GNS Footnotes

There’s entirely too much handwringing over threefold models. The arguments happened over a decade ago of course, so it’s possible the authors have changed their opinions. Or simply forgotton in a drunken stupor.

I’m gearing up to discuss the draft release of Sorcerer I got after backing Ron Edwards’ kickstarter. Before doing so I went back to GNS theory, specifically two articles–Edwards’ System Does Matter essay, and the longer GNS and Other Matters. The latter is article 1 in the Forge’s archive, which says something about its significance to the members (well, at least to the founder).

The article is 15000 words long, which is a fair investment of time to begin with; it’s also non-fiction, with technical language. That alone is enough to put a lot of people off. It’s also a defence of the theory since if you don’t bother to read it, how can you credibly argue against it?

Most people’s impressions of Ron Edwards are the infamous “brain damage” comments. Edwards is candid about the trouble he’s instigated on the web, almost as if he’s wearing a badge of honour. But although you can reasonably expect someone who writes “When you point at the Moon, RPG culture hears only ‘brain damage'” (warning, also long) to be an ally of Edwards, they do have a point; Edwards’ theory has been eclipsed by his outrageous comments.

In the WMA we’re used to divining obtuse texts: Edwards’ 15000 words are nothing to Hope’s excessive verbosity.

Executive Summary

The essays give the reader several terms that help define the inner workings of a game, as well as the motivations for playing that game. The author posits different ways these terms can and have been combined in game design, and notes successful (“coherent”) and unsuccessful (“incoherent”) designs. These are successful from a theoretical standpoint but not commercial, since he points to AD&D as an incoherent design (from what I know of AD&D, I mostly agree).

The author also provides defences against misconceptions relating to the terms; these defences occupy a fair chunk of text.

The essay is broken into six chapters:

Chapter 1 talks about why people play games, introducing the concept of Premise

Chapter 2 introduces the new threefold model GNS, and goes into some depth on how Premise applies to G, N and S as a priority for playing.

Chapter 3 talks about Stances, the way people actually interact with the character they are steering.

Chapter 4 breaks down game design by looking at different elements of a system.

Chapter 5 talks about Coherence.

Chapter 6 talks about actual play, and the pitfalls thereof.

Acknowledgements wraps everything up, pointing to some of the sources of definitions and some good reading material (Over the Edge, Everway, and of course Robin Laws).


I’m going to take it for granted that anyone reading this (yes, both of you) is familiar with threefold models and GNS. There are links at the top of this post if you’re interested.

One comment on GNS: there’s a tendency to polarise games into one or more styles. The reality is few of us would rally behind a G, N or S banner; threefold theory is useful as Myers-Briggs theory is useful. If I as GM tend towards S but want a to run a game that demands G, I need to be conscious of that and “flex my style” towards G.

Now, here’s where I think the GNS theory adds value.

  1. Premise (ch1) is something every one of us should take note of–whether we’re running, designing or just playing in games. It’s a key to preference and therefore “selling” the game to your players (and so good for a player to know what questions to ask of a game before play). Chapter 2 approaches Premise from the GNS standpoint. Even if you don’t really like threefold models it’s useful to have the Premise broken according to the different motivations of the players, to note what sells where.
  2. GNS (ch2) also discusses ways one aspect can override another in a game, with negative consequences (because player expectations aren’t met).
  3. Stance (ch3) is a high value concept that determines how a player makes decisions for their character, and the metagame implications (for example, the Author stance may cause decisions that are “out of character” but result in a more entertaining scene for other players). Stance is attributed to John Kim.
  4. Character (ch4) draws attention to three aspects of a PC–measures of Effectiveness, Resource, and Metagame–and considers how property like “money” might fall into each measure. It also considers the Currency of the game and how it applies to these measures.
  5. Currency (ch4) is the relationship between those components. I’m drawing attention to it here, because it’s a subtle but effective part of Sorcerer, and arguably goes beyond just these three measures.
  6. System (ch4) is powerful but really not that difficult–Edwards mentions the Karma, Drama and Fortune of Everway. I don’t feel the need to go further.
  7. Switches and Dials (ch4) are buzz-words that pertain to broad and fine settings of character. A Switch may be set to a character class, which in turn contains a Dial to do with choosing which class abilities to optimise at first level. For a point-buy system, this case is inverted. I think this is interesting because of how players respond to it and decide on their preference for e.g. D&D over Runequest. Whether it’s actually useful given my preference for very light system is another matter.
  8. Coherence (ch5) and its associated words are a concept that equate to good vs bad design. This forms part of a long and rambling chapter (oh, the irony) that includes many historical examples, the evolution of the metagame, “hybrids”, “drift” and other concepts. There’s a lot of emphasis on coping mechanisms for different preferences–for example, drift is the answer to some players having priorities orthogonal to the system being played. Overall Coherence is a useful term if only to illustrate a point: to get consensus between group and GM you need a system that behaves as players and GM expect it to over and again.
  9. Search vs Handling Time (ch5) is barely a footnote in this essay (it’s explained in System Does Matter), but it bears repeating. Complex system leads to higher search and handling times. The message here is this should be the only foundation of Rules-heavy vs Rules-light; the notion that Rules-Heavy=realism and Rules-Light=story should be thrown out.
  10. Social Contract (ch6) is always useful to bear in mind, even if the response is “and your point is?” We come together and agree to play a game by a set of rules and in the spirit of friendship. There is an interesting twist further down the chapter where Edwards mentions the shift from “tourney play” where several groups would be objectively tested against a single scenario, to single-group play (i.e. what we generally mean by a session). I think he loses an opportunity to reinforce the concept of social contract here.
  11. GM/Player relationship (ch6) as it pertains to GNS is useful–it determines the role of the GM as referee, opponent to the players, channeler of external source material, and so on. It also illustrates the broad spectrum of Narrativist gaming from the traditional player-GM dichotomy to the Storygame “GMless” (Edwards would say GM-full) mode.


