D&D General Respeckt Mah Authoritah: Understanding High Trust and the Division of Authority

Snarf Zagyg

Notorious Liquefactionist
It's been a hot minute, eh?

Over on another thread, people have started talking about a perennial topic that tends to rear its ugly head here on the EnWorld forum- who sucks more, the BAD DM or the BAD PLAYER? And while the contours of that debate are interesting (for certain low values of "interesting"), I thought it might be interesting to look at the conversation from a slightly different point of view- to discuss what people mean when they talk about "high trust," and, more importantly, why different game designs have different divisions of authority- and why those different designs have inherent advantages and disadvantages. It is only from there that we can fully understand why D&D, traditionally, has put such a strong emphasis on the DM's (GM, referee, etc.) role within the game.

1. What is High Trust, and Why do Some People Keep Yammering On About It?
There are two motives for reading a book; one, that you enjoy it; the other, that you can boast about it.
To understand a term, it's helpful to understand the context in which the term arose. Mostly because people will often throw around terms like frisbees ... or manhole covers. So, why did people start using this term, and what was it supposed to mean?

As far as I can tell, "high trust" first became a term utilized in the OSR movement in the early 2000s. It was likely borrowed from a similar usage in the corporate world at the time, which referred to companies and workplace cultures as either "high trust" or "low trust." At this point, we should all be thankful that (1) they didn't decide to use the terms "synergy" or "leverage your core competencies," and that (2) OSR didn't arise a decade later, in which case we would all be trying to pivot to video. Ahem. Anyway, in the corporate world, these terms refer to how a company relates to its employees; a high-trust organization is one that has employee empowerment, management oversight is not obtrusive, and employees have the opportunity to independently solve problems. Low-trust organizations, on the other hand, monitor employees closely, do not provide employees the agency (oh boy!) to solve problems on their own, and the organization will often have detailed rules that the employee must follow instead of using their best judgment.

And so you can see why borrowing the term has appeal. OSR was not just a movement for a certain style of gaming; it was also a critique of the direction D&D had taken with 3e and 3.5e. As an aside, the reason that a lot of jargon in TTRPGs can be contentious is that the jargon itself usually arises out of a specific movement that is trying to elevate one style of gaming while critiquing others; this is why you will often see endless debates over terms like "player agency" or "high trust." It's because the terms themselves arose not out of a neutral analysis, but from a critique of a mode of play.

Now the thing to remember with 3e (which I am using as a synonym for 3e, 3.5e, PF, etc.) is that while is a "trad" game, it was also a trad game that a subset of trad players did not like. One of the main reasons behind that was that while 3e (in the eyes of someone used to a narrative game) still had a lot of DM authority, it deliberately set out to make a game that was more rules-bound in terms of DM discretion. For fans of the older TSR-style games, this made 3e a "low-trust" game. Moreover, to the OSR movement, the lack of trust extended to players as well- they were no longer trusted to devise any solution that they could think of, instead depending on certain approaches that would be prescribed by the rules (or the abilities demarcated on the character sheet).

More to the point, proponents of high-trust gaming believed that players should be trusted to use their own creativity to choose what to do and how to do it, and that the DM should be trusted to react appropriately to the player's ideas. This concept of bi-directional trust animated the OSR, and the later FKR, movements.

The reason that people keep talking about "high trust," then, is because the concept is what animates a subsection of the gaming population. Moreover, we can see that this concept is at least partly incorporated into 5e in the concept of "rulings, not rules."

2. Dividing Authority at the Gaming Table
I hate writing, I love having written.
One of the primary design decisions in games is a simple one- who has final authority? If you have only played D&D, then the answer might seem obvious- the DM! But it doesn't have to be that way. There are games that can be played with any central authority (Fiasco, for example). There are games that can be played that allow for significant player authority over the fiction (any one of a number of games, such as the PbTA family). Even 5e has an optional rule (in the DMG, that no one has ever read) that allows characters to have "plot points" that would allow them to author the fiction. Roughly 80% of the threads that go on for more than 50 pages on EnWorld (numbers are approximate, yet completely true) end up being about the division of authority, but it also is a lot less of a big deal than most people think it is. Here, for example, is the description I like to quote from the Rules Lite version of Cthulhu Dark-

Who decides when to roll Insanity? Who decides when it’s interesting to know how well you do something? Who decides when something disturbs your PC? Who decides whether you might fail? Decide the answers with your group. Make reasonable assumptions. For example, some groups will let the Keeper decide everything. Others will share the decisions. These rules are designed to play prewritten scenarios, run by a Keeper. If you try improvising scenarios or playing without a Keeper, let me know.
Who decides? Who cares! Do what works for you. And yet ... it does matter. Just not for the typical reasons that usually get argued about.

