Reductionist Magic 1: The Problem
It’s hard to deny that magic is foundational to the entire genre of TTRPGs. First commercially available roleplaying game - Dungeons & Dragons - very heavily revolved around magic. Current trends in what TTRPGs are popular, as seen by this roll20 industry report, confirm this. All of the most popular systems involve magic as a central element, and it’s not until you get into fractional percentage points that you see ones where it may be rare. It is therefore unfortunate that none of these systems have any idea how magic works.
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.
What do I mean by this? Don’t all these systems have mechanical rules for how magic works? How can they not know how it works while telling you the rules for it?
Well, this depends on what you mean by knowing how something works. All those systems have rules for how magic interacts with other systems of the game - health, visibility, skills, and so on. But these rules generally have two fatal flaws: they are only concerned with the most direct effects, and they don’t deal with edge cases.
Let’s take DnD as an example, specifically Pathfinder 1st edition. There is a spell called Shrink Item that, sure enough, shrinks an item significantly. The spell has rules for how the mass and dimensions of the item change as a result. So far so good.
But as soon as you try to poke at the spell a bit, you very quickly find what I term “undefined behavior”. For example, suppose we put the shrunk item into a container too small to contain the original item, and then tried to unshrink it. What would happen?
The spell doesn’t say, and neither do any other rules that I can find. This behavior is undefined by the rules. GM would have to come up with what happens themselves. Some of the consequences I would consider reasonable may be:
Unshrinking fails entirely.
Unshrinking proceeds until the item touches the walls of the container and presses into them with some small amount of pressure, then stops.
If the container is removed, then:
the unshrinking proceeds as before immediately.
unshrinking requires a second command word to be started.
item is permanently stuck in its current, smaller size (no magic effect is present anymore).
Container is destroyed, regardless of how strong it is, as the magic keeps the form of the item effectively indestructible throughout the process.
Item is destroyed, regardless of how strong it is relative to the container, as the magic attempts to expand the size of the item by adding more mass while not pressing on any surrounding material, which is not possible while preserving the shape and consistency of the item. As a result, the item is permanently destructively warped or shattered by the process - imagine how much you would have to deform a 1m long steel rod to make it fit within a container 20cm to the side.
Container and item press against one another, and whichever is stronger at the current size of the item doesn’t break (e.g. a 256cm2 rod shrunk to a 1cm2 cross-sectional area is only as strong as a rod of that material with 1cm2 cross sectional area).
Container and item press against one another, and whichever is stronger at the original size of the item doesn’t break (e.g. a 256cm2 rod shrunk to a 1cm2 cross-sectional area is as strong as the original rod).
This is what I mean when I talk about undefined behavior in the rules when indirect consequences of using magic are considered. Some fairly common-sense methods of interacting with spells simply give neither the GM nor the players any guidance in regards to what would happen.
Why is this bad? After all, GM can make rulings about this, right? Well, there are multiple reasons.
First of all, as I have already talked about when discussing Rules as Written, having explicit rules for what happens in different situations helps players plan things. Requiring GM to answer questions does not help, because that is exactly what we should be trying to avoid.
Secondly, it can be extremely hard to predict all of the consequences of a particular ruling while you are making it. For example, several of those rulings would have important consequences for manufacturing: you could either create extremely tough materials (by e.g. making a composite out of shrunken steel rods and some matrix), or make precise cuts in large objects by shrinking them first so they would be easier to work on. This would in turn have consequences on your worldbuilding, if players notice the connection. On the player's side, some of these rulings would make the spell amazing as a weapon.
Third, just because you ruled on one undefined behavior of the spell, doesn’t mean you have removed any other undefined behaviors. For example, what would happen if we shrunk the container instead - do the items stay in it, or do they break it apart? What if we shrunk someone’s armor while they were wearing it? Rulings on different undefined behaviors can interact unpredictably, and it’s hard to retract a ruling once it’s made. It’s especially bad if you later realize two different rulings are mutually contradictory.
