getting it right, prioritizing our own subjective interpretation of the character’s internal logic and completely dispense of advocating for their preferred outcomes. This means treating the NPC as an autonomous entity and focusing on faithfully portraying their motivations and decisions rather than pushing for what they want to achieve in the fiction.

And as I said before, this brings us back to resolving conflicts of interest in one of three ways:

1. Pure puzzle-solving, where the challenge is to figure out the NPC’s personality keys and push the right buttons—without any real active opposition, just a mechanical or fiat time limit ("Patience") or other constraints.
2. Role-playing it out, where we talk until one of us decides we’ve had enough and concedes, or until someone exerts situational authority to force an outcome—which could very well be leaving the conflict unresolved, subtly implying that the initiator doesn’t actually get what they wanted.
3. GM as the arbiter of aesthetic and creative judgment, where the GM ultimately decides whether the PC’s approach satisfies their personal sense of what feels right for the fiction and rules accordingly. The problem here is that it effectively makes the GM the final authority on what is convincing or earned in the fiction, rather than letting the process of play and its instrumentation carry that weight. This risks creating an implicit approval/disapproval dynamic, where the players’ ability to succeed depends not just on the internal logic of the game world but on the GM’s taste and narrative instincts, which are inherently subjective, SPECIALLY so in the context of NPC Deception/Persuasion.

I’m open to hearing other alternatives, but I think these are the natural results of that approach. Certainly the ones I'm familiar with having had personal experience with all three.

Are any of these inherently not fun or wrong? No, I’m not prepared to argue that.

Looking at number three, perhaps the most controversial one: Is it wrong for players to enjoy letting the GM act as an arbiter of what feels right for the fiction, if the game ends up being fun and everyone is having a good time?

Like, who am I to say that that's wrong? But I do think it’s worth acknowledging the trade-offs involved.

When the GM takes on the role of arbiter, the flow of play becomes subtly dependent on their personal sense of narrative satisfaction. This isn’t necessarily bad, but it does mean that what “feels right” is ultimately filtered through one person’s instincts rather than emerging from the shared structures of play. This can lead to moments where players find themselves intuiting or negotiating what the GM wants more than engaging directly with the fiction on its own terms.

More importantly, in the specific context of NPC deception and persuasion, this approach risks making social interactions feel less like a structured game space and more like an appeal to the GM’s sensibilities—what they find convincing, what they think the NPC would buy, what they deem a sufficiently clever or compelling argument. At worst, this can blur the line between character logic and GM fiat, making it unclear whether success was due to in-fiction dynamics or simply because the GM found it aesthetically satisfying.

So while I wouldn’t call it wrong, I do think it’s fair to ask: is that really the kind of game you want to be playing? And, more importantly, does everyone at the table know that this is what’s happening? Because if they don’t, that’s where real problems can start to emerge.

 

[Sepulchrave II]

Generalising these points - one part of the skillset of playing a PC in a RPG can include being able to portray the character as the game requires them to be portrayed.

I think this sums it up very succinctly.

It seems to me that there are many, many instances where a character's agency in the game is reduced or denied, and the player should be able to represent that faithfully, without any kind of forceful intervention by the GM: fear effects, enchantments, illusions, intoxication, whatever. It does seem odd to me that a social interaction which is skill-governed should deserve special consideration; I still don't see why.

Given that - in the real world - it is the social interaction which we would encounter (barring intoxication, I suppose), I can only state that I have encountered situations where I have been goaded into precipitous action against my better judgement, and I have backed down in situations where I felt intimidated, I have been deceived etc. No magic was involved, and it seems only reasonable that any character which I might portray in an RPG be subject to the same emotional and situational pressures.