What Classes Should be in D&D?

For me, choosing a class has always been one of the most fun and important decisions to make while playing Dungeons & Dragons. I can still remember the feeling of pure excitement I had when I first cracked open the 3rd Edition Player’s Handbook and saw that Monk was a core class. I also remember our friends all having multiple discussions about what exactly the Sorcerer class was and how it was different from the Wizard. I view these discussions along the same lines as what would happen if the Fighter, Wizard, or Cleric were left out of the first Player’s Handbook for an edition. With the next edition of D&D now in open playtest, I felt it was a good time to discuss the varying levels of class distinction in D&D.

Considering the Monk is the class I played the most of in 3E, I was surprisingly happy it wasn’t included in the first PHB for 4th Edition. It never felt quite right to me as a class presented as an introductory option for D&D players unless it was specifically for an Oriental Adventures style of game. I think that if you boiled down the options for character classes to the most basic you would end up with Fighter, Cleric, Wizard, and sometimes Thief/Rogue. Beyond these 3-4 options the list of secondary classes can vary greatly. Depending on player preferences, classes such as the Bard, Druid, Paladin, Ranger, Barbarian, and many others can all be seen as important options for players. My opinion is that beyond the four main classes, most of the other options readily fall into two categories: specialization or combination.

Combinations / Multi-Classing

I would like to go through a quick experiment. Let’s put aside many of our assumptions/favorite characters and look at some of D&D’s classes in their raw forms. I don’t think it does any disservice to the Paladin class to say that it is a combination of the Cleric and Fighter concepts. In the same way, I believe you can say that most Bards fall somewhere between Rogue and Wizard, and that most Rangers can be put somewhere between Rogue and Fighter. The toughest combination I find is the Wizard/Cleric, but I keep coming back to the Druid as a class that often feels like  it is somewhere between those two classes. Surely you can come up with a Bard, Ranger, or Druid that is nothing like any of the other classes, but I think having a class paradigm to begin with makes those break out concepts even more exciting and this is, after all, only an experiment.

This gives us a fairly clean wheel of class relationships:

Fighter – Paladin – Cleric – Druid – Wizard – Bard – Rogue – Ranger – Fighter (loop)

I really enjoy seeing a nice, clean, logical layout for classes and how they relate to one another. I enjoy thinking about the sliding scale between the primary classes and imaging what a Paladin would look like closer to the Fighter end (maybe close to or having already lost faith) or closer to the Cleric end (extremely dedicated, focused, and not rearing for a fight). I also think it gets interesting when you think about pushing a primary class closer to one or the other secondary classes. Imagine playing a Cleric or a Fighter that leans towards the Paladin concept between them, or a Rogue that dabbles in music/song and discovers a yet unknown arcane spark within themselves leaning towards Bard.

Another reason I really like this wheel of classes is that it shows the distant relationship between the Fighter/Wizard classes which leads to classes like the Spellsinger or Swordmage, which are often bizarre or hard to balance with other classes. I haven’t heard a lot of demand for a Cleric/Rogue class, but I’m sure it’s existed in the past and that people love playing that combination.

The important distinction for these classes is whether or not you can play a Paladin from the beginning, or if you have to start as a Fighter (or Cleric) and multi-class to become a Paladin. I believe that with the secondary classes I’ve presented here most players want to start playing as a Paladin/Druid/Bard/Ranger, they don’t want to slog through a level or three of being “another class” and then become what they wanted to be all along. Plus, while it might make sense for a Paladin to start out as a Cleric or Fighter in some worlds, it doesn’t always make sense and it stretches even more thin for Druids, Bards, and Rangers.


The first thing I think of for a class that defines this category is the Assassin. To me the Assassin class falls clearly into the wheelhouse of a specialized Rogue/Thief concept. Certainly not all Rogues are Assassins, but to me it seems a safe assumption for all  classes involved to say that all Assassins are conceptually Rogues. Essentially, I’m saying the Assassin is a specialized Rogue. Several other classes feel like specializations as well, though they may not have as clear a connection to any one class. To me classes like the Barbarian, Monk, Sorcerer, Warlock, Necromancer, and even the Psion could all be considered specialized classes.

