The Warlock Problem

Today is the twenty-fifth anniversary of the release of one of my favorite video games of all time, Final Fantasy VI (or Final Fantasy 3 if you’re the type who hasn’t played a video game since 1998). I spent a while thinking about whether I could do a post about it here. Could I talk about adapting Final Fantasy’s magic system to D&D? Could I make alternate rules for changing job classes mid-campaign? Should I come up with custom monsters or races? As I thought about the confluence between FF and D&D, I realized something about 5E that’s been bothering me for a while, though I haven’t had occasion to put it into words.

I’m calling it the Warlock Problem. It’s a pithy term but a bit of a misnomer. The warlock isn’t a problem. It’s a sole solution, and that in itself is a problem.

I can’t turn off the part of my brain that tries to adapt other media into the D&D ruleset (nor do I want to, honestly). I’ve spent some time lately reading into Final Fantasy, watching speed runs and the like, and during this time I thought about how I might use 5E to play an FF-like character. For example, consider the summoner. They’re not like a D&D summoner who conjures an ally and lets them run around for several turns. An FF summoner is a spellcaster who summons monsters for a single, usually short-term, magical effect. The most popular summons are offensive; Ifrit makes a fire attack then disappears, Siren disables enemy spellcasting then disappears, Knights of the Round spends a solid half-hour making thirteen attacks then disappears, etc. But some summons buff allies, or cast heal spells, or have some weird effect like making the party jump. New summons tend to be hard to get, they’re powerful but very expensive to cast, and often the summoner doesn’t have a lot of other offensive options besides their semi-unique weapon, the whip.

So to play an FF summoner in D&D, we need something that has fewer spells than other characters but doesn’t sacrifice any of the power behind them, has some variety in spell types, and also can use a weird signature weapon. It’s possible to approach this from a few different directions, but from what I see the quickest path is through the warlock. The warlock can only cast a few spells between rests, akin to casting spells with a high MP cost, but it has a decent list of spells known, even if the list is smaller than for other classes. Unlike a character who dips into wizard for a level or two, a warlock’s spells do increase in power and they do get access to high-level spells, and they can use cantrips (reskinned as tiny summons or low-level black magic) to stay relevant round-by-round. The pact depends on what’s important to you as a warlock: the pact of the blade grants a whip, the pact of the chain grants a persistent mascot-like summon, and the pact of the tome puts the warlock on a path to high-level magical rituals. With just a little tweaking, a warlock is a great way to play an FF-style summoner.

Now consider the blue mage, perhaps my favorite class. A blue mage is a decent fighter who can use decent armor, and usually they can analyze enemies to understand their weaknesses, but their trademark ability is blue magic. When an enemy attacks a blue mage with certain spells, the blue mage learns the spell and gains the ability to cast it later. Again, with a little reskinning, the warlock works here. It has a weird spell list, its spells get stronger over time, and it can get monster powers like darkvision or at-will spells. With the hexblade patron it also gets weapons and armor, and the hexblade’s curse can work like an enemy-analysis power. There are other ways to do it, but the warlock is quick and direct.

With the right reskinning, several other FF classes fall backward into the warlock. The geomancer? Use the pact of the tome to grab some druid cantrips and present all spells as terrain effects. The chemist? Focus on buff spells and treat the cantrips like guns or hurled potions. Gambler? Weapon and spell attacks use dice or cards, and spells are slot powers. It’s not a universal solution—obviously the D&D fighter is closer to an FF dragoon than a warlock is, and a small spell list limits the warlock’s accessibility somewhat—but it’s a close match to enough classes that it’s a little disconcerting.

That’s because 5E really doesn’t have anything else like the warlock. Its wide array of options makes it perfect for weird applications. A warlock could be an esoteric sort of wizard, a fighter with some supernatural tricks, an expert in magical knowledge but not spells themselves, or an actual monster who grows more monstrous over time. It’s not a full spellcaster, but it can leverage its few spells at a decent level without worrying about what its available daily slots are. It’s not a full warrior, but it can wade into melee and dish out pain almost as well as a martial class. All of its abilities are so unlike each other, and the one single class can branch into so many concepts it can almost fill a party all on its own without anybody duplicating a single option. The same can’t be said of any other class (except maybe cleric—it’s not as mutable as a warlock, but it’s far more mutable than a fighter, a wizard, or anybody else).

That’s great. I love the warlock. However, by comparison I can see how hard it is for other classes to break out of their comfort zone. I can’t fold the paladin, or the fighter, or the wizard into as many shapes as I can fold the warlock. They’re too much like themselves, while the warlock is almost a blank slate. But character classes are supposed to be blank slates. You’re supposed to mold them to fit your character concept, not change your concept to fit the class. That’s the warlock problem: the warlock is so good at representing so many different ideas, it makes starkly clear how other classes are not.

The other classes aren’t bad in and of themselves. They offer a lot of variation for playing within a European medieval fantasy milieu with a specific level of magic. But if you try to apply something different to them, like a character from a popular video game or movie or comic series, they break down. Look at Building Character, a blog about adapting popular characters to 5E rules. Each build is thorough and exhaustively researched (and excellently written), but the author still has plenty of occasion to say “okay, this doesn’t entirely work, but we’ll go with it” or some version thereof. Most characters only work as some hilarious hybrid of classes and powers. 5E at least allows players to multiclass, unlike 4E’s tight constraints or Pathfinder’s grudging acceptance, and that does open up character design opportunities. It’s great that you can play the lovable goof from the Kingdom Hearts series in D&D. But you should not have to build a warlock 14 / bard 3 / fighter 3 to do it. That’s a level of system mastery beyond most players, and it’s a crying shame that it’s required for a character so simple.

But it’s a hard problem to solve. A system has to be constrained in some way lest it spiral out of control. As Pathfinder shows, just having thirty different base classes can cause the exact same problem—each class is designed to fill a specific role or a specific archetype, not to let players tweak the class into the character of their dreams.

I think the solution isn’t more, specific classes; it’s fewer, generic classes. That is, use a small number of classes that each pertain to one core concept (martial / arcane / divine / primal, or tank / dps / healing / control, or something similar) and give each class a wide array of options related to that concept. You don’t have separate sorcerers and wizards, you have one class that lets you decide what your casting ability and class features are. You don’t have a paladin, you have a character with levels in defender (with a focus on heavy armor and weapons) and levels in healing (with a focus on buff spells). Give players building blocks, design with the expectation that players will multiclass instead of merely tolerance for the possibility, and let them do what they want.

I’m not saying this method is viable in 5E as it stands. It would basically be an overhaul of the entire class system, top to bottom. Also, 5E is much more interested in whether classes are balanced against each other and whether they fit the official, published, copyrighted setting than whether they’re capable of representing a wide range of characters. That’s not necessarily wrong, it’s just not terribly important to me. I’m more about freedom in class design than structure, and I’ve expounded on this before. So when I see a class that does actually offer some freedom, I get excited, and when I’m trying to adapt non-D&D concepts into D&D, that class is the first place I go.

Of course, this only applies when you’re trying to fit a specific character into a campaign. Creating a full adaptation of a setting does require an overhaul. I’ve toyed around with several such adaptations, but they’re all theorycraft, unlikely to see the light of day because I just don’t have the time to perfect them enough to make them game-ready. Maybe I should start.

This entry was posted in Game Design, Gaming Systems and tagged . Bookmark the permalink.