D&D 4E Pemertonian Scene-Framing; A Good Approach to D&D 4e

Not open for further replies.


I'm also rather baffled as to what the choice to replace DMing with published adventures and encounter tables has to do with the actual mechanics of the game.
I don't really understand your reference to "replacing DMing". But on mechanics I have a strong view: different mechanics definitely suit different approaches to play.

Here are three examples, all relevant to 4e:

In many fantasy RPGs, including D&D, fighting, and therefore healing, is a recurrent element in action resolution. The more that a system mandates keeping close track of time outside of the resolution of an encounter, in order to determine the consequences of injury and the healing of it, the harder it makes it to run a scene-framing game. This is because those system imperatives distract attention at the table away from the emotional and tactical challenges of the scenes, towards and onto the minutiae of who is bandaging whose wounds with what, and how long that takes, and whether the Heal roll is a success, etc.

This has never been as big an issue in D&D, with its hit point system, as in games like Runequest and Rolemaster. But 4e, with its short rest mechanic, makes it a complete non-issue.

Effect durations
This is similar to the point about healing. When effect durations are measure in units like 1 minute per level or 10 mintues per level (very common in pre-4e versions of D&D) then adjudicating those effects requires careful tracking of time, especially but not only by the GM. Which takes the focus of play off the scenes, and onto the transition periods between scenes. Which elevates exploration over engaging scenes as an aspect of play.

4e has very few durations like this (maybe some rituals?).

Detection spells
The detection spell - Detect Evil, Detect Magic, ESP, Locate Object - is a big part of classic dungeon-crawling D&D play. As well as spells, some classes have detection abilities, and they are also common properties of intelligent swords.

As Lewis Pulsipher in particular emphasised in some of his early White Dwarf articles, if the GM is to be fair to players who have access to these spells then the GM has to map out, in advance, those areas of the dungeon that the PCs might explore by use of them. So then the cautious player whose PC detects first gets the benefit of that; and the rash player whose PC just blunders in gets the consequences of his/her rashness.

That level of preparation is pretty much the opposite of a scene-framing approach, which is all about placing and modulating challenges and rewards in resopnse to the ongoing dynamics of play, rather than laying it all out in advance.

4e is the first version of D&D to basically have no detection magic of this sort. (There is the Arcana skill for detecing magic through barriers, but because it's a skill check that doesn't involve resource consumption, the GM can treat it as a trigger to make something up without ripping off the player.)​

Here is a different example, that (for me, at least, based on my experiences) captures a key difference between Rolemaster and Runequest:

Both Rolemaster and Runequest resolve melee combat as a type of contest between attack and parry. But in RQ, the attack and parry are distinct skill bonuses. Whereas in RM, the PC has a single combat skill pool, and allocates it between attack and parry on a round-by-round basis.

One result of this is that the RM player has a lot of freedom, from round to round, to choose the level of risk/boldness for his/her PC. This makes it easier to drift Rolemaster combat resolution in a non-simulationist direction, as players make choices about risk/reward that don't just reflect the PC's point of view, but the player's emotional investment in the scene.

There are a number of other ways in which Rolemaster differs from Runequest by giving the player the opportunity to metagame decisions involved in both PC building and action resolution, all of which support this possibility of non-simulationist drift.​

4e has even more metagameable action resolution than Rolemaster, which - together with other features, some of which I noted above - make it especially well-suited for a scene-framing approach.

'90s games of the White Wolf style (eg Vampire, Dream Pod 9's Heavy Gear, or even Call of Cthulu*, arguably) are different, but seem to lack the open scene resolution that determines the next scene; they focus on leading the PCs through a pre-written story - pre-mapped scenes.
The easiest way to understand The Forge, I think, is as a massive reaction against Storyteller style, which has all this "story gaming" rhetoric but in practice involves railroading based on massive GM force. (It sounds as if [MENTION=87792]Neonchameleon[/MENTION] may have run them in a different fashion.)

I don't have a lot of personal experience with the White Wolf style games, but definitely experienced this phenomenon - GM force and railroading under the guise of "story" - in 1990s 2nd ed AD&D play. Of published 2nd ed modules, I would pick out Planescape's Dead Gods as a particularly egregious example.

I have a lot of games, but the only one that largely meets my OP definition in its description of how to play is Ron Edwards' Sorceror and Sword.
It doesn't surprise me that Edwards spells it out.