I believe that how classes are presented in any given edition of D&D is actually a big indication of how that game is going to feel. Being able to play a level 1 Assassin or Monk is a pretty big difference from having to go through 5 or even 10 levels of another class first. Taking all of this into consideration, I must say that the last thing I want to see in any D&D game is a player who wants to play a Psion being forced to play a classic Wizard character for 10 levels. I’m less scared by the idea of an Assassin player being a rogue that specializes in backstabs and then officially becoming an Assassin at a higher level. Would it all that different to play a Fighter specializing in unarmed combat/mobility for a few levels and then become a Monk than to play a Monk from the start?

I could even see adding a third level of classes to the scale I presented above, where Barbarian is placed somewhere between Ranger and Fighter. However, I feel that starts to get very muddy and I start to see classes getting restricted or feeling shoved into a place they don’t quite fit. I started thinking about this whole discussion because the playtest for the new version of D&D and the distinction of the Cleric as a more warrior-like class and the possible addition of a Priest class that would cover the less “fighty” side of the Cleric concept.

My personal feeling on the matter is if you can accomplish something simply and easily by providing more options for existing classes, then there is no need to add more classes. For example, a Monk can be a lot more than just a fighter that punches things and a Barbarian more than just a Fighter that gets angry. I feel that a Priest class would simply be a Cleric that casts spells instead of fighting, not a lot more to distinguish it other than the assumption that all Clerics wear armor and fight in melee. My inclination is that the Assassin should also not be an available class in the early game, because it mostly feels like a Rogue that specializes in killing, but if you explore ideas like mastering poisons then it starts to come into its own a little bit more.

What Does it Mean to be Level 1?

Something that I keep coming back to is that the spread of classes available to a player at 1st level can imply some incredible things about what it means to be a level 1 character. Surely a 1st level Wizard could be seen as only an apprentice/novice, but how much distinction is there between a 1st level rogue and a 1st level assassin? Does playing an Assassin or a Monk from level 1 limit a player’s opportunities for character growth and adaptability? There have been several times over the last few years where I, as a DM, have encountered a situation in play where it would be an interesting story for a player/character to switch classes and had backlash from it because the players were very hesitant to keep the same character but change their class.

As I continue to grow as a DM, I find myself questioning things like starting classes more often. If players can choose from a wide variety of very specialized classes right from the beginning, then what they choose becomes a very large part of their character’s identity. I feel that players could be just as satisfied making an archer Fighter and simply give the character the in-game title of “Ranger” (think about the Night’s Watch in Game of Thrones) than they are if there is a Ranger class available to them, and this keeps the player’s options open to change and adapt with the story if the opportunity arises.


  1. I was having a very similar discussion over on fellow RPG Bloggers Network member Points of Light under his post looking at the 5e Paladin and saying much the same thing. The design element I questioned was whether or not we need “dual-class” characters when secondary classes like the Paladin are already a combination of core classes and one with more story behind it than the simple mechanical act of combining abilities from two classes. I would totally go for a streamlined system that featured combo classes with their own role and flavor in the game world, and use the concept of Themes (formerly Kits) to allow for specialization as you call it, based on culture or professional focus. In the discussion we did end up compromizing and allowing for story-relevant dual-classing, allowing characters to “pick up” abilities from other classes, trading out core abilities to simulate training focus.

    Of course, I freely admit I haven’t read a D&D book since 2nd ed. (and I have yet to hear anything about 3-4 ed. that makes me want to “upgrade”. In the last 20 years, I’ve played a lot more games that were essentially of wholly classless. Players create a core concept with whatever abilities they can afford through the point-buy system (usually) and set goals for where they want to take the character, though such plans may change based on stories and opportunities. I think you’re correct in saying that classes are a limiting factor on story telling. It IS a component of D&D that won’t go away, and can be useful in teaching players how to build characters, balance them, even just know what to pick! However, more experienced players can certainly do more with a more complete and flexible tool kit.

    One thing you do get with classless is a complete and satisfying core concept for your character, rather than a framework that requires a lot of play to flesh out. Classes and leveling are great for very regular campaigns, classless for occasional or infrequent play. If we can’t play often, you’ll like a character that can already do the stuff you want it to do.

    I think 5e may be moving towards more useful/powerful 1st level characters as a way to compensate for this, which is a reality facing more and more gamers (getting older, etc.).

  2. Rather then a continuum, arrange the classes in a flow chart or family tree, with the primary ancestors being Fighter, Rogue, Cleric, and Wizard in the original continuum. Then the secondary classes flow off from those, followed by the tertiary classes.

