When playing a class-based fantasy RPG, are classes diegetic for you?

[Lanefan]

Yes, it certainly is. I say that because while some elements of class might be 100% diegetic, no class is entirely so. It's not aesthetics, its design. If someone wants to show a class that's 100% diegetic they should jump in, but the burden is very much on them to demonstarte that it's even possibly the case.

That's true of pretty much every element in the game that has to be abstracted so as to work at the table, not just class. The combat roll to hit isn't diegetic - the character isn't rolling dice - but what that roll abstracts is completely in-character. Same goes for class and (often) level: what's abstracted to the table are a bunch of numbers and abilities, with that abstraction trying its best (one hopes) to reflect what the actual characters experience.

 

[Fenris-77]

Still seems like your position is rather binary on the subject.

You say in reply to a non-binary reply on my part. You're a funny fellow.

 

[Sepulchrave II]

For all casters, level - measured as degree of competence - can very easily be an in-character thing, as delineated by the best spell(s) you can cast.

For some martials - i.e anything that might use a belt-colour or similar competence measure - level is baked in from the start, measured in-character as belt colour rather than by number.

For some other martial classes it can get a bit fuzzy. Simple sparring can determine who is better than who between a few individuals but anything more universal is tough.

And in any setting or system where training is required in order to advance it becomes much more obvious for all classes: how many training courses have you had?

This is just bizarre to me.

The selective reification of class (and now, level), while leaving the rest of the mechanical scaffolding (hp, saves, whatever) firmly on the players' side of the screen, is simply unprincipled. You're either committed to the full absurdity of a world where people walk around knowing their AC and hit points, or you accept that all of it - class included - is a modelling apparatus, not a description of the world.

Even if we grant that the fiction contains something that class is trying to represent - warriors, rogues, mages - the class itself is not that thing. Between the mechanical abstraction ("Fighter") and the imagined person swinging a sword in the fiction, there must be an intermediate concept: the idealized Fighter, a kind of Platonic form that the class is supposedly tracking. But this phantom third thing does a lot of suspicious work. It silently absorbs all the weirdness of the mechanics (why can a Fighter do exactly this many attacks? why do their abilities improve at these specific thresholds?) and launders them into apparent naturalness. Once you name the intermediate concept, the argument for class-realism collapses into an argument about that phantom instead, and now you have to explain what it is, where it lives in the fiction, and why the mechanical class maps onto it so neatly. The fiction contains warriors. The rules contain Fighters. The idealized Fighter is just a conceptual placeholder people use to avoid confronting the gap between them.

 

[pemerton]

It silently absorbs all the weirdness of the mechanics (why can a Fighter do exactly this many attacks? why do their abilities improve at these specific thresholds?)

I think a system of rather abstract purely mechanical devices - the distinction between to hit and damage, the roll to hit that doesn't correspond to any particular action in the fiction (where the attack roll or rolls sum up 6 seconds of simultaneous action), etc - is particularly inapt to be reified in the way that your post is critical of.

 

[bsss]

Look again - he specifically references "climb walls" as well as stealthy and-or graceful movement, making it obvious he's after more than just a skill focus.

I noticed, but I think this only weakens your example. What structure exists in the fiction that explains why these two concepts combined are "more" than a skill focus or that this is best or only achieved by becoming a class (which drags in the ideas of levels in a class in order to take them)? If someone walked into the scene and demonstrated aptitude in both climbing walls and stealthy and graceful movement, would people go "oh wow, a rogue, clearly" [Rogue class] and know they also have a bonus to stabbing people in the shadows [Sneak Attack]? Could they then have a climbing challenge for sport, and see that this interloper climbs as fast as they walk [Climb Speed = Speed] and ergo must be a thief [Thief Archetype, Second-Story Work] and also a rogue of some experience [Third Level Rogue] to have gained the access to those features [Roguish Archetype]? And from that, conclude they must, as a matter of course, also be able to quickly dash, disengage, or hide when in a fight [Cunning Action, Second Level Rogue]?

If you want the actual terms such as class names and level numbers to be consistently used in order to make those concepts in-character, that might be asking a bit much.

I perhaps don't need the terms exactly used in-world, but I do submit that my example above needs to be as logical a situation in world as it is in game terms, for the whole idea of a class to be considered diegetic. And that as @Fenris-77 and @Sepulchrave II have put far more eloquently me than me, that these concepts cannot be selectively ignored or discarded, else you are not talking about class diegesis, but rather just facades of elements of class pieces.

 

[Sepulchrave II]

A few final thoughts about this whole question, FWIW:

I’ve suggested that a Phantom Ideal Fighter underpins arguments for class-realism.

Aside from what I consider the most fundamental error - treating an abstraction as though it were a concrete thing - we know that "Fighter" is a category created by game designers for play purposes. The reification move takes this category and grants it independent existence - first in the player’s head as a design concept, then into the fiction itself. But the class “Fighter” is a sorting device, not a natural kind.

