Dungeons & Design 10: Magic
LFQW--Linnear Fighter, Quadratic Wizard. I've referenced it before, but if it's not obvious, it refers to the ability of wizards to not only get better at what they do as they advance in experience, but to gain more abilities. As a wizard's spells increase, their capabilities increase, and there are enough printed spells to basically cover any sort of hole or weakness if the wizard can but find them. Fighters just punch things in the face better, and get no inherent ability to move through locked doors, make friends, fight magical monsters, or anything else with level.
It's worse with clerics, because they just get their spells from their patron and don't have to search through holes in the ground for scrolls.
I don't find this to be a particularly acceptable situation, but how to solve it while still allowing the low fantasy, sword-and-sorcery feel? Giving powers to everyone is a reasonable solution, and while Exalted is an excellent game and the Book of Weeaboo Fightan MagickBook of Nine Swords is an excellent supplement, they aren't the feel I'm going for.
What Does Magic Do?
In Greg Stolze's excellent Reign, the chapter on magic begins with some of the best advance for magic worldbuilding I've ever seen. Simply put, it tells you to ask yourself what magic can do, and not to accept an answer of "anything."
And that's really the foundational problem with D&D magic. It can do anything. Literally anything, in the case of the Wish spell, but the sheer number of spells published over the years fill in most of the cracks even if you ban Wish entirely. In pre-3.x editions, wizard power is constrained by the ability to find spells, but there are pretty much no guidelines for this and it also relies on the GM having a good understanding of the relative power level of spells and how likely a new one would be to break their game. Clerics, as I said, just get everything, so every new spell published increased their powers directly.
In Reign, magic is specifically limited in that it cannot affect the mind or heart. You can threaten someone with horrific death to convince them to go along with your plans, but there's no Charm Person to make someone your best friend. Furthermore, of the actual magic that does exist, a lot of it is highly limited and themed. Most of it requires "attunement," which allows the caster to use the more powerful spells in a specific school but forever limits them to spells of only that school. The choice then is between power and flexibility, and furthermore, between different kinds of power. One school of magic is about weather control and lightning, one is about emulating plants, one is about manipulating fire through dance, one is about wolves and the night, one is about unconvering knowledge through mathematics, one is about manipulating earth and stone through drumming, etc. Each school is tightly themed and mechnically limited.
That is interesting. "You can do anything" isn't interesting, and it's toxic to game balance.
Codifying Magic
Using D&D's basic spell list and existing classifications, there's no easy way to make a list like that work. The existing classifications for magic--Abjuration, Alteration, Conjuration/Summoning, Divination, Enchantment/Charm, Illusion/Phantasm, Invocation/Evocation, and Necromancy--seem like they're designed to develop a mechnical scheme for how magic works in the rules instead of arising organically from the way it works in the setting. Each school is way too limited to function on its own, which is why Necromancers get spells from a bunch of other schools too. Plus, there's no mythic or literary resonance for some of those schools. Diviners and necromancers, sure, and conjurers summoning fiends, but non-alchemist transmuters (other than witches turning people into frogs or themselves into cats)? Enchanters? Abjurers who aren't exorcists? It's just a mess.
Elementalism is one of the things that I tend to turn to when I'm trying to split magic up while still keeping it thematic. One of the neat things Al-Qadim did with its magic, in addition to the whole "jann fetching spells" mechanic, was to divide all the spells into elemental classifications, though substituting Sand for Earth. Air, Earth, Fire, and Water--or as I like to split it, Air, Earth, Fire, Metal, Water, and Wood--has a lot of mythic resonance, and it's relatively easy to determine where spells should go if you set it out beforehand.
Warhammer does something similar with its Colour Wizardry, which break things down a bit further while greatly limiting the total number of spells any individual caster has to keep track of. In D&D terms, a powerful WFRP wizard would have probably 20-25 spells, and a master archmage 30+. That certainly stops the GM from having to remember too many interactions, or the player from forgetting what they can do, and keeps the QW part from getting too out of hand. A wizard is always going to be able to do more than a non-wizard under any gaming paradigm where the world is "like ours, but some people have magic," which is the way most sword and sorcery fiction is structured. The trick is to keep it limited to an appropriate niche.
I've actually already converted a bunch of WFRP stuff into D&D terms for later use, because I love the way the magic is structured, and the basic colors in the link above plus ice magic is a neat nine-fold system. I suspect I'd go with that. I'm running a WFRP game right now, so I can at least test the mood and theme.
You may have noticed that I didn't mention the cleric/wizard split. That's because it appears essentially nowhere before D&D, only makes sense as a D&Dism and doesn't have any wider or pre-existing basis, and I don't like it. I prefer spellcasting priests to just be wizards who are also priests. Mechanistic gods granting miracles raises a ton of questions about the world structure that I don't really like, unless it's some kind of genius loci, demon, or mad thing from beyond time and space.