  3. Zoarmt: In that scheme, do combination classes look like tangled branches? (i.e. is the Paladin in both the Fighter and Cleric tree?)

  4. Alhazred says:

    There sure are a vast number of ways to slice and dice class hierarchies, especially when you start throwing things like themes into the mix. There are of course no ‘right answers’. 4e is interesting in that it has very high granularity. A fighter for instance isn’t just a couple of mechanics, it is a whole suite of things, each one of which is individually pretty small but that add up to a pretty specific concept. It also tends not to care too much what you use something for. People complained a lot about 4e fighters and archery, but it just didn’t matter that much because you could base your concept on the Ranger class and just select the options that worked for your character. If you needed some elements that class couldn’t provide then you could MC a bit or even make a hybrid, or often just pick the right feats (usually with a basic MC feat). There wasn’t really a strong reason to actually need a lot of in-between classes (they exist in spades, but honestly I think that was a mistake).

    One question that arises here is just how attached SHOULD specific things be to one class? In 4e you HAD to be a wizard (of some ilk) to have a wizard power. It felt to me like they made a mistake there and powers should have been more broadly accessible pools (like broken down into lists by power source instead of class). 5e will ask the same questions. Do you HAVE to be a wizard to have wizard spells? Is there even a list of spells that is labeled that way? Many choices exist. You could water down classes too much by making pools like that, or you could miss a good opportunity by not doing it (like 4e seems to have). I don’t expect any set of answers will be perfect. Whatever ends up being the 5e system will turn out to have limitations and quirks.

  5. I wrote about basically the same subject just a few weeks ago on RPGMusings. Basically, I was saying that there should be 4 base classes (comprised of what 4E referred to as Roles), and everything can build off of that through multiclassing (using what 4E referred to as Themes). This would allow players like myself who dislike the constraints of picking one specific class path to mix and match to our heart’s desire. At the same time, if a player wants to just ride along a narrowly defined path, he or she can simply do that.

  6. Philo Pharynx says:

    Some of this seems to be a part of themes. Looking at the open playtest, it seems that themes can be used to reinforce your existing class or as a sort of dabbling that’s less than multiclassing. For example the “Magic User” theme lets you take two at-will magic spells. I could see a rogue with the magic user theme being very cool.

    Still I’m a fan of having separate classes besides the combination. I’m also a fan of having classes you can’t take at first level. I could easily see somebody starting as a barbarian, but I see assassin as something you have to earn.

  7. The reason I don’t really belong in any discussions about D&D is I look at this, and I think, if we are going to have classes, a class should represent some fairly specific concept. Monk would be pretty good if it didn’t have a bit of being racially problematic, Wizard is worryingly broad, and Fighter (maybe even in 4e) is hopelessly uselessly broad, Paladin is pretty great, Cleric is a bit of a confusing marriage of convenience, and so on. If anyone who primarily uses “non-magic” and is focused on fighting is the same class, what’s the point of having the Fighter word?

    When you start to say that only three or four classes make sense, I start being really confused about why you have classes at all. Why not assign character competences some other way?

  8. Because then it wouldn’t be D&D (and remember, the original game didn’t very much more in the way of classes). There are many artificial elements in the game and that have always been in the game. Maybe that’s what makes it a “game”. Classes, hit points, alignments. None of that anything like the way the real world works. What the game says is that regardless of game world, if you use D&D, people are pigeonholed by their choice of profession, get physically tougher the more experience they have, and are all psychically aligned with the good/evil and law/chaos spectra.

    D&D5 has tried to fix the second of those affirmations, equating hit points not only with toughness, but with cosmic importance. I personally like that, but is it too New School for the Old School? It’s a dramatic conceit in a game that many play in a simulationist (as opposed to narrativist) way.

    But when it comes to classes, various things have been tried to make them more reality-compliant.
    -Multiply the number of classes so that more roles in society are covered (but you’re still stuck in a job all your life)
    -Dual-classing (jack of two trades, master of less)
    -Thematic overlays that differentiate one member of a class from another more effectively than class multiplication (Kits and specializations)
    -The ability to switch gears mid-career (some wonky effects on reality there)
    -Leveling up to a super-class with experience, with thematic branching off (Prestige classes)

    I might be forgetting some, but which do you prefer?