If you accept the Phantom Ideal Fighter as a real thing in the fiction, you immediately face the problem of why the boundaries fall where they do. Why is the Paladin a separate ideal from the Fighter? Why does the Ranger occupy a distinct ontological category? These divisions are transparently the result of design decisions - Gygax needed the classes to feel meaningfully different at the table, to give players distinct play experiences. There is no natural joint in fictional reality at which "Fighter" ends and "Ranger" begins. The phantom ideal smuggles in the assumption that these arbitrary design seams correspond to genuine divisions in the world.

The Phantom Ideal Fighter is invoked to bridge the gap between the mechanical class and the fictional warrior. But it doesn't actually bridge that gap, it just inserts a third thing that has the same gap on both sides. You still need to explain how the mechanical Fighter maps onto the Ideal, and how the Ideal maps onto the fiction. You've doubled the explanatory burden without discharging any of it. This is structurally identical to the homunculus fallacy in philosophy of mind: explaining vision by positing a little man inside the head who watches the images, which just pushes the problem back one step and generates an infinite regress.

To take the Phantom Ideal Fighter seriously as an ideal is to commit to a fairly heavy piece of metaphysics: something like Platonic forms. There exists, in some sense, the Form of the Fighter, of which actual fighters in the fiction are imperfect instances, and of which the mechanical class is a representation. This is an enormous philosophical commitment to smuggle in through the back door of a game rule. And, really, it's Platonism done badly: real Platonic forms are arrived at by reasoning about essential natures, not by reading a class description that was written to differentiate play styles.

The Phantom Ideal implies that there is some essential Fighter-nature that all fighters share: some cluster of properties that makes something really a Fighter as opposed to merely fighter-adjacent. But try to specify what that essence is without just redescribing the mechanics, and it immediately evaporates. Is it martial training? Barbarians have that. Weapon proficiency? So do Paladins and Rangers. Lack of magic? Not in most editions. The supposed essential nature of the Fighter turns out to be either empty or just a redescription of the mechanical chassis, which proves the point: the "ideal" has no independent content. It's a shadow cast by the rules, not a light source.

There's a peculiar asymmetry at work. People who claim class is real in the fiction might say "my Paladin wouldn't do that," using the class as a constraint on fictional behaviour. But they rarely push it the other way: they don't argue that because their fictional warrior is brave, honourable, and skilled, the Paladin class must have a Courage stat. The Phantom Ideal is used selectively: to import mechanical constraints into the fiction, while the fiction's own logic is never allowed to revise the mechanics in return. This asymmetry reveals that the mechanics are doing all the real work, and the "ideal" is just the name given to the direction of flow.

Perhaps most damningly: the Phantom Ideal Fighter cannot be described in purely fictional terms. Ask someone who believes in it to tell you what the Ideal Fighter is as a person in the world: their training, their social role, their psychology, their history. The answer must either be something so generic as to be trivial (a person who fights well), or it translates back into mechanics (well, they'd have high Strength and Constitution, they'd be proficient with most weapons...). The moment you try to look directly at the Phantom, it either disappears into banality or reveals itself to be the mechanics wearing a fictional disguise.

I don’t think there’s a “right” way to play D&D, but I’d consider myself quite old-school in many of my sensibilities. As far as I know, Gygax never seems to have explicitly addressed the diegetic status of class - whether it "exists" in the fiction - as a philosophical question. That particular framing is really a later RPG-theory concern and Gygax predated that conversation (and largely showed no interest in it). What he did articulate - and quite forcefully - was something adjacent. In a long piece in Dragon magazine issue 16 (1978), he argued that D&D was a game first and foremost, not a simulation, and that the mania for "realism" that dominated the hobby in the late 1970s was a category error, worshipping what he called the "false deity" of realism. His core claim was that "as a game must first and foremost be fun, it needs no claim to 'realism' to justify its existence."

Gygax was defending mechanics against the charge that they were insufficiently realistic, which implicitly treats mechanics as modelling tools, not as descriptions of in-world reality. In the Gygaxian view, class, hit points, AC, and so on are all just game apparatus; tools for making play fun. He was hostile to the conflation of game mechanics with simulation of reality, and he seemed to think of class primarily as a player-facing tool rather than a feature of the imagined world.

The irony is that the "class is real" position is in some ways a byproduct of players taking D&D more seriously as a fictional world than Gygax himself apparently intended.

 

[Not a Decepticon]

No, classes exist as a way to mechanically reflect the story and world on the table, you should NEVER build your world or story around game rules - that's how we got Kender, after all.

 

[pemerton]

@bsss, I take your comments to be in the same general vein as mine from upthread:

The reason I think they're not diegetic is because a class establishes a bundle of abilities as markers of being a particular sort of person/protagonist; but in the fiction, there may not be any such type of person as a saliently existing thing.