Fourth, if you don’t write your rulings down, you are liable to forget which way you ruled in any particular case, which introduces its own set of problems.
While it’s possible to keep making these patches to individual interactions when necessary, I think it’s generally a poor approach, because it quickly makes your houserules list bloated while not providing any assurance that future issues will not show up. Furthermore, it makes magic seem like a hodge-podge set of rules (possibly suspiciously player-centric set of rules), as opposed to a fundamental part of the world. Generally, it would be better to find a way of consistently resolving all such problems at once.
Fortunately, humanity already discovered a consistent way of figuring out what would happen when disparate parts of a very large system with very complex high-level effects interact. It’s called physics. Real world isn’t inherently less complex than magic described in TTRPGs, especially when it comes to biology or engineering. Nonetheless, we have discovered how to consistently model it in order to make internally consistent predictions, even in cases where nobody ever tried to do something before.
What is required is a general Theory of Magic. Something that would break magic down to the level of basic interactions with regular physics. Then figuring out what happens when some novel edge case comes up would be as simple as taking the theory, doing some figuring, and getting a consistent answer out. It would not require the GM to intervene, it would be guaranteed to be consistent between different sessions, and it is not going to lead to internal contradictions.
Unfortunately, this is not humanly feasible. While I am sure it’s possible to make a simple set of physical rules that will simulate DnD magic to the level of fidelity of modern physics, I don’t see any human managing such an effort on their own. Fortunately, we don’t have to do this either.
Let’s look back on the development of regular physics. At first - back before writing was invented - humanity only knew surface level facts. They knew how hard specific kinds of rock were, how much you could bend specific branches before they broke, what kinds of things burned and what didn’t. But there was no real connection between these individual facts, and predicting properties you didn’t already observe was basically impossible. This is similar to how magic currently works in TTRPGs - each spell is, typically, it’s own little set of rules, independent of all others.
Then people slowly started to discover physics, and by the start of the 19th century we had a good number of connections established. We knew how pressure of a gas would behave when it was compressed, we knew some chemical elements, and their properties. At this point predicting the future, unobserved interactions becomes possible, and engineering things based on those interactions.
As the 19th century progressed, we connected these disparate connections into broad laws, like the Ideal Gas Law, or the periodic table when it comes to chemistry. Finally, in the 20th century we started to delve into truly fundamental laws, such as quantum physics, relativity, and so on.
However, when it comes to jumps in explanatory power of higher-level effects, I would argue that the 18th and 19th centuries were the largest ones. And that’s the sort of effect we need when it comes to TTRPG magic. There is no need to break the magic down to the level of atoms and fundamental forces, but having broad laws for how it all works - similar to ideal gas law - would already be incredibly helpful. Even if those laws would have obvious gaping holes, points of contradiction, or inability to explain some effects - so what? 18th century physics had massive holes, but it still gave us the steam engine. There is no need to explain everything or anything close to it in order to gain massive gains in broad explanatory power.
Development of such a Theory of Magic when it comes to DnD magic is what I would like to explore in this series. I hope it could be useful to some people.
If you are looking for other posts on my blog, check out this list of all other posts, and if you enjoy what I write, you can subscribe to receive updates by email:
Title picture uses and modifies "Rune Circle CCX 07" by Filter Forge, licensed under CC BY 2.0, and is thus itself licensed under the same license.
I'd like to point out that in Call of Cthulhu, casting magic is extremely difficult and usually dangerous. It is also quite uncommon for a PC in most campaigns. On a lark I once allowed players to spawn in knowing one spell, but they almost never got to use them. It quickly became a running joke, such as the PC who knew "Brew Space Mead": his ONLY character trait was his insatiable desire to acquire the ingredients to brew the spell. Even this proved virtually impossible.
But CoC was always an odd duck compared to the other popular tabletops.
A strong start! I'm eager to read this series