08 Nov 2010

Class Consciousness

Blog 16 Comments

Are classes products of role and power source? Or are they something else? Behind the cut, you’ll find my musings on how to think about character classes.

I believe some early criticism about 4e identified targeted a perceived disconnect between concept and character. With the limited palette in Player’s Handbook, certain concepts seemed harder to realize than were others. This perception, I suspect owes its origins to class functionality in the previous edition compared to the present one. In 3rd edition, a class was really a menu of benefits awarded at specific levels. Since classes were generally open, you might dip into one, two, or more at whatever levels you like to acquire the class features that served your character concept. It was possible to be a pure representative of a class by playing a fighter from level 1 to 20, but odds are the character underwent some fundamental shift through multiclassing into another class or picking up levels in a prestige class. And the more classes you accumulate, the less your character is truly a member of the original class.

Fourth edition shored up class boundaries by returning to a model closer to 1st and 2nd edition. Features, powers, and other benefits are reserved for actual members of a particular class. By declaring your membership to a particular class, you embrace all the story implications that go with that class. Playing a fighter means you favor defense over offense, you wear heavy armor and wield military weapons. Playing a cleric means you improve other around you, keep allies fighting, while still deliver powerful attacks. Although there are clear concepts associated with each class, there are ways with each class to customize a character to best fit your vision. As a wizard, you might favor illusion, summoning, control, or area attacks. These are largely tactical options and choosing one over the other doesn’t always reinforce a particular character concept. Still, there are enough stylistic difference to accommodate several different expressions of that class.

More than providing a concept template, I really believe class choice frames how you engage the imaginary world. Playing a fighter feels different from playing a wizard. There are commonalities between them both, but as a player you concern yourself with different areas in the game. Fighter players probably focus on individual tactics and containment, while wizards likely look at broad strategy and thwart it. Class, in my opinion, shapes how we experience game play.

An interesting consequence arises from all this. I believe the menu approach from the last edition served the game well, at least when you had a cogent character concept in mind. You could use all the game’s tools to tailor make a character. The problem as I see it stems from reducing a class’s importance by reducing it to a set of benefits from which you can pick and choose. Sometimes this leant itself to highly effective characters. Other times, it meant being trapped by a sub-optimal combination that could diminish a player’s ability to contribute to the group effort in a meaningful way. For me, making classes function as the character concept, a broad template designed to frame your choices along one of several available paths within the class identity, combined with mechanical parity results in a dynamic group armed with complementary techniques to overcome the challenges faced in the adventure environment.

Understand, building character concept into class design is not, in my mind, a flaw, but a benefit. Also, it’s not at odds with how the game used to play in the old days. When you create a character, you have two options. One, you can decide on the concept first and then find the class to match the concept or just pick a class and adopt the concept. It’s crucial to also add that while mechanics tend to be somewhat fixed, the story elements framing those mechanics are completely flexible. Just because the battlerage fighter is a hero who can shrug off damage and channel wrath into punishing attacks doesn’t mean you can’t explain the way this occurs for your particular character. You can easily alter this story so the mechanical effects result from some other story element. For instance, perhaps the character is possessed by a spirit and taps into that spirit’s power whenever he or she hits with an invigorating power. A control wizard’s spellbook might be crystals pregnant with power left over from creation and thus when he or she draws power from those crystals, he or she rewrites creation in a small way.

4e has evolved since the Player’s Handbook’s release and with it so has the class concept. I think the game has embraced the class as archetype now more than ever. Essentials achieves this by divorcing mechanics from a class and instead assigns them to a class’s specific expressions, or subclasses to be more specific. What we have now is a hierarchy of character definition that begins with a broad concept and then moves into specific expressions of that concept. There is the platonic idea of the fighter and then there are fighters in the world, fighters who are alternatively defined as knights, slayers, and weaponmasters. As each subclass is a discrete representation of a nebulous idea, there’s far more freedom in how we create and modify subclasses, allowing use to manipulate traditionally fixed elements such as power source, role, and mechanisms used to reinforce the role. Furthermore, we can still include options within a subclass depending on the subclass’s needs. With a knight and slayer, choice is more limited, while a weaponmaster has a wide range of options for maximum customization.