It's not like the cleric has a classical or sword and sorcery inspiration anyway. It was developed from players complaining that they needed a way to deal with the guy playing a vampire named Sir Fang. Cue Hammer Horror films starring Peter Cushing, and bam, cleric. No thanks.
Limiting Magic
Okay, so if the spells lists are cut down to a more manageable size, how to further prevent spellcasters from waving their hands and solving all the problems the characters run into?
One thing is to increase the casting time so it's not just snapping fingers and having people burst into flames. This is the route that 4e took with moving most utility spells into rituals. When the ability to open a door using magic requires a mere gesture, mundane skill at picking locks is devalued. This is admittedly less true in a pre-3.x Vancian system, since memorizing Knock prevents the wizard from memorizing other spells to deal with other situations, but even there there are a ton of things wizards can do that take very little time that can solve all kinds of problems. If opening a door with magic takes ten minutes, though, then there's a real question to be asked. Is it worth it to use magic that you know works if it means you might get interrupted? What if the people on the other side of the door hear all that chanting? Why not just have the thieftreasure hunter pick the lock?
Increasing casting times and adding components does happen in the original D&D spells, but the components are a joke (literally--guano to cast fireball?) and most casting times are pretty low. Also, if you do it to combat spells, then actually being a battle wizard is really boring. Most actions are just I keep casting. That's okay in a video game, when it still only takes a few more seconds, but in a TTRPG when you have to wait for initiative to go around the table it's really boring. On the other hand, letting a wizard cast combat spells the same way archers can shoot arrows isn't going to break anything, and requiring extra time to cast AoE spells, maybe a round or so more to make them easier to stop, won't either.
To bring up WFRP again, all of the spells that require less than a minute to cast are personal in scope. They hurt people within 50 yards, or affect you, or people you can touch, or maybe ten people in a close radius. Anything more powerful or wide-spread requires a ritual, which takes hours to cast and usually rare or dangerous ingredients--excessively so, in my opinion, but the basic principle is sound. I personally like that interpretation of magic, where wizards can affect things nearby pretty easily but anything else is more difficult or time-consuming. Exotic components encourage adventures and interaction with the world, and rituals that need to be performed in a certain way or at certain times are also good plot-fodder. Minor magic is easy, but really magical things require effort. That's a good niche and, along with limited spell lists, prevents the flying, fireballing, invincible wizard.
Vancian Magic
I don't like it.
Not for any logical reason. It makes sense as a way of resource management, where spells are essentially tools that the wizard takes with them into the dungeon and then expends them on various problems, but I just don't like the feel. I greatly prefer a fatigue system, or a spell misfire system.
True20 has Fatigue Checks after spells, and failed one causes Fatigue, and if you're Fatigued and fail again, you become Exhausted. WFRP checks for doubles, triples, or quadruples on rolled spells and triggers mishaps, ranging from a cold wind blowing through the area or flowers near the caster dying all the way up to a demon appearing and dragging the screaming caster bodily into Hell.
There's also spell point systems, but while they're workable, they aren't really evocative in my opinion. Video games use them because it's easy to program (much like hit points), and by now it's tradition, but we can model a lot of other, more interesting mechanics in TTRPGs.
How Common Should Magic Be?
This is something I go back and forth on. Typically in fantasy settings, magic is a discrete thing partitioned off to the side. Wizards use it, some monsters use it or need it to survive (insert "how do dragons fly?" debate here), but like I said above, it's our world, but while some few people have magic, most people do not.
That doesn't actually mesh with the way magic is treated in the real world, though. Magic, even in the D&D sense of repeatable actions designed to provoke a supernatural response, is ubiquitous on a low level. Knocking on wood. Wishing on a star. Blowing out all the candles at once. This kind of folk magic is how most people thought it worked in the past. If you left out a bowl of milk for the faeries, it wasn't because you were a lechemancer with a supreme understanding of the power of cow secretions, it's because faeries loved milk and it just worked. Rubbing crushed shamrocks mixed with holy water on your eyelids to see through faerie illusions doesn't need any extra knowledge because the power isn't in the user, it's in the materials or in the way reality works.
I like the idea of washing your spear in water bathed in moonlight to give it power against supernatural monsters. I like the idea of chanting over athelas to awaken its healing power. I like the idea of burning herbs to drive out disease-causing spirits. I like love-potions and health charms and cradle blessings and rune-inscribed bandages and the evil eye.
D&D doesn't touch that space at all, really, but Runequest has a mechanic called Folk Magic that's pretty much entirely about this, and Exalted's thaumaturgy is quite similar as well. Since I'm ditching the whole class-and-level thing, having a skill for minor magics wouldn't throw anything out out of whack. If anyone can take it, and if it costs the same for everyone, then there isn't a balance problem to the mechanic as a whole.