    I do agree with you, Patrick, that part of the problem of classes is their tradition. At some point, the game gave us Clerics and Monks and those character types have been called that ever since, except they are too culture-specific to fit seamlessly in any given campaign world. The Cleric word is fine, but should indicate a priest more than a crusader. Similarly, the Monk should be a martial artist of some kind, or really, a Fighter specialized in hand to hand combat, not a Shaolin Monk. (When I was a kid, I just didn’t get that Friar Tuck was supposed to have all those crazy abilities.) I mean, they didn’t call the Paladin a Samurai, right?

    I really do think that when it comes to classes, Wizards should take a good honest look at what has gone before and ask themselves if they’re really just another class with the numbers filed off. I think Fighting/Magic/Faith/Subterfuge may be very well all you need, and then offer different builds, themes, kits, what have you, for each. And I think I’m being kind to spellcasters by separating Magic and Faith there.

    I guess I’m an old AD&D2nd GM, but I kind of liked what the game was doing back then. Your basic classes, with a number of secondary classes which you could overlay with Kits and Roles, and/or specialize by School, Sphere, and Deity. At that point, it seemed fairly easy to introduce a new school (like Wild or Elemental Magic), Kit (like Planewalker) or even secondary class (Dark Sun’s Gladiator) based on how the others were built. Easy to build for GMs and Players, and yet the possibility of publishing splatbooks which the gaming companies clearly love.

  9. Philo Pharynx says:

    @Patrick; A lot of players want to be able to play many different concepts. If you have narrowly focused classes, then you need to have dozens and dozens of classes to make people happy. If you have ways to customize classes then you can have fewer classes with more options.

    It also makes good sense to not have a completely separate class for a “Guy who fights with sword and shield” and a “Guy who fights with a two-handed weapon”. To me, both are fighters.

  10. I am not sure that there needs to be a specific class for every variation. Maybe when the line blurs too much would you need a specific class. The classic Fighter-Paladin-Cleric situation. It may depend on how WotC views advancement of the character. If you want to play a monk, could that not just be a theme? What if I want the thief-acrobat, like we had in the old UA. If they used a central XP chart for each core trope then you could simply make choices along the way and advance on that tropes chart. With a combination of skills, themes and backgrounds you can build any “class”. For that matter you could build any character you wanted. This is one of the things I always found lacking since the inclusion of the prestige classes. People complained about pre-generated characters, I complain about pre-destined character class.

  11. I agree with @Patrick: other than tradition (as stated by @Siskoid), why must we keep trying to shoehorn a skill system onto a class-based structure? Way back when, we played classes because we were 12 and our imaginations were limited to whatever fantasy fiction we had just read. Given new game theory /designs, correcting for past mistakes, and allowing for greater player input when designing a character “theme”/”template”, how about going the way of another classic (in this case, Runequest) and eliminating the artificial class concept? Beef up the skill system, new vistas suddenly open wide.

    Sword-swinging sorcerer? Check.

    Spell-lobbing warrior? Sure.

    Lock-picking, pocket-picking, sneaky, hidey rogue who can cast a few cantrips, work as a scout/spy (maybe even — *GASP!* — an assassin[!]), and plays backup lute in a friend’s band on the weekends? No problem.

    What makes D&D, well, “D&D”? Throw some dragons in some dungeons, mix in a healthy dose of other fantasy tropes, slap the label on it and, ta-da! “Dungeons & Dragons”. Squint and it might be “Swords & Wizardry”. Look out of the corner of your eye and it’s probably “Labyrinth Lord”. Blink fast and it could even be “Barbarians of Lemuria”. All fantasy pencil-and-paper RPGs; all capable of giving you a similar experience.

    Gary & Dave are dead. We all miss them, but moving on is part of life. Stop kibitzing and play.

  12. Classes are indeed artificial, which tends to be irritating to people who want realistic, flexible play, but they are really, really good for helping the game run smoothly. Classes make it really easy to give every player something to do, and are an easy shorthand for a character’s general power set.

    If I know that I’m going to be running a game for a party of two rogues and a wizard, I know exactly what I’m getting into, even though there will probably be some variation. If I’m playing a point-buy game with two skilled, high damage characters (who happen to resemble rogues), and one magic-using character (who happens to resemble a wizard), I’m going to have to take a good long look at the players’ character sheets before I know what I’m dealing with, and if I don’t read carefully enough (it’s happened) I might not realize that everybody has low hit points, and nobody can heal.