It differs though in that the three-stage scene-framing system it describes is couched within heavy 'Dramatic Premise' Narrativism. Pemerton showed me it could be used in a much lighter, non-Narrativist play style.
I have never read a decent description of the process prior to reading Mr P. I certainly did not read a decent description of the process in the 4e rulebooks, or in those of any other RPG I own, other than Edwards in S&S. And Edwards had so much baggage in there with it (actually harmful baggage, IME) that it would never have occurred to me from reading Edwards that this was a technique to use in D&D.
First, thanks again for the kind words.

Second, for me the epistemic process was the opposite. I discovered The Forge more-or-less randomly in 2004 (I can't remember what link or Google search took me there) and found the GNS essays hugely interesting. (And had no prior knowledge of the Usenet discussions and analysis.) And naturally enough they made me think about my own playstyle. At that time I was GMing Rolemaster, and had been doing so for many years, and I was easily able to identify it as a purist-for-system simulation system (= process simulation). But what I and my group were doing with it seemed a bit different from sheer process simulation and world exploration: it seemed to have more in common with the vanilla narrativism that Edwards described. In particular, morality in our game emerged out of play and mostly at the metagame level of player decision and response, rather than ingame as part of the fiction. (I have a long time hatred of mechanical alignment!) And a lot of my play approximated more towards No Myth (improvised NPCs, locations etc) than heavy pre-prep, and that seemed to be a strength rather than a weakness.

So reading those essays, plus other posts, blogs etc, plus starting to look at some of the games Edwards referenced (Maelstrom Storytelling, HeroWars/Quest, etc), got me thinking more theoretically about my game and the techniques I was using. And I came to understand the Forge style in terms of my own play, rather than encountering it externally and not noticing it could be relevant to my own (very non-avant garde) fantasy RPGing.

This meant that when 4e started to be revealed by WotC, and some of its key features started to be revealed, I felt I had a fairly good handle on what was motivating the designers and what sort of play their game was meant to support. And therefore was pretty sure it would be a game I would enjoy - a level of mechanical crunch comparable to Rolemaster (for no very sensible reasons my group is pretty crunch-loving), but action resolution mechanics that would better support my preferred approach. (RM's PC build rules are pretty good for light fantasy narrativism play, but quite a bit of its action resolution is not.)

It may well be that Pemertonian scene framing is actually a completely routine mode of play in the sort of games talked about in the RPGnet General forum
As I posted in a thread we were both in late last year, I'm surprised that it doesn't seem to be more common in D&D play. I really stumbled into it when I started GMing Oriental Adventures in 1986/87. I knew that I didn't really enjoy, and was also quite bad at, designing and adjudicating dungeons in the Pulsipherian/Gygaxian style. The difference that OA made was that the PCs had pretty clear inbuilt hooks (honour, family, etc etc) and so did the monsters (the Celestial Bureaucracy, etc), so it made it easy to build and adjudicate fun and engaging encounters on the fly.

After that I ran a two-person thief game in a similar way - of all the AD&D archtypes, I think the thief has the easiest inbuilt hooks (which in my view also explains some of the notorious problems of thieves in dungeon exploration, because it means having to ignore those hooks). And then I strated my series of long-running RM games, which is also how I met my current group.

I guess maybe I'm not explaining well what I find important about what Pemerton describes, how it's different from what I did before, or how it's different from what most GMing advice advocates doing. I know there is a difference, I know it has affected my games - and made my recent 4e games much more enjoyable - but it must be fairly subtle.
I agree that there is a real difference. My path has been a bit different from yours, but since I've become more self-conscious about my techniques and have deliberately cultivated some and changed others, I've felt the difference too.

EDIT: One example of a deliberate change in technique - being a lot more upfront about stakes, for instance by table-talking with the players, and by using many fewer secret notes/one-on-one reveals and instead doing many reveals in front of the whole group even though only one or two PCs would know - thus setting up an emotional tension between what the players know and what their PCs know and can do about it. And giving them clear options for pushing the game forward to resolve those tensions. (Having been doing this for several years now, I've recently discovered that Robin Laws talks about this very technique in his "On the Literary Edge" essay in Over the Edge.)

Also, for what it's worth your explanations make sense to me, but I'm probably not the best test audience!
Last edited:

log in or register to remove this ad


I feel that this thread would strongly benefit from a very simple example illustrating the two systems.
Example 1
This example is made up, but is very close to some real play from my 4e game.

The PCs are investigating an old catacomb. One of them is a plading of the Raven Queen. The player of that PC says "I'm looking out for any signs of Orcus infestation, and trying to sense if his evil influence is present".