That was a lot. I think I'll end it here. If I have any more thoughts about it, I'll obviously write another post. Otherwise, next up will be social mechanics. (^_^)v
It's worse with clerics, because they just get their spells from their patron and don't have to search through holes in the ground for scrolls.
I don't find this to be a particularly acceptable situation, but how to solve it while still allowing the low fantasy, sword-and-sorcery feel? Giving powers to everyone is a reasonable solution, and while Exalted is an excellent game and the Book of Weeaboo Fightan MagickBook of Nine Swords is an excellent supplement, they aren't the feel I'm going for.
What Does Magic Do?
In Greg Stolze's excellent Reign, the chapter on magic begins with some of the best advance for magic worldbuilding I've ever seen. Simply put, it tells you to ask yourself what magic can do, and not to accept an answer of "anything."
And that's really the foundational problem with D&D magic. It can do anything. Literally anything, in the case of the Wish spell, but the sheer number of spells published over the years fill in most of the cracks even if you ban Wish entirely. In pre-3.x editions, wizard power is constrained by the ability to find spells, but there are pretty much no guidelines for this and it also relies on the GM having a good understanding of the relative power level of spells and how likely a new one would be to break their game. Clerics, as I said, just get everything, so every new spell published increased their powers directly.
In Reign, magic is specifically limited in that it cannot affect the mind or heart. You can threaten someone with horrific death to convince them to go along with your plans, but there's no Charm Person to make someone your best friend. Furthermore, of the actual magic that does exist, a lot of it is highly limited and themed. Most of it requires "attunement," which allows the caster to use the more powerful spells in a specific school but forever limits them to spells of only that school. The choice then is between power and flexibility, and furthermore, between different kinds of power. One school of magic is about weather control and lightning, one is about emulating plants, one is about manipulating fire through dance, one is about wolves and the night, one is about unconvering knowledge through mathematics, one is about manipulating earth and stone through drumming, etc. Each school is tightly themed and mechnically limited.
That is interesting. "You can do anything" isn't interesting, and it's toxic to game balance.
Codifying Magic
Using D&D's basic spell list and existing classifications, there's no easy way to make a list like that work. The existing classifications for magic--Abjuration, Alteration, Conjuration/Summoning, Divination, Enchantment/Charm, Illusion/Phantasm, Invocation/Evocation, and Necromancy--seem like they're designed to develop a mechnical scheme for how magic works in the rules instead of arising organically from the way it works in the setting. Each school is way too limited to function on its own, which is why Necromancers get spells from a bunch of other schools too. Plus, there's no mythic or literary resonance for some of those schools. Diviners and necromancers, sure, and conjurers summoning fiends, but non-alchemist transmuters (other than witches turning people into frogs or themselves into cats)? Enchanters? Abjurers who aren't exorcists? It's just a mess.
Elementalism is one of the things that I tend to turn to when I'm trying to split magic up while still keeping it thematic. One of the neat things Al-Qadim did with its magic, in addition to the whole "jann fetching spells" mechanic, was to divide all the spells into elemental classifications, though substituting Sand for Earth. Air, Earth, Fire, and Water--or as I like to split it, Air, Earth, Fire, Metal, Water, and Wood--has a lot of mythic resonance, and it's relatively easy to determine where spells should go if you set it out beforehand.
Warhammer does something similar with its Colour Wizardry, which break things down a bit further while greatly limiting the total number of spells any individual caster has to keep track of. In D&D terms, a powerful WFRP wizard would have probably 20-25 spells, and a master archmage 30+. That certainly stops the GM from having to remember too many interactions, or the player from forgetting what they can do, and keeps the QW part from getting too out of hand. A wizard is always going to be able to do more than a non-wizard under any gaming paradigm where the world is "like ours, but some people have magic," which is the way most sword and sorcery fiction is structured. The trick is to keep it limited to an appropriate niche.
I've actually already converted a bunch of WFRP stuff into D&D terms for later use, because I love the way the magic is structured, and the basic colors in the link above plus ice magic is a neat nine-fold system. I suspect I'd go with that. I'm running a WFRP game right now, so I can at least test the mood and theme.
You may have noticed that I didn't mention the cleric/wizard split. That's because it appears essentially nowhere before D&D, only makes sense as a D&Dism and doesn't have any wider or pre-existing basis, and I don't like it. I prefer spellcasting priests to just be wizards who are also priests. Mechanistic gods granting miracles raises a ton of questions about the world structure that I don't really like, unless it's some kind of genius loci, demon, or mad thing from beyond time and space.
It's not like the cleric has a classical or sword and sorcery inspiration anyway. It was developed from players complaining that they needed a way to deal with the guy playing a vampire named Sir Fang. Cue Hammer Horror films starring Peter Cushing, and bam, cleric. No thanks.