    Likewise, for a player joining a pre-existing game, it’s so much easier to understand “we need a cleric” than “We need someone who has healing magic, diplomacy, anti-undead magic, and melee combat.” Classes make it easy to find a niche and fill it. Point-buy systems, in my experience, inevitably result in parties full of characters who don’t complement each other’s weaknesses, and don’t synergize well. They also make it really easy for a new player to create a character who’s utterly useless. That’s exactly the kind of play experience that a lot of players prefer, but it’s not the audience that D&D has catered to in the past, and it doesn’t fit well with D&D’s role as most people’s introduction to RPGs.

  13. Jake Nelson says:

    Sometimes I wonder if I’m so weird… almost all my characters are multiclass, and it often feels like that’s some kind of aberration.

    In 4th, I was terribly frustrated as having started as a wizard, I took up cleric due to story reasons a few levels in, and it wasn’t until hybrid characters came along that I was able to make it even function mechanically, and it still was badly… off. The system simply couldn’t handle the concept of changing classes partway in due to story reasons.

    The lack of custom magic items makes my longest-played character essentially impossible to translate to 4th without vast amounts of handwave and guessing. (Warforged artificer, mostly made of magic items after a point, basically.)

    I really liked the progress 3.5 was making towards more freeform characters, and wish it had continued, instead of 4th’s “whatever you were when you left home defines you forever entirely” ideology. (I should note, I like 4th a great deal aside from those two very important points.)

    Not all of us play defined archetypes, and even when we do, they may change.

  14. Unwinder makes a good point about D&D: It’s most players’ gateway drug. If I want to play classless or freeform advancement, I’ll play something else. There are plenty of other sword&sorcery games out there if that’s the genre I really want to play.

    HOW did it become a gateway drug, aside from its market visibility (it’s essentially the Kleenex of RPGs)? It must in part be because its system makes certain things easy. It’s not the easiest system to run or play, nor are the campaign worlds easier to imagine than, say, those based on our modern world, so what’s the hook? It doesn’t make rules or setting easier, but I guess it makes imagining your characters easier. Classes make it easier for you and others to know what role you play mechanically and socially. A lot of video game RPGs use a similar conceit, so new players are likely to understand it.

    Whatever D&D becomes, it probably should retain entry-level facilities like class and alignment, even though both those things can be interpreted as crutches for role-playing. How many gamers have you ever asked about their characters only to get class-race-level and the only notion of personality being alignment? It’s both a quality and a defect, but it seems to work on a recruitment basis. Older players and GMs will experiment more, either moving to other games (like I have) – and I do think even hardened D&D gamers should try other games if only to learn new lessons they can bring back to their beloved D&D campaign – or experimenting with class concepts, dual-classing, etc. like many in this discussion.

    It’s why I advocate a streamlined approach for 5e classes that features a very basic class system (for new players) that can be easily built upon by more experienced players.

  15. Patrick says:

    @Philo mind my WFRP 2nd Ed book came with dozens of starting careers. So (you’d need lots of classes) isn’t necessarily an insurmountable obstacle.

    @siskoid classes serving as functional niches works well when classes are designed to fulfill valuable niches (like Roles in 4e) but when these niches are poorly defined it can be worse than useless. For a very minor example you mentioned diplomacy as something a Cleric brings to the party. The number of versions of D&D where all clerics are good at diplomacy is 0. (pre-3e few social rules beyond reaction rolls, relies on players; 3 and 4e cleric has diplomacy class skill and Charisma is a secondary attribute but may not be trained) I know I’ve made Clerics who were not particularly diplomatic in little or big D senses. One non-class game that handles roles very well is the Gumshoe space opera game Ashen Stars where crew position packages ensure you can fulfill a role, but aren’t your whole character.

    My actual suggestion for an introductory RPG is make character creation optional. Have lots of pre-made characters (at least 12) that are easy to tweak backstory and gender wise. Give options to customize in certain ways, and then give full chargen. I’m too lazy to dig up the link but Monte Cook had a thing about this on his site recently.

    In many ways the best approach is to come from a media liscense so that your pregens are literally characters people know. (cf Marvel Heroic Roleplaying, Doctor Who:Adventures in Time and Space, and the upcoming Atomic Robo RPG) for D&D famous fantasy characters with the serial numbers filed off might work, or might not.

    I don’t think alignment is helpful for new role players at all, but that’s a separate argument.