In a simulationist game, the GM consults his/her notes, or perhaps rolls an encounter check. In a scene-framing game, the default answer to the player's question is "You see a niche with a statute in it. It's of Orcus." Or, perhaps,"Yes, he's here. [Roll d10] Take 5 psychic damage as the sense of evil ovewhelms you!"

What does the statue mean? What is the focus of Orcus's malign presence? That's to be worked out by the GM and players in the course of play: the little narration I described has framed the scene (it's a non-combat one involving the presence of Orcus in the catacomb) and now it's up to the players to engage it via their PCs.

Example 2
This example is from my 4e game, and is part of this post. It is a combat variant on my first example.

The PCs stumbled upon an ancient underground temple to Orcus. . . .

As GM, it hadn't occurred to me to place an Orcus temple until one of the players said that his PC spent time going around town trying to find out where the cultists had come from and what they were up to. But I found a suitable poster map in Death's Reach (the one with the statues, pillars and altar) and had some monsters statted up - immoliths and other demons.

After clearing out the demons and dealing with the Altar of Zealotry in the temple (that had fun dominating the dwarf fighter), the PCs opened the back door behind the altar and could see Moria-like stairs descending further into the depths (the map is from one of the 4e Dungeon adventures, Siege of Bordrin's Watch). As they were checking out the stairs they were attacked by a nightwalker and its bodak servants. The idea of putting the stairs there, and using a nightwalker as a balrog substitute, was something I came up with in the time between sessions. But it was easy to narrate a weakening of the barrier between world and Shadowfell in the vicinity of an ancient Orcus temple. This encounter was quite a bit of fun, as a couple of PCs got knocked off the stairs (but, being 17th level, survived), and I got to use the bodaks' Death Gaze successfully at least once (maybe twice) and also Finger of Death from the nightwalker.

Once the PCs had beaten the nightwalker and bodaks, and were searching around, one of them tried to sense if there was any more shadow energy leeching through. I said that there was, and expected the PCs to try to seal the breach. But instead they took up positions and prepared for combat - so I had a dracolich attended by lost souls (levelled-up wraith figments) come through the barrier. This was another interesting combat - I used the MV rather than the MM dracolich, which included more domination of the fighter - and I got to reuse my poster map!​

Notice how the framing, and the necessary backstory to support it (eg the existence of the temple, the weakening of the barrier betwee world and Shadowfell) is done mostly on the fly in response to cues from the players - I am following their hooks rather than them following mine.

Example 3
This is another example from the same post.

As the PCs continue through the Underdark tunnels, I described them coming to a cleft in the floor, and got them to describe how they would cross it. The drow sorcerer indicated that he would first fly over (using 16th level At Will Dominant Winds) and then . . . before he could finish, I launched into my beholder encounter, which I had designed inspired by this image (which is the cover art from Dungeonscape, I think):


I'm not sure exactly what the artist intended, but to me it looks as if the central beholder is hovering over a chasm, with uneven rocky surfaces leading up to it (archer on one side, flaming sword guy on the other). I drew up my map similiarly, including with the side tunnel (behind the tiefling) which on my version ran down into the chasm, and the columns, stalactites, etc.

I didn't use four beholders, only 2 - an eye tyrant (MV version) and an eye of flame advanced to 17th level and MM3-ed for damage. And also a 15th level roper from MV, introduced on a whim when the player of the wizard asked, before taking cover behind a column, if it looked suspicious. (Response to result of 28 on the Perception check before adding the +2 bonus for knowing what he is looking for - "Yes, yes it does!")​

This encounter was preprepared, but notice two things: (1) the travel through the tunnels is not a principal focus of play, but is rather a quickly-narrated bridge between scenes; (2) the roper is not part of the preparation, but is introduce on the fly in response to the player's choice to look for one.

I don't know if [MENTION=463]S'mon[/MENTION] has any comments on these examples, but I've chosen them to try and illustrate some features of the approach both in and out of combat, and with varying degrees of prep vs improv. Notice that a scene can be non-combat and involve exploration (as with my Example 1) but the stakes of the exploration are framed in relation to the signals sent by the players via their build and play of their PCs.

4e makes this signal-sending by players easy, I've found, because so many player build elements (races, classes, patron gods etc) have clear fantasy themes atached (honour, loyalty, betrayal, life, death, servitude, glory etc). And many elements on the GM's side hook onto these too.

