I think what you're describing is fleshing out what, in my formulation, is called the reliability of the rules. Or, more fully, the reliability of the rules to achieve your known goals.
I think that's right.
I would describe the situation you present like this. The 'world' in an rpg is simply an imagined boardstate. It is a common feature of many rpgs that the only participant who is allowed to change the 'world' (ie the boardstate) is the GM. It is therefore self evident that any goals requiring a change in 'the world' can only be done by the GM - and so only the GM has agency to achieve them.
Up to this point, I want to say that the GM is the only one with
authority to make them. But I want to suspend - at this stage - the judgement as to whether other participants can use their authority to guide or control the GM's exercise of the GM's authority.
What happens in practice is that the sometimes the GM changes the boardstate in accordance with my wishes and then claims I exerted the agency. This is the heart of illusionism. Part of the same illusion is GMs who declare the boardstate independent of them, autonomous, with it's own causal and motive power. It's alive on its own. Utter nonsense when examined, and yet repeated again and again on these boards with complete seriousness.
I agree that this is something that happens often in practice.
But I don't think it covers the field. If there are
rules (formal or informal) that govern the GM's exercise of their authority over the board state, and if I - a player - can use
my authority to enliven one or more of those rules, then I can exercise control over the board state even though I don't have
authority over it.
(A side note: I think it may be because I work in law and legal theory that I emphasise this point. Because part of one ideal of the rule of law (not the only way of making sense of that ideal) is that while it is the
judge who has the authority to pronounce, a litigant is nevertheless able to exercise
their agency by activating the judge, because (i) the judge is bound by law, and thus (ii) activating them isn't just like opening Pandora's box, but (iii) is actually a way of generating effects that the litigant wishes to achieve but does not themself have the authority to bring about. Contrast, say, dobbing on a classmate to a teacher, which is much less reliable as a mode of exercising agency because the teacher's response is not very strongly rule-bound, making this much more of a Pandora's box situation.)
It is my thinking along these lines that makes me especially sensitive to claims that
the GM is never bound by any rules. And this is also why, a bit like a broken record, I keep going back to Gygax on successful adventures. Because he
clearly is assuming that the GM is bound by rules (and there are more illustrations of that assumption in his DMG, although perhaps a bit more opaque): because otherwise, none of the advice that he gives would make any sense at all!
And as I already alluded to, I think there is value in making it clear that Gygax excepted the GM was rule bound, because of the cultural and historical legacy. Of course Gygax never quite comes out and states the rules - this is a well-known gap in early RPG books - and some of the rules are rather informal. But they're nevertheless not hard to work out.
And while I'm rambling on about these things, here's another idea that - for me - is closely related to all of the above (and also to my comment that, at a certain degree of skill differential, the weak player may barely count as playing the game at all). And it also relates to a comment on an old RPG discussion site that you'd be pretty familiar with I think, that some RPGing texts tend to recycle the author's play experience as
material for new players' own play. Anyway, here's the idea:
The original Monster Manual is full of monsters that make no sense, and are frequently regarded as "gotchas": mimics, trappers, lurkers above, piercers, rot grubs, ear seekers, and the various oozes and moulds are the stand-outs. I have a conjecture about these monsters, though.
In the context of an evolving culture of play, at an actual table among actual participants who play together, the rules and heuristics that govern the way the GM exercises authority over the shared fiction can evolve over time in a way that is (at least roughly/loosely) knowable to all the participants. And in this context, those bonkers monsters don't have to be "gotchas": they are elements developed and deployed over time as part of a game play among the participants that becomes ever more intricate, ever more self-referential (like the "in jokes" and so on that become part of a group's culture), and that reflects the development of shared tropes, expectations, etc.
As well as monsters, the same thing can happen with traps and tricks: eg initially pit traps are a reasonable thing, but then over time it becomes acceptable, as part of a challenge for
these players at
this table, to include pit traps which open when the floor 5' in front of them is poked. Just as rot grubs step up the challenge of searching dead bodies, and ear seekers step up the challenge of listening at doors, so the baroque pit traps step up the challenge of prodding with poles.
How much fun this is overall is obviously going to vary with the participants, and likewise whether it is still fun when computer games can do at least some of it a lot better. And how many sessions it can sustain for an individual player is also going to vary, but probably be finite - at a certain point, say when you can get through ToH or some analogue thereof, you've effectively beaten this particular game.