    @Jake Nelson to the degree I like class, I think I like it much better for character generation than advancement. I like advancement to have a relatively high degree of freedom. Or at least a more menu driven advancement like say Apocalypse World

  16. Unwinder brought up the Diplomacy bit, I don’t particularly agree that a Faith-driven world view makes one necessarily good at diplomacy (especially in a warrior-priest such as the Cleric as so often presented). But I do get your point that mechanically, classes must fulfill specific roles to be useful in party/adventure balancing. I’m afraid that in my experience, even rookies want to make their own characters even in licensed games (I’ve played tons of DC Heroes and Doctor Who with neophytes and can back that up). I think they’re useful as examples though, and to help guide character creation, that’s for sure.

    I am wary of any game that requires such a regimented balancing act however. If the presumption is that every story will need each of the core classes (by which I mean, the core mechanics of those classes, i.e. damage dealer/tank, spellcaster, healer, subterfuge), that limits the kind of adventures that can be told. The thief is a particular aberration that seemed (I’m talking early days) specifically designed for dungeon delving, which made the game dungeon heavy from the start. I understand it’s in the title, but…

    With most games – but again, this is coming from an experienced GM of 25 years, not from what should be a new edition’s sole target demographic – I will build the game/campaign around the core concepts the players want to play, or inversely, will tell them the campaign premise and they’ll make characters that fit in it. So for example, I might run a fighters-only military campaign, or goings-on at a magic college, or one of the fantasy settings on the day the magic dies. If D&D is my system of choice, I want to be able to do all that with a minimum of fuss, and somehow still grab a published adventure and insert it into my campaign without it being impossible for any given group of characters. It’s hard enough finding players without having to find players that enjoy each of the niches.

  17. Patrick says:

    @siskoid and yet the experience of the people involved with the Smallville and Marvel Heroic games is that people love playing the pregens. When I did some Doctor Who stuff some people wanted to make a companion, but the Doctor and Jack were popular.

  18. Good to know some groups are using them. I’ve handed of pregens to guest players in the past, and they enjoyed them. It just wasn’t the preferred way to do things in any of the groups I’ve been involved with for some reason, even when I’ve strongly suggested it.

  19. Matt F. says:

    Two concepts introduced/codified in 4e D&D (albeit lifted from MMOs and other RPGs) were the ROLE and the POWER SOURCE. To me, it all comes down to that. Everything else is just a theme or a build or a kit or whatever you want to call it. That’s where 4e dropped the ball, I think, because, to me, this simplification of things as belonging to roles and power sources would seem to make the ever-complicated multiclassing the easiest thing ever. You move up a level when you reach a certain amount of XP, which you then SPEND on improving skills, purchasing feats, improving attack bonus, whatever. Yeah. Spend the XP you earn. And with that XP you could also have the option of buying additional power sources or roles; i.e. if you’re a striker, you can also buy into being a controller at the cost of SLIGHTLY lowering skills or attributes in the long run. Don’t we already do this? Lower our hitpoints by multiclassing with wizard (3e) Lower hit points by not selecting Toughness as a feat, lower attack roles by not selecting an Epic Destiny with a stat increase? I guarantee I can make any fantasy class or theme or specialist or concept based on just those two categories: Power Source and Role. Anything from the Paladin to the Halfling Whistler, from the Swashbuckler to the Psychic Warrior to the Arcane Archer.

  20. Interesting. I do have a deep aversion with mechanics-driven games however (which is how GURPS lost me at 4th ed.). While building from role and power source sounds fine, as long as role is striker, healer, etc., it is completely flavorless to me. Yes, I can use these elements to craft something with literally ANY flavor, but as core game, it’s dull and doesn’t make me WANT to play. Would be fine if D&D was the only game in the world and I HAD to play it. I would them customize classes to my tastes. But it’s really not the only game in the world. I would want to play D&D for its feel or for its settings, but not for its buts and bolts, which role an power source reduce it to.

    However, if you dress it up with evocative language, it would provide the basic underpinnings I’ve been advocating. Possibly, your role in the STORY (as opposed to combat/adventuring) should be chosen first, followed by your role(s) and power source. A player shouldn’t want to be a “striker”, he should want to be a Fighter or a Wizard. Then when the GM asks what KIND of fighter he wants to be, he can decide that he wants to be a Tank powered by Inner Strength (or Chi) of whatever. Even games like WoW which essentially popularized the idea of tank-DPS-Buff-Healing etc. schemes made you choose from an evocative class that had different builds.