There may be some uncertainty with the new direction mainly because of confusion in what class meant and means now. Before, I think we put too much emphasis on power source and character role. Power source flavors the class, while role lays out the mechanical objectives for what the class should do in the game. As others have noted, this model leads to madness. It creates a weird grid wherein folks feel each power source demands a class to fill a specific role. In the world of arcane, we have a controller (wizard), defender (swordmage), leader (artificer, bard), and striker (sorcerer, warlock). Thus should we not have a martial controller? Likewise, should we not have a shadow leader? No. Emphatically no. Power source and class role are merely tools to guide design and should have nothing to do with world’s story. Does this mean there could be a martial controller? Maybe, but only if the concept works within an existing class or is broad enough to accommodate several different subclasses.

Looking at the classes populating the game currently, it feels to me that several have concept problems. Mechanically, they are fine, but their concepts are too narrow to exist as independent classes. Compare the fighter to the ardent. There can be all kinds of fighters in the game, knights, duelists, berserkers, and so on. But ardent? Now I’ll fight to defend the ardent’s existence, but I do feel that ardent would have been better served as a subclass of the warlord. It’s sounds like heresy to introduce new power sources to existing classes, but remember, a class is a broad concept that contains any number of expressions, even those that dip into other power sources.

I can imagine a spellsword subclass of the fighter. This character engraves runes and sigils on his or her weapon to better channel arcane energy. Most attacks might be melee basic attacks, but as the character gains levels, he or she can release arcane bursts on hits using the weapon. The subclass might be an arcane and martial controller. Here’s a taste:

Blazing Strike            Spellsword Attack

You channel arcane energy through your weapon so that when it hits, you unleash a torrent of multicolored flames.

Encounter * Arcane, Aura, Fire

Free Action            Personal

Trigger: You hit with a melee basic attack using a weapon.

Effect: You activate an aura 1 that lasts until the end of your next turn. Whenever an enemy ends its turn in the aura it takes fire damage equal to 3 + your Intelligence modifier.

Other classes aren’t so flexible and the problem with these classes is that the concept and expression are one and the same. A swordmage is a swordmage. Although certain features might vary, they have identical stories. Should the swordmage exist as a class then? Or would the swordmage function better as an expression of a different class. My gut tells me the swordmage should be a subclass of the wizard. Likewise, the seeker should be a subclass of the ranger, the ardent under the warlord, the battlemind under the fighter, runepriest under the cleric, and so on. Although I can see arguments for moving many classes under major classes (warlord to fighter, paladin to cleric, sorcerer to wizard), the thing to we must ask ourselves is whether the class by itself can accommodate many different concepts. If the answer is yes, then it should remain a class. If no, then the class should, in my opinion, become a subclass.

The following is purely an academic exercise designed to demonstrate what I’ve been chewing on here. I’m under no delusions that this will amount to anything more than mental masturbation, but it will at least, I hope, be interesting to you.


Concept: A ruthless killer who develops combat techniques to eliminate enemies one at a time.

Assassin (Dragon Magazine, shadow striker)

Avenger (Player’s Handbook 2, Divine Power, divine striker)

Executioner (Dragon Magazine, martial and shadow striker)


Concept: A devoted servant of the gods, armed with raw divine energy to further their patron’s interests in the world.

“Cleric” (Player’s Handbook, Divine Power, divine leader)

Invoker (Player’s Handbook 2, Divine Power, divine controller)

Runepriest (Player’s Handbook 3, divine leader)

Warpriest (Heroes of the Fallen Lands, divine leader)


Concept: A staunch protector of the natural world who calls upon primal magic to destroy nature’s enemies.

“Druid” (Player’s Handbook 2, Primal Power, primal controller)

Sentinel (Heroes of the Forgotten Kingdoms, primal leader)

Shaman (Player’s Handbook 2, Primal Power, primal leader)

Warden (Player’s Handbook 2, Primal Power, primal defender)


Concept: A warrior whose talent at arms is unmatched on the battlefield.

Barbarian (Player’s Handbook 2, Primal Power, primal striker)

Battlemind (Player’s Handbook 3, Psionic Power, psionic defender)

Knight (Heroes of the Fallen Lands, martial defender)

Slayer (Heroes of the Fallen Lands, martial striker)

Weaponmaster (Player’s Handbook/Class Compendium, Martial Power, Martial Power 2, Dark Sun Campaign Setting, martial defender)


Concept: A mystic warrior who transcends normal limits by perfecting body, mind, and spirit.