And 4e's mechcanical transparency and reliability makes GM improv easy too. I can roll a d10 for psychic backlash damage to the paladin and know, mechanically, how severe this is. I can stick a roper into an encounter on a whim and know whether or not I've turned it into a TPK. That's why, even though 4e is mechanically quite heavy, I find it in an important way "light" or "flexible" on the GM side of play.
Last edited:


I don't know if [MENTION=463]S'mon[/MENTION] has any comments on these examples, but I've chosen them to try and illustrate some features of the approach both in and out of combat, and with varying degrees of prep vs improv.

Yes - I don't improv all the time, as I'm usually running or riffing off published adventures; but I have done so even down to the ad hoc damage, and in 4e it works very well and doesn't come across as punitive the way it would in an OD&D dungeon-exploration game. The ring of the Empress Nemeia mentioned in Dungeon Delve #5 has become a major focus of my Loudwater campaign because of the way players expressed interest in/fear of it, basically exactly the way Pemerton describes. It turned into an intelligent, seductive artifact trying to win over individual PCs and inflict Necrotic damage when they spurned its advances. They have been carting it around several sessions now, trying to get it destroyed, seeking a sage to mask its aura, fighting evil forces trying to seize it, etc. And it has no stats; nothing written down, no DMG Artifact Track. :D


Once more I'm going to extensively quote from the Leverage RPG...
...That's the sort of GMing advice I've been picking up from modern games.

Sure, good example. I think better than in any of the 'modern' games I own - while I'm mostly a grognard buying archaic games, reprints and retro-clones, I do have eg Buffy the Vampire Slayer RPG (21st century) and recently got All Flesh Must Be Eaten (late '90s). Those games seem very much stuck in the White Wolf paradigm. Indeed the time I played AFMBE at Gencon, it was a complete railroad and it actually turned out at the end of the scenario in the Big Reveal that everything was foreordained and nothing the PCs did made any difference at all - we were damned souls doomed to experience the same pre-written scenes for eternity - in-world! :lol: It seems a shame because the Pemertonian style ought to suit the Buffy setting etc very well, but the system and GM advice really weren't directed that way.

Maybe published GMing advice has got better in the past 5-8 years and it's just passed me by? That could explain our different experiences.


Open World Sandboxing is straight-forward. I believe S'mon uses the "You are here. What do you do?" way of conveying it. This could be in the market, in a dungeon, in a tavern, etc. This would be followed up by more color, introduction of hooks, and arcs and more PC exploration of the nooks and crannies of the built world. The DMs job is not to set adversity or pressure against the PCs. It is to set the scene, the color and play the relevant parts and adjudicate outcomes as the PCs explore. The PCs job is to explore, not to respond to the DM pressure in-kind as is their role as protagonists in the story.

Yes, the opposite (non-Pemertonian) style, the exploratory style in an objective world, IME requires that (1) the GM take an objective approach in modelling the simulated world, and (2) that the PCs are pro-active in seeking to explore that world. You see this in Old Geezer's talk on RPGnet of how Gygaxian-dungeoncrawler play is supposed to work. There is really no role for the 'reluctant adventurer' in this style - the world does not owe you an exciting life of high adventure, you have to go out and seize it! Book-Conan type freebooter PCs seeking wealth and power work well. Hobbit PCs that would rather stay home smoking pipeweed are a disaster in this style, and can easily trash a game.
I think this style of world-sim is the best for supporting genuine Gamist challenge (contrary to Edwards' GNS model) - with an objective environment, all player achievements are 'real' - comparable I think to the challenges we face IRL, though of course our PCs probably have different/superior assets for overcoming them.


I'm mostly a grognard buying archaic games


Maybe published GMing advice has got better in the past 5-8 years and it's just passed me by?
I don't buy that many new games, as opposed to modules to crib story ideas and maps. But of the games I do buy, they are probably equal mix old (too many copies of Runequest, for instance, but I must have them all!) and new-ish - most recently the 20th anniversary release of Over The Edge, which has some excellent GMing advice.

But my favourite GM book is the Burning Wheel Adventure Burner. It is so frank about what the game is about and how a GM can help get it there, even moreso than Laws in HeroQuest revised, I think. None of the coyness or incoherence or refusal to talk about metagame that is such an irritating feature of WotC's efforts.


I think this style of world-sim is the best for supporting genuine Gamist challenge (contrary to Edwards' GNS model)
I've never got much into this style, though can enjoy it better as a player than a GM (I suck at GMing it!). But for those who like it, I would think a game like Tunnels & Trolls must look pretty inane and/or juvenile?

Not open for further replies.

An Advertisement