Every table is different- a different mix of personalities, a different mix of people. When we discuss D&D, one of the topics that often comes up is that of a DM shortage, or of DM burnout. And this points us to both the great advantage and the great disadvantage of those games that vest more authority in the DM. On the plus side, you only need one player to be the DM. You only need one person that really knows the rules. You only need one person to have a higher level of investment and to "bring it" every gaming night. But that strength is also a weakness- that is a lot to have riding on one person. If that person is ... the BAD DM, then that will be a catastrophic failure. And putting that burden on the same person on a regular basis can lead to burnout.

The flip side of that is a model with distributed authority. The best thing about distributed authority (especially when you're playing a Story Now / Fiction First game) is that you don't have all of this prep time foisted on one person. Everyone is invested in the game. Everyone is creating fiction. But this isn't all pop rocks & soda- the downside is that you really do need everyone to be invested in the game. You need them to "bring it" on a regular basis, and be ready to create their fiction. And not everyone is prepared to do that. So while providing the GM more authority provides one point of failure (albeit a catastrophic one), providing more narrative authority to everyone can end up allowing for multiple points of failure.

Which brings us back to why D&D continues to use a DM-centric model....

3. Why D&D Will Continue to be DM-Centric
I don’t care what is written about me so long as it isn’t true.
"I don't get why more people don't listen to opera; it's so much better than that hippity-hop music that they're listening to!" D&D is always designing for the mass market. For the maximum number of gamers. For as many people to play as possible. Because they want to continue to be the 800lb gorilla in the room. And, in my opinion, this desire naturally leads to game that will be designed with a DM as the final arbiter of fiction. And I say this not because it's necessarily the best way to do things, and not to excuse the lack of DM support in 5e, but simply because it makes the most sense in terms of designing for the market.

We've all had groups like this-
Amy loves to optimize and get into combat.
Brad loves to roleplay, and is playing a bard again.
Chad is mainly there because he's married to Amy and likes to hit stuff and not read things, so, um, Champion?
Derek likes to drink. MAXIMUM DEREK!
Emma knows all the rules, and believes gaming night is a good excuse to argue those rules.
Fiona likes game night, but she's more into drawing the maps and writing down the treasure. It's your turn Fiona. Fiona? Fiona? Fiona?

We all have friends like this. We all have mixed gaming tables at some time or another. People who are playing for different reasons, and with different expectations- but they want to have fun. Sure, maybe you could get them to play a one shot of Monster of the Week, maybe. But for the most part, they aren't looking to author fiction and "bring it" every night (except DEREK, who thinks "bringing it" means going through your liquor cabinet). And D&D, by placing the final narrative authority in a single person, allows these tables to function. Allows them to have a good time.

It's not a perfect solution for everyone, but it's a good solution for a lot of people. When you move away from the theoretical to the practical, it becomes obvious why D&D continues to use the DM-centric method when it comes to the division of authority. Because there's a lot of very casual players out there, and that's a much bigger market than TTRPG afficonados.

Anyway, it's been a month. Thought I'd post something. Enjoy, and tell me why I'm completely wrong in the comments. :)

log in or register to remove this ad

Micah Sweet

Level Up & OSR Enthusiast
I have to say the same. DM authority has a long precedent of being the best method for the most people. Not for all people, of course, but more often than not.


I haven't played D&D in a looong time, but by the definition given above, I prefer "low-trust" game design, but perhaps for different reasons than most.

I am a software engineer, and for the majority of my career, have worked with and contributed to Open Source software. Open source seems like the epitome of meritocracy or democracy, but it isn't. That is why you see a lot of forks of software, because someone disagrees with the direction that the project is going and forks a new project based on the old. Some will tell you that is horrible for an open source project, because it divides the community and splits the amount of engineering resources that can contribute. And yet, forking is the ultimate in freedom and "agency". It basically says, "I think I can do this better, so let the best code and maintainer style win".

What I learned, IMHO, is that the best run open source communities have what we call BDFL's or "Benevolent Dictators for Life". For example, Guido van Rossum (who relinquished his BDFL title not that long ago) of the python programming language, or Linus Torvalds for the linux kernel. Other projects which had no central authority, such as a council, steering committee, corporate sponsors or some other diffusion of power ultimately devolved into political infighting. Meritocracy, or what should have been the best code, no longer always won and got committed. It was "which group has the most influence" to get their code committed into the project. I feel that games which allow players to "direct" the narrative, rules, or other game decisions ultimately becomes like these "design by committees". They look and feel fair on the outside, but I think it can devolve into politics. Also, design by committee lacks a centralized "vision". While some artists and engineers do create great works while working collaboratively, often, there needs to be a single "source of truth" or vision as to where things are going.