When Gygax talks about the importance of letting knew players learn the game themselves, and not have it spoiled by old hands;
and that this needs to happen in a dungeon suitable for new players; I think he (at least in part) has this sort of thing in mind. That is, both (1) a player can't just begin the game in a dungeon full of the most baroque and gonzo elements and be expected to meaningfully exercise their agency, because no reasonable person is going to think of mimics and ear seekers and 10'-pole-defeating pit traps a priori - these need to be gradually learned and internalised as part of the table's heuristics and informal rules; and (2) being
taught those rules by the old hands, rather than picking them up via the play experience, is like being told the cheat codes for the game from the outset - it robs the experience of its fun.
Assuming what I've said above is at least in the right ballpark for how the game got going, it then shows how
ridiculous certain things are which have nevertheless gone on to be treated as quite standard. First is the treatment of the MM as a mere menu for GMs to choose from, with no discussion of
how to use the different elements inside it to build a shared play experience. Second, perhaps, is the presentation of an MM with those baroque/gonzo creatures at all, as opposed to a book (comparable, say, to the BW Monster Burner or Magic Burner) that tells a GM how to gradually increase the complexity of their game, riffing on the actual play experience at the table and the informal rules and heuristics that have developed. And this would also extend to tricks and traps.
Third, the idea that things like the MM, and dungeon design more generally, should be understood through some sort of "naturalistic" paradigm, rather than through the idea of emergent local rules and heuristics that step up the challenge to players over time, while still permitting them to use their agency (because they grasp the rules and heuristics, having been participants in the play, or at least in the local culture, that saw them emerge).
This has been a long series of paragraphs, covering quite a bit of territory. But I think it relates to the theme of agency by trying to identify the way in which certain approaches to RPGing, which have departed in fundamental ways from the original D&D experience even while retaining some surface-level resemblances, have really contributed to a reduction in expectations and experiences of agency in play.
Earlier, another poster asked why debates about agency are so frequent. My answer is because the predominant rpg playstyle is to give the GM all the agency, while denying that this is the case.
I agree with this. And to build on what I've said just above, I think there are key features that explain why there is little player agency:
*A strong commitment to the GM not being rule-bound;
*A reinforcement of that commitment that arises from an apparent hostility to the reality and importance of informal norms - expectations, heuristics, procedures, informal rules, etc;
*A corresponding insistence that the only constraint on the GM is their own conception of what makes sense in the fiction.
It might be a little bit of an oversimplification that it all went wrong when the MM was published in the mode of a
bestiary rather than
a tool for playing games - but for the reasons I've given above, I don't think that is entirely incorrect either.
I played AD&D with friends (but no GM) and used the random dungeon generator and random encounter tables by dungeon level, random treasure tables. The game worked totally fine. We had reliable processes to generate what we needed. We could assess our ammo, spells, hit points and make judgements whether to move forward another room based on what we might encounter and what we might gain. All a GM would have done in this game is introduce unreliability into all these processes and overwritten our agency with their own!
I think there is more to be said about the possible role of the GM, that is positive - consistently with what I've said above, a GM can introduce an element of intricacy into the puzzle-solving aspect of dungeon play which need not be agency defeating
provided the GM is bound in knowable ways that can become known via known ways.
And too be clear: "knowable ways that can become known via known ways" is a deliberate choice of words, modelled on the basic structure of the successful adventure. Eg the first time the GM includes some new baroque trick or trap or creature, it has to
bind the GM - via their notes, write-up etc - but may not be
known to the players - part of the puzzle is
coming to know how the new thing works. But the
ways of coming to know that must be known, or else it's just arbitrary and blind decision-making, as would be experienced by neophyte players thrown in at the deep end. The known ways in classic D&D include (say) using torches to test for fire vulnerability, using poles and ropes and pitons intelligently, etc.
When I GMed a session of White Plume Mountain a few years ago, the players - who are all RPGing old hands, and so familiar with the basic D&D tropes - found themselves surprised by some tricks, but they worked out solutions to some of them. I'm not going to say that it was the highlight of my RPGing experiences! But it was not un-agential play, because the puzzles were beatable because the ways in which the GM (ie me) was bound in respect of them were also knowable.
I would contrast this with some of my nadir experiences GMing RM, where the only heuristics available were "make a naturalistic decision", and these generated almost no constraints at all, and so there were not knowable ways in which I was bound, which could come to be known to the players via known ways. There were two such moments that I still remember, which really were low agency RPGing, and disappointing both for me as GM and the players. Thankfully after the second time I was able to develop new and better informal rules and heuristics, that the players also participated in making part of our game play, and subsequent experiences were much better.
Anyway, this has been a long reply but I hope not too tedious!
