(Sorry if this should be in the House Rules section. Since I am not proposing any specific rules, I figured it might as well go here.)
One thing I liked about late-era 3.5 D&D was the inclusion of specialized spellcasting classes that were given access to only a few of D&D’s traditional schools of magic. So the dread necromancer (bad name) got necromancy and a bit of fiend-binding, the beguiler (worse name) got enchantments and illusions, the summmoner (good name) got conjurations, the elementalist got elemental spells, and so on. This is kind of like Erikson’s Malazan Book of the Fallen, if you’ve read that. It’s also not completely dissimilar to what Gygax what planning for his proposed second edition.
One thing I have always disliked about D&D and its iterates is the presence of the Cleric. This is partially because of the prevalence of the healbot meme, and partially because I like D&D most when it has a pulp fantasy feel. The cleric, as influenced by Judeo-Christian tradition and implemented in D&D, is not necessarily a common archetype in pulp fantasy. (This is by no means an original insight on my part.) Clerics of Pelor or Titivilla can turn sticks into snakes and call down insect plagues because of the Bible and Gygax, not because of what Pelor or Titivilla represent as deities. The idea of a priest is still important and powerful, but it doesn’t have to be tied down to a specific class. Most priests of Torm are probably paladins or fighters with the Knowledge (Church of Torm) proficiency. Most clerics of Wee Jas are necromancers or assassins. And similarly with other deities.
So what precisely is my point? The natural conclusion from my two premises is to get rid of the cleric and mage classes (and the priestess, warlock, and witch classes if you are using them) and introduce a bunch of specialized classes in their place. This means abandoning the arcane-divine divide, which is okay because it never made a huge amount of sense in the first place. Fictional magic being what it is, the number of ways you could accomplish this division is almost limitless. An easy set of classes would be:
Necromancer (necromancy and soul-magic)
Sorcerer (summoning and other conjurations)
Elementalist (elemental spells; not just a damage-dealing class because that’s boring)
Enchanter (mind-affecting spells)
Abjurer (protective and many healing spells)
We collectively refer to these classes as mages.
Obviously a lot more could be added, or you could combine the sorcerer with the elementalist or necromancer or whatever. To reiterate, these classes will be assigned what used to be both arcane and divine spells, including healing. Cure light wounds could easily be a spell for necromancers, sorcerers, elementalists, and abjurers, for example.
To make these classes both interesting and distinct, a lot of spells would have to be added to the base game. Luckily, a lot of spells is precisely what D&D has. So that’s okay. Actually going in and divvying up spells would be a nontrivial but not insurmountable amount of work.
The more troubling aspect is spell acquisition. In their 3.5 iterations, these classes automagically knew every spell on their spell list. But the acquisition of new knowledge through scrolls and collaboration is a super fundamental (and fun) part of the older editions ACKS emulates. If your group’s mages are a sorcerer and an elementalist, no one will be excited if the latest treasure trove has scrolls of animate dead and feeblemind in it. There are a couple of solutions.
- The best solution is no solution: Accept this situation as another aspect of the randomness inherent to roleplaying. Players will get over it, particularly since broad coverage of the mages can be obtained with henchmen.
- Return policies: Assume that sufficiently large numbers of friendly or neutral mages exist such that unusable scrolls can be traded for usable ones. Such an assumption would impose large ramifications for the makeup of your setting, and not all of these would necessarily be positive.
- Allow scrolls to be consumed in magical research to learn new spells.
- Allow mages to use scrolls of spells that are not from their class’s list with some chance of failure. You could make this chance flat, or have it be affected by your level and/or the level of the spell. Playtesting and mathhammering would reveal what works best.
- As 4., but also give mages some chance of being able to learn spells not on their class’s list.
Has anyone on this forum experimented with similar ideas? Did they come up with any elegant solutions to the main problem, or did they discover any other issues that I have not foreseen?