The 6-part essay is a useful collection of definitions, framing devices and provocative questions with some commentary on how historically games evolved. For that alone it’s worth attention.

I start to have trouble with the model around chapter 5, where it blatently diverges from a clinical theory into a historical discourse–a useful one with many contextual examples, but a divergence nonetheless. The biggest problem is the equation of incoherence with the emergence of metaplot, and the illusionism baggage tacked on the end of that chapter. Edwards is right in that it’s nonsense to pretend the GM is a “storyteller” writing some great pre-planned chronicle and yet the players are masters of their own destinies. But honestly I don’t think anyone I knew played like that–scenes were framing devices and the Storyteller was a marketing buzzword, nothing more. We knew they were the GM.

Metaplot’s a marketing tool, too. The fact that we had Shadowrun and Dragonlance and other third-rate fiction pressed on us was irrelevent. If that fiction hadn’t existed the same consumers would have gobbled up Gemmel and Eddings and Donaldson and other third-rate sub-Tolkien fantasy. Yes, the White Wolf publishing model was extremely cynical. So what?

GNS theory is like any theory that boxes the game into camps. Some people resist being boxed. GNS is at its best when considering hybrids and priorities in play, not absolutes (which to be fair it does not; that appears to be an assertion of its critics).

The other thing GNS does nicely is provide a toolkit to dissect games and determine why some components don’t work. What it doesn’t do for me is tell me how to put those games back together again–but to be fair it doesn’t need to; I have enough experience to choose and build my own games.

You then have to wonder who this essay is aimed at. Certainly not a newcomer to the theory. Perhaps we would be better served by hearing Edwards’ reasons for writing this essay in some preface that gives better historical context. The current one that asserts the theory is not Dogma but it is the single coherent theory of the Forge doesn’t do it any favours, nor the assertation that most roleplayers are “tired, bitter and frustrated”.

Edwards has been demonised for some of the things he’s said: I don’t feel the need to defend or admonish him, since he has a fanclub for that. I do feel the need to look at what he’s written critically and without emotion. If that’s a problem with anyone, I’d like to point to a couple of examples from my English MA experience for contrast:

George Silver is one of the foremost exponents of English martial theory, laying down his Grounds and Governors succinctly and with great passion; at the same time, he makes no bones about the fact that he really, really hates the Italians.

Sir William Hope is author of a body of extremely well considered fencing treatises, mild-mannered and sensitive, if verbose. He’s also implicated as a bully landowner who once knocked out a woman’s teeth during a mob-handed eviction (I don’t have a reference for this; it may be another landowner named Hope).

Thomas Page’s Use of the Broad Sword has considered contemporary to other Highland Broadsword texts, despite the suggestion that it’s a plagiarism of several earlier manuals (and an incoherent one at that) whose primary purpose was to attract readers to the advertisements for Pages’ own watch and clock shop in Norwich.


p>Make your own mind up. TTFN. 

3 thoughts on “GNS Footnotes

  1. Stance is probably the bit of Forge Terminology I’ve had the most use out of. Indeed it is (perhaps ironically) the piece of Forge Terminology I tend to find most useful when describing people’s individual preferences about gaming. I’ve often found that being able to say “okay, the reason this person is getting confused or frustrated is that they like to stick firmly to Actor stance, while our group tends to play mostly in Author stance” is actually quite helpful.

    By contrast I’ve never seen the practical use of G, N and S as terms (with the possible exception of N, and only insofar as it describes a very narrow set of games). One of the players in my current AD&D 2nd Edition game enjoys the abstract tactical challenge of the game. I could label this as “Gamist” if I want to, but doing so wouldn’t help me actually make the game more fun for the player. Indeed I suspect it would make the game *less* fun, since the kinds of tactical challenges he likes are the ones that arise organically from the situation (that is to say, a game which is run in a primarily “Simulationist” way). If I tried to introduce deliberately “Gamist” challenges this would probably dent his enjoyment as much as everybody else’s.

    I’ve had similar problems with the concept of “incoherence”. Back when I first read the Big Essays on the Big Model nearly a decade ago I found incoherence to be a really compelling idea with profound game design implications. Much later I realized that all I was really doing was using Forge language as a stick to beat White Wolf with. I don’t much like /Vampire/, but the reason I dislike it has nothing whatsoever with the reasons for which the Big Model says I should dislike it.

    I think I also (on the basis of recollections from a *long* time ago) parted ways with the essay around the time it started conflating theory and analysis with history and (what I perceived as) a strong dose of personal preference.

    • Stance and the foundation threefold stuff come from John Kim, as noted. This is a criticism I’ve seen of the theory–the best bits aren’t even original.

      And yes, the theory stops being useful when it quite clearly diverges into a history lesson (as interesting as it is).

      My problem with the GNS theory is I don’t see a useful workflow for design. Maybe that’s in other essays or the Big Model. Without that it’s an analytical tool that’s good at stating the obvious but not really necessary.

    • Based on my dim recollections, the essays and the like actually take it even further away from being a useful workflow for design, because they’re very big on the idea that Creative Agendas are about decision making in play, which isn’t something I particularly want a big complex model to help me with.

Comments are closed.