You've framed your post more as a question and a requirement:

If someone walked into the scene and demonstrated aptitude in both climbing walls and stealthy and graceful movement, would people go "oh wow, a rogue, clearly" [Rogue class] and know they also have a bonus to stabbing people in the shadows [Sneak Attack]? Could they then have a climbing challenge for sport, and see that this interloper climbs as fast as they walk [Climb Speed = Speed] and ergo must be a thief [Thief Archetype, Second-Story Work] and also a rogue of some experience [Third Level Rogue] to have gained the access to those features [Roguish Archetype]? And from that, conclude they must, as a matter of course, also be able to quickly dash, disengage, or hide when in a fight [Cunning Action, Second Level Rogue]?

<snip>

my example above needs to be as logical a situation in world as it is in game terms, for the whole idea of a class to be considered diegetic.

But if I construe your question as at least somewhat rhetorical, than our views converge.

we know that "Fighter" is a category created by game designers for play purposes. The reification move takes this category and grants it independent existence - first in the player’s head as a design concept, then into the fiction itself. But the class “Fighter” is a sorting device, not a natural kind.

If you accept the Phantom Ideal Fighter as a real thing in the fiction, you immediately face the problem of why the boundaries fall where they do. Why is the Paladin a separate ideal from the Fighter? Why does the Ranger occupy a distinct ontological category? These divisions are transparently the result of design decisions - Gygax needed the classes to feel meaningfully different at the table, to give players distinct play experiences. There is no natural joint in fictional reality at which "Fighter" ends and "Ranger" begins.

<snip>

The Phantom Ideal implies that there is some essential Fighter-nature that all fighters share: some cluster of properties that makes something really a Fighter as opposed to merely fighter-adjacent. But try to specify what that essence is without just redescribing the mechanics, and it immediately evaporates. Is it martial training? Barbarians have that. Weapon proficiency? So do Paladins and Rangers. Lack of magic? Not in most editions. The supposed essential nature of the Fighter turns out to be either empty or just a redescription of the mechanical chassis, which proves the point: the "ideal" has no independent content. It's a shadow cast by the rules, not a light source.

<snip>

Perhaps most damningly: the Phantom Ideal Fighter cannot be described in purely fictional terms. Ask someone who believes in it to tell you what the Ideal Fighter is as a person in the world: their training, their social role, their psychology, their history. The answer must either be something so generic as to be trivial (a person who fights well), or it translates back into mechanics (well, they'd have high Strength and Constitution, they'd be proficient with most weapons...). The moment you try to look directly at the Phantom, it either disappears into banality or reveals itself to be the mechanics wearing a fictional disguise.

I think this is also in the same general vein as my view that I posted upthread: I referred to "saliently existing things" rather than "natural kinds", but I think the point being made is much the same.

As far as I know, Gygax never seems to have explicitly addressed the diegetic status of class - whether it "exists" in the fiction - as a philosophical question. That particular framing is really a later RPG-theory concern and Gygax predated that conversation (and largely showed no interest in it).

<snip>

He was hostile to the conflation of game mechanics with simulation of reality, and he seemed to think of class primarily as a player-facing tool rather than a feature of the imagined world.

The irony is that the "class is real" position is in some ways a byproduct of players taking D&D more seriously as a fictional world than Gygax himself apparently intended.

I largely agree; but I also think that Gygax muddied the waters a bit by using the player-facing classes to build NPCs. We see this in the MM entry for "Men", and in the Appendix C encounter tables, and at a few other points also in the DMG.

Once classes are being used to build NPCs, the idea that these are, in the world of the fiction, salient/natural kinds does get a bit more heft to it.

This is why I think that the 4e approach is superior. (And more generally, I regard 4e as taking the Gygaxian approach to mechanics - AC, hp, to hit rolls, saving throws, etc - and taking that approach closer to its logical conclusion than Gygax did.)

 

[Sepulchrave II]

I largely agree; but I also think that Gygax muddied the waters a bit by using the player-facing classes to build NPCs. We see this in the MM entry for "Men", and in the Appendix C encounter tables, and at a few other points also in the DMG.

Once classes are being used to build NPCs, the idea that these are, in the world of the fiction, salient/natural kinds does get a bit more heft to it.

This is a fair point, although as someone who generally advocates for the interoperability of PC and NPC mechanics, I’m not of the opinion that Fighters are real.

 

[GrimCo]

I voted no, but it's mostly based on D&D since it's only class based ttrpg i play regularly. In Dark Heresy or WH Fantasy, my answer would be yes, cause classes are real carrer paths in the world. Guardsman, Arbiters, Tech priest, those are in game world occupations.

In D&D, specially in 5e, i view classes as packed preselected set of skills and abilities that you get as you level up. Class defines what character can do, not what his position is in the world. My fighter/wizard is noble knight, rogue is traveling minstrel, bard is charismatic diplomat. I pick concept, think what skillset would that character need, then pick appropriate class or mix of classes to flesh out that concept.