“Monk” (Player’s Handbook 3, Psionic Power, psionic striker)


Concept: A divine crusader who combines skill at arms with divine magic to triumph over darkness.

Cavalier (Heroes of the Forgotten Kingdoms, divine defender)

“Paladin” (Player’s Handbook, Divine Power, divine defender)


Concept: An individual armed with astonishing mental powers.

“Psion” (Player’s Handbook 3, Psionic Power, psionic controller)


Concept: A warrior, typically a loner, who walks the roads between wilderness and civilization.

Hunter (Heroes of the Forgotten Kingdoms, martial and primal controller)

“Ranger” (Player’s Handbook, Martial Power, Martial Power 2, martial striker)

Scout (Heroes of the Forgotten Kingdoms, martial and primal striker)

Seeker (Player’s Handbook 3, primal controller)


Concept: The consummate adventurer, with a solution to almost any problem.

“Bard” (Player’s Handbook 2, Arcane Power, arcane leader)

“Rogue” (Player’s Handbook, Martial Power, Martial Power 2, martial striker)

Thief (Heroes of the Fallen Lands, martial striker)


Concept: A character embodying raw, magical power.

“Sorcerer” (Player’s Handbook 2, Arcane Power, arcane striker)


Concept: An individual armed with magical power won from a pact with an otherworldly agency.

Hexblade (Heroes of the Forgotten Kingdoms, arcane striker)

“Warlock” (Player’s Handbook, Arcane Power, Dark Sun Campaign Setting, arcane striker)


Concept: A battlefield commander, whose superior tactics and leadership ability can turn the battle’s tide.

Ardent (Player’s Handbook 3, Psionic Power, psionic leader)

“Warlord” (Player’s Handbook, Martial Power, Martial Power 2, martial leader)


Concept: An arcane scholar whose learning grants access to vast magical energy.

Artificer (Eberron Player’s Guide, arcane leader)

Mage (Heroes of the Fallen Lands, arcane controller)

Swordmage (Forgotten Realms Player’s Guide, Arcane Power, arcane defender)

“Wizard” (Player’s Handbook, Arcane Power, arcane controller)

If you enjoyed this post, please consider subscribing to the RSS feed to have future articles delivered to your feed reader.

16 Responses to “Class Consciousness”

  1. Steve Townshend says:

    I not only agree, but wholeheartedly agree. I have a hard time quantifying a certain number of existing classes. This is partially due to my general lack of wit, I’m sure, but if you were to introduce those classes to me (as you have here) explaining that an ardent is a warlord build that uses psionics and that a warden is a druid build, etc, I would have a much more solid idea what to do with them. If I recall correctly, I remember 2e firmly embraced the “THESE ARE THE FOUR CLASSES” model, and then there were kits and spinoff classes for each of those, which made it pretty simple to figure out what you were supposed to do with your class.

  2. Sarah Darkmagic says:

    I’m struck by how narrow a few of these are, such as the warlock, sorcerer, monk and psion.

    I agree with you, the 3rd edition model was better if you had an idea of what you wanted your character to be. For someone like me, who was just learning how to create a character, the 4e model rocked. Similar issues happen in tech all the time. For instance, some people like statically-typed languages (where the types of variables are known when you write the code) because it’s harder to shoot yourself in the foot while others desire the flexibility that only dynamic languages can provide.

  3. robertjschwalb says:


    Although they may have only one representative class, I believe it would be easy to develop several subclasses for the warlock, sorcerer, monk, and psion. But that’s me :D

  4. Will says:

    This sounds like an excellent way to organize classes, and done right, I think it would make 4e-style multiclassing even easier; basically you could swap out your “subclass” features for those of another subclass. So a fighter/druid gets his “major” features from the fighter catalog, but pulls his “specialty” features from one of the druid subclasses (or a special druid menu, similar to hybrid powers). I think 4e multiclassing tried to get something like this with the power-swap feats but it didn’t quite work. With your two-tier distinction of superclass and subclass it could be more clear what the mix of abilities would be.

    And don’t discount multiclassing — for all the benefits of a class system, a lot of players still hate it. To this day, other RPGs still advertise themselves as “classless” as a major selling point because some people don’t like the restrictions of being forced into a class. The 3e mix-n-match splash-class system worked well to allay this but it weakened the idea of what a class was. A more consistent view of what a class is should also help create a multiclassing system that is both appropriate and easy to understand.