As I learned more about programming language theory and learned new languages, I came across, at the time, the near religious war between the dynamic type proponents, and the static type proponents. For those who don't know, a dynamic language is one where some variable's type can change (eg, in one part of the code, it can be a string, and in another part of the code it can be a number). More importantly in dynamic languages, in the code that you wrote, you don't have to declare its type ahead of time. The dynamic proponents argued that:
  • there exist valid programs that can't be typed (true, but in practice this is almost never encountered..static languages are still Turing or Church Complete)
  • it stifled creativity by forcing a programmer to fix the type
As type theory became more advanced and graduated from academic languages like Haskell or Idris to more mainstream languages like typescript, rust or even python, some of the arguments that the dynamic proponents used lost their power. For example, type inference removed the need (in most cases) to specify the type of a variable in the code because the compiler could figure it out. But more importantly, companies and engineers realized that as their code base grew in size and complexity, the "freedom" to dynamically change types was too costly. Refactoring code became a nightmare for dynamic types, and performance was also lost (because the compiler interpreter couldn't make optimizations knowing that a type would always take up a fixed amount of memory for example). To put it another way, the "creative freedom" of the individual developer had to come second to the group as a whole.

I would say that dynamic type proponents lost the war. More typescript is written than javascript (the former is the statically typed version of the latter) and even most python code has type annotations now. The last popular languages I can think of that were mostly dynamic were Elixir and Julia, both coming out in the early 10's. However, both have support for type annotations too. So "creativity and freedom of expression" lost out due to the problems they posed.

I also came across a quote by Mokokoma Mokhonoana while learning Clojure (a lisp dialect)
A blank cheque kills creativity
If everyone gets a say, it become "design by committee". If everyone is allowed to influence the outcome (at a metagame level. not through their actions while playing), then I think it becomes a kind of "blank cheque". The irony is that everyone having a say in the direction, narrative, rulings, etc of a game actually can have an opposite effect. The same I feel is true even of "rules-lite" systems where much of the arbitration and adjudication is done by whomever is in charge (hi or low trust). The rules themselves should help so that less arbitration and adjudication is even needed in the first place. The rules act as constraints, and I am a big believer in constraints.

To go with a programming analogy again, you can have something called a Generic. It's like a placeholder for a type. Maybe the Generic will become a String, or an Integer. But here's the catch: if a type can be anything, there's really nothing you can do with it except to contain it. If I don't constrain what the type can do, how do I know what can be done with or to that type? For example, I can't add two generics together, because they may not support the "+" or "add" operators. In Programming Language Theory, this is usually called a bound or a constraint to specify that some type supports some set of operations.

The rules of a game system can act as a kind of constraint or bounds. Depending on the rules, they specify not just how something is done, they can also limit what can be done. They take away the blank cheque. And I am fine with that, either as a GM or a player. It's the limitations imposed on me as a player that actually can give more meaning. If everyone is special and has a say, then no one really is special. Of course the game rules can't cover every possibility, and hence the need for human intervention. But once it comes to that, I would prefer a ruling by one individual. If one asks "where is your agency if you give it all to the GM/DM?", I would say my agency lies in the fact that I can choose another GM, solo roleplay, GM myself, or not play at all. That's it...I don't get any special rules.

As an engineer, I deal with reality. I have to work within its confines. I don't get to cajole, manipulate, coerce, suggest, or otherwise
get reality to conform to my liking. So I am okay with low-trust, limited agency games. I realize that is not to everyone's tastes, and that's just my personal preference.

EDIT: Removed lots of superfluous parenthetical comments, and fixed some grammatical issues. I'm an Engineer, not an English Major Jim!
Last edited:

Thomas Shey

Of course I have to note that one of the consequences of the DM-centric design (and I don't really disagree with Snarf why its liable to be the dominant paradigm in not only D&D but most trad designs indefinitely) is that there's a big power imbalance there, and when that happens you're inevitably going to have a lot of people wandering through the hobby who are have various sorts of issues related to it. That's just going to be a reality, and outside the people who will pull the plug if they don't have a group lacking those issues (which in some areas adds up to "bail out of gaming"), everyone else has to deal with that as best they can.

Voidrunner's Codex

Remove ads