    Experienced players will houserule everything anyway, but the core game should perhaps present a baseline worldview complete with context for its mechanics.

  21. It seem to me that classes work best when they represent broad, strong archetypes. I think for a concept to merit its own class, it should meet at least one of these two criteria:

    1. Is it sufficiently different from existing classes (or is there a lot of overlap)? For example, Monk is fairly different from other classes, suggesting it should be its own class. OTOH, Paladin and Assassin are very similar to Fighter and Rogue, respectively, suggesting they would be good sub-classes or themes or prestige classes or whatever.

    2. Is it broad enough to cover multiple different sub-classes or themes or prestige classes or whatever? This one is somewhat more subjective. For example, I’d say there are a bunch of interesting variations on Paladin, but I see all Monks as more or less the same, and likewise all Assassins are pretty similar.

    So by my criteria, Monk should be its own class (due to criteria #1) and Paladin should be it’s own class (due to criteria #2), but Assassin should not (it fails both criteria). Obviously a lot of this is personal taste and is debatable, but I think you need good criteria to have such a debate.

  22. Patrick says:

    @Will Here’s a criteria I have. For a class to be narrow enough to be good, it should be something that there’s a stereotype for that a player can embrace or choose to go against. That’s my problem with Fighter, not-so-magic-fighting-dude is too vague. Now in 4e the fact that fighters (other than Slayers) were Defenders starts to focus us in and the choices you made for class features with “Weapon Master Fighters” sometimes helped. (That is a Battlerager is a non-magic version of a Berserker archetype, a Brawling Fighter is a brawly type, etc.) Knight obviously has various implications.

  23. Overall, I like that class continuum based on the Big 4. Quite sound. And I agree, there’s something lost with easily accessible “specialists” like assassins.

    Reminds me of the original Final Fantasy where you had to play your way through 95% of the game just to reach the “prestige classes” of Knight, Master or Ninja for the Fighter, Monk or Thief.

    Doesn’t translate well into sit-down RPGs though. I mean, it would be a long, painful wait – game-wise. Although, sure, story-wise, it’d be pretty cool and feel like you really earned something incredible!

    As for the Fighter comments:

    Honestly, I always found the class name “Fighter” a little too vague, and even a bit silly.

    What movie or book comes to mind immediately where a main character says or is described as a “fighter?”

    Now try that question with Warrior or Soldier. See the difference? Much stronger imagery, much more identifiable stereotype.

    Fighters should be called *Warriors* in D&D Next!

    And yes, I like the sound of Mage over Wizard too 🙂

  24. Stole the words right out of my keyboard, Tony. Fighter is about the same as Magic-User, a function, but not evocative of story.

    Thief, on the other hand, is TOO specific a word, begging the question as to why adventuring parties (and clerics!) would hang out with a dishonest man. I sometimes think D&D is a prisoner of its naming traditions.

  25. Philo Pharynx says:

    @Will “For example, I’d say there are a bunch of interesting variations on Paladin, but I see all Monks as more or less the same, and likewise all Assassins are pretty similar.”

    I see it the opposite way, only a couple of variant paladins, several assassins and lots of monk variants.

  26. Regardless of how you view and organize classes, they all should be available from day one when a new edition is finally physically published. My problem with 4e was that they made 3 PHBs, and if you wanted to play a monk or a psion (or a gnome) you had to wait. Yet they put compleately new stuff in the first book (dragonborn, warlord, tiefling, warlock). I’m wondering if it would just make more sense to publish all classes and races in a PHB, and shift all the ‘rules’ (combat, etc.) to a DMG. Because, really…does someone who isn’t DMing the game really need to know how that works? don’t they just need to know how to make a PC and play a PC? And you can’t teach ‘role-playing’ in a manual.
    I find the 4 core classes (like in the playtest) to be harshly limiting. I’ve NEVER played any of them in 25 years of playing and DMing the game. I like rangers, monks, sorcerers, psions and bards. I also think they are just as iconic as fighters wizards clerics and rogues. And I really hope they don’t make you be a fighter first then prestige class out into a monk or ranger later. That makes NO sense. Monks are often trained in monastaries, and rangers in the woods/wilds. Level 1 is apporpriate for that.
    Basically, I believe D&D should either go totally classless (like some videogames, eg: Fallout) or make ALL classes/races available to anyone who just buys ONE book.