  5. Sarah Darkmagic says:

    Oh yeah, I just found it interesting is all. Although I’ll admit I thought the Essentials builds were more similar to the original class builds than new classes such as brutal scoundrel and artful dodger for the rogues. I like the essentials builds because they bring joy to some players, make some interesting design tweaks and give the designers something interesting to work on, but I’m selfish and wish there was “more inside” baseball talk about them.

  6. The Angry DM says:

    I am reminded of a question I asked when Heroes of the Fallen Kingdom first came out: under the Essentials design model, does class really mean anything? When you look at the Knight and the Slayer, what does it mean to say they are the same class? Mechanically speaking, there is little to suggest that those two subclasses are more closely related to each other than to, say, the thief rogue. Conceptually, I can see how they would be tied together except that you have to broaden the concept of fighter to such a degree that you wonder why warlord and ranger aren’t also just fighters.

    I’m also reminded of a debate I had once with a friend. He said he wanted to play a cleric who was a controller and complained that the game didn’t support that. I told him that the invoker is just that: a controller cleric in all but name. His response was “but I don’t want to play an invoker, I want to play a cleric who is a controller.” I was especially confused because the only difference in the flavor between the two seems to be in the ordainment process – clerics are ordained by mortal organizations, invokers are ordained by the gods themselves.

    The addition of the subclass, to me, seems to make the whole idea of class a little vestigial and your division of the classes seems to support that. After all, once you say that barbarian, warlord, slayer, knight, weaponmaster, and ranger are all fighters, what purpose does the label of fighter actually serve. They have different roles, different HP, (in some classifications) different power sources. The concepts become too broad and serve no mechanical purpose. Why not just say Slayer, Knight, Warlord, and all the rest are classes and drop the broad classification.

  7. Dan says:

    While you are at it, we should add back in racial restrictions on classes. Seriously.

    I would not mind seeing Sorc, Psion, and Warlock listed under Wizard (or Magic User even), Monk listed under Assassin or Rogue, Paladin under Cleric, Assassin under Rogue, etc.

    Get down to as few Superclasses as you can then work the Subclasses off of that list.

  8. Dan says:

    Angry – you posted as I was posting.. and you bring up a great point.

    If you took away all afiliations sub classes have with the superclasses you would just be left with classes… and that might not be bad way to go either.

  9. robertjschwalb says:

    I believe classes are important in much the same way that power source is important. While maybe not as useful to the audience, from a design perspective they help direct game design, story, and mechanical concepts. What I will readily admit, the current model pushes class further into the background since players now interact with class expressions rather than the overarching classes themselves.

  10. Ronin_Randy says:

    Couldnt you classify everything under the aegis of Cleric (Avenger is a priest of splashed with rogue, Invoker is a priest splashed with wizard, Runepriest is a priest splashed with fighter, Druid is a priest with a natural bent, etc.), Fighter (Monks fight with martial arts, Rangers fight in the woods, etc.) along with Rogue (bard and assassin) and Wizard (warlock, sorceror, and psion).

    Better yet, couldn’t you just use roles like origins in Gamma World? You’re a Striker/Striker, so you’re a Rogue. You’re a Defender/Striker, so you’re a fighter. You’re a Leader/Leader, so you’re a Warlord.

    It’s why I worried during 3E every time a new base class came out. I’m all for variety, but when you’re doing one of four things, you only need so many ways to do them.

  11. The Angry DM says:

    Thanks for the answer, but that leaves me very curious. I understand that role and power source were added as a way to facilitate tighter design over, say, in 3.5. For instance, in Classes and Races, it was mentioned that the bard and the monk suffered from a lack of direction and so ended up not filling any strong roll well enough to be effective. On the other hand, the cleric and druid lacked focus and could do almost everything very well, making them outshine other more focussed classes.

    So, given power source and role to guide design, how does the class itself provide additional design guidance. For instance, how would classifying the psionic controller as a wizard subclass have changed the design over creating itself as a class of its own or two seperate classes like “Telekineticist” and “Empath” (if my memory of the builds serves me correctly)? Is that too loaded a question? If so, I’m sorry.

  12. The Angry DM says:

    Err… except you didn’t suggest calling the psion a wizard. Sorry about that. I meant Ardent and Warlord.