Limiting Magic
Okay, so if the spells lists are cut down to a more manageable size, how to further prevent spellcasters from waving their hands and solving all the problems the characters run into?
One thing is to increase the casting time so it's not just snapping fingers and having people burst into flames. This is the route that 4e took with moving most utility spells into rituals. When the ability to open a door using magic requires a mere gesture, mundane skill at picking locks is devalued. This is admittedly less true in a pre-3.x Vancian system, since memorizing Knock prevents the wizard from memorizing other spells to deal with other situations, but even there there are a ton of things wizards can do that take very little time that can solve all kinds of problems. If opening a door with magic takes ten minutes, though, then there's a real question to be asked. Is it worth it to use magic that you know works if it means you might get interrupted? What if the people on the other side of the door hear all that chanting? Why not just have the thieftreasure hunter pick the lock?
Increasing casting times and adding components does happen in the original D&D spells, but the components are a joke (literally--guano to cast fireball?) and most casting times are pretty low. Also, if you do it to combat spells, then actually being a battle wizard is really boring. Most actions are just I keep casting. That's okay in a video game, when it still only takes a few more seconds, but in a TTRPG when you have to wait for initiative to go around the table it's really boring. On the other hand, letting a wizard cast combat spells the same way archers can shoot arrows isn't going to break anything, and requiring extra time to cast AoE spells, maybe a round or so more to make them easier to stop, won't either.
To bring up WFRP again, all of the spells that require less than a minute to cast are personal in scope. They hurt people within 50 yards, or affect you, or people you can touch, or maybe ten people in a close radius. Anything more powerful or wide-spread requires a ritual, which takes hours to cast and usually rare or dangerous ingredients--excessively so, in my opinion, but the basic principle is sound. I personally like that interpretation of magic, where wizards can affect things nearby pretty easily but anything else is more difficult or time-consuming. Exotic components encourage adventures and interaction with the world, and rituals that need to be performed in a certain way or at certain times are also good plot-fodder. Minor magic is easy, but really magical things require effort. That's a good niche and, along with limited spell lists, prevents the flying, fireballing, invincible wizard.
Vancian Magic
I don't like it.
Not for any logical reason. It makes sense as a way of resource management, where spells are essentially tools that the wizard takes with them into the dungeon and then expends them on various problems, but I just don't like the feel. I greatly prefer a fatigue system, or a spell misfire system.
True20 has Fatigue Checks after spells, and failed one causes Fatigue, and if you're Fatigued and fail again, you become Exhausted. WFRP checks for doubles, triples, or quadruples on rolled spells and triggers mishaps, ranging from a cold wind blowing through the area or flowers near the caster dying all the way up to a demon appearing and dragging the screaming caster bodily into Hell.
There's also spell point systems, but while they're workable, they aren't really evocative in my opinion. Video games use them because it's easy to program (much like hit points), and by now it's tradition, but we can model a lot of other, more interesting mechanics in TTRPGs.
How Common Should Magic Be?
This is something I go back and forth on. Typically in fantasy settings, magic is a discrete thing partitioned off to the side. Wizards use it, some monsters use it or need it to survive (insert "how do dragons fly?" debate here), but like I said above, it's our world, but while some few people have magic, most people do not.
That doesn't actually mesh with the way magic is treated in the real world, though. Magic, even in the D&D sense of repeatable actions designed to provoke a supernatural response, is ubiquitous on a low level. Knocking on wood. Wishing on a star. Blowing out all the candles at once. This kind of folk magic is how most people thought it worked in the past. If you left out a bowl of milk for the faeries, it wasn't because you were a lechemancer with a supreme understanding of the power of cow secretions, it's because faeries loved milk and it just worked. Rubbing crushed shamrocks mixed with holy water on your eyelids to see through faerie illusions doesn't need any extra knowledge because the power isn't in the user, it's in the materials or in the way reality works.
I like the idea of washing your spear in water bathed in moonlight to give it power against supernatural monsters. I like the idea of chanting over athelas to awaken its healing power. I like the idea of burning herbs to drive out disease-causing spirits. I like love-potions and health charms and cradle blessings and rune-inscribed bandages and the evil eye.
D&D doesn't touch that space at all, really, but Runequest has a mechanic called Folk Magic that's pretty much entirely about this, and Exalted's thaumaturgy is quite similar as well. Since I'm ditching the whole class-and-level thing, having a skill for minor magics wouldn't throw anything out out of whack. If anyone can take it, and if it costs the same for everyone, then there isn't a balance problem to the mechanic as a whole.
That was a lot. I think I'll end it here. If I have any more thoughts about it, I'll obviously write another post. Otherwise, next up will be social mechanics. (^_^)v