RAW and Schelling points

Today I’d like to give some unique perspective on a pretty common conflict in PNP RPG spaces - the problem of Rules As Written (RAW), Rules Lawyers, Rule Zero and how they relate to shelling points. I haven’t seen this perspective elsewhere, so I hope this will prove instructive.

As usual, if you are confused about any terms, make sure to check out the glossary, and if you are looking for other posts, check out the blog map.

Let’s start with the basics. What are “Rules as Written”?

Most PNP RPGs out there have very large sets of rules for how the game is played - rules about conflict resolution, how objects are damaged, combat, survival in the wilderness, and all sorts of other things. These rules are described in various rule books, which come in various levels of clarity and internal consistency.

GM and the players, when playing the game, use these rules to guide their decision making and conflict resolution. Of course, rules are written by human writers, and as such have multiple limitations. First of all, they only cover some subset of situations that may come up over the course of the game - no system can encompass everything. Furthermore, sometimes the rules are written in ways that make it unclear what they actually say - much like the regular laws, it is difficult to write a text that accepts no two interpretations.

Finally, sometimes the literal reading of the rules results in an outcome that doesn’t conform to common sense or the in-game world. This happens especially often when multiple parts of a large ruleset interact.

For example, in DnD, your characters may carry a very large shield, called a tower shield. This shield blocks what is called Line of Sight: that is to say, you are covered by the shield to such an extent that any arrows or spells have to go through the shield before reaching your body. This is a perfectly sensible way for this shield to work.

In another part of the rules, we have rules for stealth. One key rule for stealth is that if an enemy lacks Line of Sight to your body, they also can’t see any of your equipment. This is likewise a sensible rule: if you are hidden behind a wall, the enemy shouldn’t be able to see the sword you are holding in your hands. Rule covers all equipment in order to prevent arguments over how large of a backpack you are carrying, and wherever anything pokes out of cover (seeing how generally DnD leaves the exact geometry of a lot of things up to interpretation, these would be very hard arguments to arbitrate). It also gives players some more freedom in how their characters may look without sacrificing effectiveness: otherwise, all characters would try to minimize their silhouette whenever possible, which would lead to fairly bland appearance.

Of course, a canny reader may see an unfortunate consequence when these two rules collide. When you are hidden behind a tower shield, enemies lack Line of Sight to your body, meaning they also lack Line of Sight to all your equipment. But the shield you are hiding behind is part of your equipment: meaning that enemies also lack Line of Sight to it. Therefore, if you hide behind a tower shield, you are completely invisible.

This is a problem, because within the logic of the fantasy world, this makes little sense.

This is where we involve another key concept: Rule Zero. Some PNP RPGs write it explicitly in the text of the rules, while others leave it implicit. Usually, it is some variant on “The Game Master is always right”. This means that GM is in principle free to override any rules in the rulebook, and has the last word in any dispute of interpretation. To draw an analogy with real laws, in case of a dispute of interpretation of laws courts decide which interpretation is correct. Final court is a supreme court, whose decisions are typically inarguable: in PNP RPGs, GM is the supreme court.

So in our case of the shield, GM may say “This may be what the rules say, but I think the sensible ruling here would be to say that tower shield is an exception to the general rules about all equipment being hidden”. When this happens, we call the literal reading of the rules “Rules As Written” or RAW, and the ruling GM makes “Rules As Intended” or RAI - intended by the writers of the rules, presumably.

Rule Zero isn’t just used for dispute resolution when rules are unclear or nonsensical: it is also implicitly used when neither the GM nor the players remember the actual rules, and neither of them wants to look them up (e.g. because it will take too long). Then GM may simply make a ruling on the spot to handle some situation, and the game goes on. If this ruling is later found to be in contradiction with official rules, then it’s obvious GM exercised Rule Zero.

The final piece of the puzzle is Rules Lawyers. These are players who know the rules well, and typically object to overruling the RAW reading of the rules, or simply are ready to mention what the literal reading of the rules says in any particular situation. This tendency sometimes leads to conflict with GMs - for example, GM may be trying to describe a particular event that makes sense to them, but goes contrary to the literal reading of the rules. Rules Lawyer mentioning that this event is impossible may interrupt the pace of the game, and make the GM unhappy. Likewise, if some other player was trying to do some cool thing that is impossible according to RAW, Rules Lawyer mentioning this may make that player unhappy.