  13. Landon says:

    Clearly, I need to learn more about 4e if I want to understand this article better. I can kind of see what you’re talking about via my understanding of previous editions, though, and, if I’m reading this right, it prompts a question in my mind. In 1e, classes started to proliferate because the classes as they existed just didn’t support certain concepts very well (or at all). For instance, if you wanted to play a D’Artagnan-like swashbuckler, there really wasn’t anything in the 1e PHB to do that with – if you played a fighter, you were just hurting yourself by not wearing armor and using a lighter sword, because the class features didn’t really “help you out” on that front; if I’m reading this correctly, that corresponds to the 3e situation that you describe as “being trapped by a sub-optimal combination that could diminish a player’s ability to contribute to the group effort in a meaningful way.”

    Where 1e came at the problem with new classes (like the Duelist – seriously, a d12 for hp?), 2e came at it with kits, which I take to be the corresponding case to subclasses in the new edition. Kits helped, some, but I think the problems with kits were (a) some of them didn’t help much (try playing a piratical fighter, and see how much the kit helps take the sting out of wearing little or no armor) and (b) now you have to wade through a zillion kits to find the one that works to support your concept.

    I’ll cop to being a huge fan of the 3.x/Pathfinder multiclassing system, because it let players start with concept and build from there, rather than seeing if they can find something in (mountains of) additional rules to support the concept they’d like to play. As I said, I don’t know much about 4e, so I don’t know enough about the “feel” of the gameplay to understand the added value of returning to foregrounding the classes as archetypes, rather than menu choices. A crucial point seemed to be that it had something to do with avoiding characters of disparate power (effectiveness) in a party, which can be a concern, I’ll admit. Still, that seems like an argument against non-classed systems in general, as well as (a not-really inaccurate) suggestion that 3.x/Pathfinder is only a pseudo-class-based system, since the classes do not require any “loyalty” from their exponents, but only offer themselves up to be cannibalized for parts, as it were, in realizing each specific character concept.

    I guess I have one worry, though. I always feel straight-jacketed by systems that are heavily invested in their archetypes, rather than systems that provide me with a toolkit for character building. I want to subordinate the mechanics to the story elements I have in mind, and – I’ll admit – I tend to favor unusual or non-standard character concepts that don’t always fit the standard archetypes. I suppose my worry is this, to put it more pointedly . 2e recognized that players would inevitably want to “branch out” in terms of concept, and attempted to provide mechanical support for this by introducing kits (similar to 4e’s subclasses, though less fully realized?), but this only ended up with a staggering (literally – you trying carrying around the complete “Complete” series) proliferation of mechanical patches and supplements. What’s to stop this from happening with 4e? Is this an inevitable consequence of archetype-centered systems (as opposed to toolkit/menu-option systems)? As the newness of a game wears off and players have explored all the familiar nooks and crannies – and especially as an ever larger body of gamers sign on board and each try to realize their PARTICULAR narrative visions within the scope of its mechanics – won’t the system always be pushed to accommodate an ever-growing number of new concepts? And if that is inevitable, does our experience with 2e show that the archetype-centered approach will eventually creak and sag under the weigh of this demand?

    I honestly don’t know enough about 4e to know the answers to these questions, so I’m really curious as to what you’ll have to say about this.

  14. Bartoneus says:

    I love this kind of discussion and am very glad you opened it up, though I may not agree with everything suggested or discussed in the comments I think it’s great to bring these things up and think about the basics of each edition and advantages/disadvantages to each. I like how almost all of the 4E classes have turned out, but when you bring up the Ardent easily falling within the concept of a build for the Warlord class I can’t help but agree and think it’s a great idea.

    The Swordmage I see was one of the most interesting and difficult classes to discuss in this sense, because it is pretty strongly in the middle of both wizard and fighter, I believe it’s one of the main concepts that probably led to the way of thinking in 3E that was designed so that you could literally dabble in two or more classes at the same time, but it never shook out quite right and seems to have led to the Swordmage being its own class in 4E. Of course that’s an oversimplification of the entire design process across editions in D&D, but still the gist of it seems pretty clear.

  15. Danmarks beste trylle shop says:

    Thank you for another magnificent post. Where else could anybody get that kind of info in such an ideal way of writing? I have a presentation subsequent week, and I am at the search for such info.

Leave a Reply