  27. PS: my group did a classless mini campaign in 4e that worked. You just picked any powers you wanted ( 2 at will,1 encounter,1 daily) and any 3-4 class features at first level. We’re just friends, so we don’t care about RPGA/organized play balances, etc. No one was over or under powered, mainly because we play fair. D&D is cooprative, not competitive after all.

  28. Patrick says:

    @seti while 4e probably had too many classes by the end, if there are to be supplementary materials at all it would seem that new classes are an obvious choice. And indeed every version of D&D has released classes not in the initial materials.

  29. DarkplaneDM says:

    For me it’s all about flexibility. If I can have lots of options to make my character unique in story and mechanics, then I’m happy (creatively and strategically). It might be through multiple class options, it might be through specialization, or it might be through hybrid and multi-class options (which are my favorite).

    I think the revered cleric is just as specialized as an assassin. He’s a divine warrior that’s good at scaring the undead, has healing powers, wears heavy armor, and only uses blunt weapons. What the hell is that? I liked how 2nd ed had priests that were tailored to a specific god. And they gave you rules for creating new priest classes for custom deities. That was the sweet spot for me. A provided way to customize your character’s mechanics to fit his or her story. However it comes, I hope that’s available in 5e.

  30. TheMiddleThing says:

    The reason the four basic classes work well as core is because they approximate some of the most significant real-world social spheres of endeavor, such as commerce, politics/administration, religion/spirituality, knowledge, and leisure.

    The traditional FIGHTER is a man-at-arms, and thus fits under the political sphere (warfare is politics, that is, resolving differences through violence). The CLERIC is just a priest, and fits under the religion/spirituality sphere. The commerce sphere is rather mundane. Farmers, craftsman, and merchants aren’t very heroic or adventurous … unless you what you’re procuring, crafting, or selling is illegal. Thus the THIEF fits under the sphere of commerce. In D&D, the WIZARD fits (roughly) under the sphere of knowledge.
    Sphere:Politics(Warfare) = Fighter (Man-at-Arms)
    Sphere:Religion = Cleric (Priest)
    Sphere:Knowledge = Wizard (Ritualist)
    Sphere:Commerce = Thief (Rogue)

    The reason many of the subclasses seem to flow from these four is because the subclasses MUST fit within a social sphere, and interesting aspects of four major spheres are already represented by the base classes.

    All you really need to do to create the other classes is to define other parameters within a sphere, such as the legality of the character’s activities, whether his/her society is primitive (simple) or civilized (complex), and of course, in a fantasy world, whether the character is involved in the arcane.

    Thus the base classes become:
    Sphere:Politics(Warfare) + Civilized = Fighter (Man-at-Arms)
    Sphere:Religion + Arcane = Cleric (Priest)
    Sphere:Knowledge + Arcane = Wizard (Ritualist)
    Sphere:Commerce + Illegal = Thief (Rogue)

    The sub-classes can be defined as:
    Sphere:Politics(Warfare) + Primitive = Barbarian (Warrior)
    Sphere:Religion + Sphere:Politics(Warfare) = Paladin (Templar)
    Sphere:Religion + Primitive Arcane = Shaman

    Some spheres aren’t very adventerous, unless:
    Sphere:Leisure + Arcane = Bard

    Point is, you can build most classes by mixing REAL-WORLD spheres and REAL-WORLD social parameters like legality, with only one paramaeter, arcane, needed to create fantasy characters.

    Thus far, D&DNext is calling some of these ‘subclasses’ themes. Hopefully they’ll maintain discipline and keep it that way. If they do, there will be no need for multi-classing, which is generally just more excessive combat nonsense abused by optimizers and powergamers.

  31. Philo Pharynx says:

    @seti, Including all possible classes in the core book is impractical. It would mean you would have to extend the development time until all possible entries are finalized. It would mean that your player’s handbook would be huge. It would cause problems when somebody comes up with a great idea after the fact. It would limit the revenue of future books (for D&D, this annoys Hasbro. for smaller game companies, this limits their ability to stay afloat).

    Also sometimes you need to see how things work in the real world so you can make improvements on future revisions.


  1. […] very interesting article by Bartoneus over at Critical Hits on the topic of what classes should be in D&D? The relationships that each classes have with […]