OK. So reposting it:
Purist-for-system is a term that I learned from
this essay. It describes an approach to RPGing.
The approach is exemplified by RPGs like RM, RQ, Champions/HERO, GURPS, C&S, perhaps Classic Traveller. I mention these systems because familiarity with them helps make sense of the approach.
RM, RQ and C&S are all FRPGs that try and cover the same general sort of thing as D&D, but they all differ from D&D in some fundamental ways. I'm not as familiar with C&S as the other two, and so will draw mostly on RM and RQ to outline the differences:
*They eschew classes as the core of PC build, in favour of skill-based progression (RQ has no classes at all; in RM and I think also in C&S, classes serve as skill-packages or skill-cost determiners);
*They eschew hp-based combat in favour of hit location and crit systems, with active defence/parry broken out from armour protection, and armour acting as damage reduction;
*They eschew Vancian casting in favour of some sort of points-based spell system.
On top of these mechanical differences, the all emphasise "seriousness" in tone and setting (in contrast to some of the more "funhouse" aspects of, say, White Plume Mountain or Castle Amber or even Tomb of Horrors).
What's going on with these differences?
At their core, the resolution processes of these RPGs - that is, the methods used during play to work out what happens when a player declares that their PC does something - have two properties:
*They generate answers to the question "what happens next?" without the participants (players and GM) having to make decisions on the way through;
*As the resolution process is worked through, it generates knowable fiction at every point.
Here's a clear example of a RPG resolution process that doesn't meet the first criterion: in Burning Wheel, if a player fails a roll, the GM has to decide what happens next, having regard to (i) the task the PC was attempting and (ii) the player's stated intent for that task and (iii) the player-authored PC Belief that is at stake in the current fictional situation. It's metagame all the way down! RM, RQ and similar games eschew this.
Notice that D&D's core combat resolution engine, and core spellcasting rules, mostly satisfy this first criterion. This shows their very close connection to wargaming procedures. (Notice, also, that 4e's skill challenges don't - that's one reason they're controversial!)
But consider the second criterion. D&D's combat procedures fail this test (putting to one sie the more outre interpretations of them canvassed upthread by
@Pedantic). When an attack roll is made, and resolved, and (if it hits) hp damage is applied to the target, we don't know what happened in the fiction. Was it a lucky attack that caused a small injury? Was it a mighty blow that was parried, but the parry shook up the parrier? Or something else? We can't know that
just from the resolution procedure. Someone - typically the GM - would have to make it up!
Whereas RM and RQ answer these questions via the resolution procedure without anyone having to make anything up: we know whether the blow was landed skilfully or not, we know whether or not it was parried, or blocked by a shield; if an injury occurs, we know where it is and how severe it is.
In the case of spellcasting, the contrast with D&D is less clear but there's a general sense, in the purist-for-system ethos, that Vancian memorisation in it's D&D style is hard to take seriously, but a points-based system enables us to similarly envisage what is going on at the moment of casting: the spell user is calling on their inner energy (kind-of like Ged or Dr Strange) and shaping the magic to try and impose their will on it. It's not a coincidence that these games tend to require a roll for spell casting, rather than having it succeed automatically as it does in D&D.
This is the essence of purist-for-system RPGing. But this ethos tends to bleed from the resolution process to other parts of the game. For instance, because the resolution process tells us what is happening at each stage in a knowable way that doesn't require additional authorship, it is natural to think that the resolution process
tells us what is possible in the fiction ("rules as physics"). This then has implications even when the procedures are not being used - eg when the GM is authoring the setting, or framing a scene/establishing a situation, they should do so having regard to what would be possible given the resolution mechanics. (Contrast D&D, which at least in its classic form that these games were reacting against cheerfully permitted all sorts of elements in the fiction that couldn't be explained in terms of the extant PC build and resolution rules.)
Furthermore, the differences I've described - more "realistic" combat, less reliable spell casting - make classic D&D dungeon-crawling less feasible as a play activity (because fighting is less feasible, and spells no longer serve the same player-side resource function that they do in classic D&D). At the same time, rich and robust skill systems encourage players to focus their activities elsewhere.
The things I've described in the previous two paragraphs, in turn, are what feeds into the sense of "seriousness" about the setting. And that iterates back into expectations about what sort of fiction the resolution system should generate, so the upshot tends to be a type of "groundedness" or even "grittiness".
To reiterate what I've posted a bit upthread, you can see the same purist-for-sytem ethos in AD&D players who change the game in certain key ways:
*Introducing a wound/vitality type system to make injury more realistic, and make it clearer - in the fiction - which "hits" are really just wearing a foe down vs which are the actual injury-causing blows;
*Adjusting the falling damage rules to avoid the situation where a player knows their character can't die from a fall;
*Adjusting the archery rules so that a single crossbow shot that is not dodged or deflected can be fatal, even if the PC has more than (depending on edition) 6, 8 or 10 hp;
*Replacing Vancian memorisation with some sort of spell-point system, and/or adding in general spell failure/mis-cast rules.
These changes (i) enable tighter connection between resolution processes and knowable fiction, and (ii) make classic and "funhouse" dungeon crawling less viable as a focus of play. They increase the sense of "seriousness", just like RM and RQ and C&S and so on.
I think it's harder to graft an effective skill system onto AD&D (apart from anything else it collides a bit with thief, ranger and bard class design) but it's not impossible to have at least a bit of a go at it, as the original OA shows.
Anyway, that's the longer version of an explanation of what I think is going on in purist-for-system RPGing, and how action resolution procedures are at its core, but generate effects and an ethos that ramifies into other parts of the game. (It also explains why I think that purist-for-system RPGing is
not connected to "world simulation" in the sense of that phrase that
@AbdulAlhazred is criticising. RM and RQ and C&S GMs don't pretend that they can model a whole world using game mechanics. Rather, they approach their GMing task, and the relationship between the fiction that they author and the mechanics of their RPGs, in accordance with the ethos that I've tried to describe.)