But why would anyone do this, in principle? How does this fit into the utility theory of PNP RPG play?

Of course, sometimes people bring up RAW as just another argument when they dislike some ruling GM made. But oftentimes, I think there is a broader pattern going on. I pose it to you that RAW is a very convenient shelling point for resolving disputes, and rules lawyers simply have a much higher attachment to this shelling point than other players.

What is a shelling point, I hear you ask? Well, let me explain.

Imagine that you are on a reality TV show. The host releases you on a random street in New York and tells you that you can win ten thousand dollars if you find the other contestant of the show in the city. That other contestant gets the same proposal: they can win ten thousand dollars by finding you. The only problem is that you don’t know who one another is - the other contestant can be a man or a woman, short or tall, have any kind of hair color, and so on. You have no information about their looks.

New York has a population of 8.4 million as the time of this writing: obviously, the chances of finding one another by chance are very low.

And yet when this test is done in reality, the majority of people pick the same location: Grand Central Train Station, information booth, at noon. This is what is called a Schelling point: solution that people can cooperate on without communicating with one another. 

For a shelling point to exist, it has to be in some sense “obvious” from the question itself. For example, if people didn’t know that New York has a central train station, or even that trains exist as a concept, this kind of cooperation would not be feasible.

But what does this have to do with PNP RPGs? After all, the game master is right there - communicate at leisure.

There are two reasons why communication is not quite possible here. First reason is that sometimes the very act of communication is annoying. Imagine that your player group is trying to pull off a bank heist in game. You are a player who is planning how to go about doing this - looking at the map of the bank GM gave you, resources you have access to, and so on. As you think on the problem, you juggle a dozen parts of this plan - how to approach the bank, how to break in, how to get away, where to pawn the money, etc, etc. Naturally, as you think of various problems, you patch them - this involves very rapid prototyping, all inside your own head.

And this is where a shelling point of RAW comes in handy: you can consider rules for how objects are broken, how much damage various weapons deal, speed of various vehicles, and so on without consulting the GM, because talking to the GM is slow, way slower than thinking. If you had to cross-check every rule and number you will be relying on with the GM, planning would take ten times as long, and thus would be 1/10th as enjoyable. If you didn’t cross-check every rule with them, and the GM didn’t stick to the shelling point much, then there would be a soft limit on the complexity of plans you can possibly attempt, as the chance of GM overruling some crucial step increases in geometric progression.

Second reason why communication is not ideal and why you may have to rely on a shelling point is that you can’t be sure what decision algorithm the GM is operating under. In this case RAW takes the form of a Schelling fence - a special kind of shelling point when complicated decision problems are involved.

Imagine that you expect GM may be somewhat biased in favor of one of the players, and somewhat biased against you. It doesn’t matter why you may expect that, only that you do. Furthermore, imagine that you lose a lot of utility from being treated unfairly - it is one of your unfun types.

In this case, communication with the GM doesn’t actually help, because you do not expect them to cooperate honestly. However, if you stick very strongly to the shelling point of RAW, then it “doesn’t matter” if GM is biased: they would be stuck applying the same rules to everyone, resulting in a “fair” playing field.

(Of course, this is an illusion: GM has many ways to be unfair that do not touch the rules at all. But I think some people still derive some sense of safety from at least the literal rules being fair.)

Mind you, I don’t think GM has to do anything specific for the player to not trust them. Some people are simply fairly paranoid about digressions from such social shelling fences, for one reason or another - such is their personality, or perhaps background.

Furthermore, GM doesn’t necessarily have to be biased against the player. For some people, being biased in favor of them will be seen as a transgression - for example, they may be less satisfied with overcoming various challenges if they think GM may rule in their favor at a critical point.

I hope this gives people an interesting perspective on some of these common elements of the PNP RPG scene. Maybe next time you’d meet a rules lawyer, you’d know better what may be motivating them.

Meanwhile, if you enjoy what I write, you can subscribe to receive updates by email: