Copyright © 2002 C. Demetrius Morgan


What follows below is intended as a encyclopedic reference entry on the subject of "Role-Playing and the GNS Theory". Alas, as there are no quick and easy definitions for the GNS topic, suggestions for clarifications of the subject matter are welcome.

Please note that, like the previous posting, this entry has been formatted to be specific to topical discussion here at The Forge. This is close to the (hopefully) finalized entry. If I have retained any references that seem too intrusive please let me know so that I can re-edit those portions of the text.

Comments and criticisms designed to help this encyclopedic article achieve clarity are always welcome. EDIT: With the posting of this Article feedback is being actively sought from all members of The Forge, regardless of whether you may have read the GNS/Threefold essays or not.



Role-Playing and the GNS Theory

by C. Demetrius Morgan

Draft 3 Rough Word Count: 2873


There are games we play then there is Game Theory. One is obviously those amusements with which we entertain ourselves, pastimes and hobbies. Such are games. Game Theory can seem disingenuous at times because so few of us realize that it exists. Yet Game Theory is a very real field of study. But what is it?

Look the term up in the average dictionary and, if we are lucky, what might be found is: "Theory of Games". Doesn't say much till cross referenced, where we find Theory of Games given a definition of: "a method of applying mathematical logic to determine which of several available strategies is likely to maximize one's gain or.." whoa, let's stop right here a minute. Very dry reading. Great for Webster's Unabridged, but already it doesn't sound like it has anything to do with the titular GNS Theory of which this article is about. Does it?

Actually it does. Game Theory is a attempt to analyze decision making in conflict situations using statistical analysis, or more simply put it attempts to measure who is most likely to do what, when, and under what given circumstances. Thus it has both a sociological and economic aspect. It can thus be said to be a theory about the dynamic interactions of the decision making process, as relates to game models.

Which is precisely what the GNS Theory is, only as applied specifically to role-playing games.

In order to understand the “Threefold” or “GNS” Theory one must understand role-playing games, their origins, the context in which they are played, and the group dynamic involved amongst the players. More than that one has to realize that the theory, as a whole, draws upon a rich and detailed background of extant terminology. Terminology which is not always applied in familiar ways.

Where to begin?

For the historian the tale begins in 1915 with H.G. Wells and the publication of a rather unassuming book titled "Little Wars". A book that would become known for containing the first set of amateur wargaming rules. Rules intended for use by anyone who owned those once popular lead army men of yesteryear, now long surpassed by the ever present green army men found in most stores toy aisles.

That Mr. Wells found it necessary to provide a codified set of rules for resolution of conflicts which arose in games played with toy army men is proof that, from the very beginning, a debate has waged amongst game aficionados about resolving basic game issues. Chief amongst these is determining victory conditions, defining styles and methodologies of play, and making these “rules of play” available in a set format. We take it for granted that our games come with rules, but it was not always so. In centuries past board games were purchased and the rules of play literally passed on by example, meaning through first hand experience.

Of course it is unlikely that Mr. Wells, or the wargamers who followed, realized that they were ushering in the modern role-playing era. Yet once such parameters were put into black and white and printed for all and sundry to use it was only a matter of time before game players began to wonder why it should be that units, as a whole, were always destroyed. After all in real life warfare much could hinge upon sorties made by the individual; be they a soldier, spy, scout, or saboteur. Thus rules for running individual units were created.

But what does that have to do with role-playing?

Quite a lot. Once individuals were created they ceased to be units. Thus the players began to associate with these individuals, naming them, giving them traits, which eventually built into personalities. Beginning to see anything familiar? Odd, isn’t it, for role-playing games aren’t said to have been created until the mid nineteen seventies! Never mind that many old parlor and diner games could be said to share basic elements of the role-playing experience, or that theatrical troupes have been taking on roles far longer than the RPG is said to have existed, for these contained mere aspects of what the true role-playing game entails. Thus, for the average role-playing game enthusiast, the story doesn’t really begin until roughly 1974 and the publication of Dungeons & Dragons.

Amazing as it may sound role-playing games are not the evolutionary culmination of any singular extant form of game, at least not by conscious design. Yet role-playing games do contain elements from a myriad number of game forms. The use of characters, specifically the taking on of character roles, has existed in parlor games and murder mystery games since well before the first role-playing game was printed. Yet none of these precursors are truly a role-playing game anymore than Chess would be classified as a historical war game, despite the fact few would argue that Chess isn’t the direct precursor of the modern historical war game.

What this comes down to is perspective and definition. Such definitions can be easy and straight forward, witness the following: "In a fantasy game, each player assumes the persona of a particular character, be it witch, warlock, mighty warrior or pious priest...".(1) This from “Fantasy Wargamming”, a early book upon the subject of role-playing. Or the explanations can be vague and ambiguous non-starters like: “The very nature of RPGs makes them almost impossible to codify in any exacting terms.”(5)

Oddly enough that last quote comes from the opening paragraphs of a chapter attempting to provide an explanation of how a RPG is played. When read in comparison to the previous example one might assume this if from some older tome, yet it is not. The book was published circa 1999 while the previous one circa 1982. But is the ability to define what a role-playing game is really so impenetrable that some feel it necessary to make statements like, “Getting involved in gaming can sometimes feel like trying to join a secret society.”(5); or is it merely that the hobby, read the role-playing game itself, has evolved to the point that it now encompass not a single type of game but rather games of a type. Thus leaving hobbyists racing to catch up with the realization.

Certainly that was the case when Fantasy Wargamming was published. The facts loudly declare themselves in the terminology employed. It is, yet isn’t, role-playing jargon. But this is only because the book was written during the developmental infancy of the role-playing game. Yet it is the peculiar jargon which readily identifies a game. Poke has it’s full houses, but use the word “houses” in reference to Monopoly and the meaning shifts dramatically.

So what is a role-playing game?

In short game play was, and remains, strictly defined by the sorts of character that you are playing. Or rather the archetypal "persona" which you are going to be slipping into during the game. The actual style of play wasn't so important in those early days as was the rules of play, however this has since changed somewhat. Of course the biggest problem in the early days of FRP games was in getting gamers to understand the difference between "player knowledge" and "character knowledge", thus the IC (In Character) and OOC (Out of Character) short hand often seen in discussions of game play, which is perhaps how the larger debate of “gaming styles” truly began.

Once the distinction was made it opened the doors to debate about how the game should be played. Obviously this is a debate that has yet to be resolved. Or such is the perception. Obviously, where a game is concerned, there can be no right or wrong method of play. Only what feels right to the players involved. Too, over time, the terminology used by those debating the percieved differences has changed. Once the distinction was summed up as the difference between "roll" and "role" gaming. Of course as the role-playing hobby evolved so, too, did the terminology used to described it.

Then the early FRP games existed in a simpler time. A time before minimalist rules and storyteller systems. A time before CCGs. A time before debates about whether diceless or generic rules systems were better than systems that used one sort of dice or another, or which might be set in specific genre milieu. Thus those early FRP games could all too simplistically mark this distinction as being between whether or not a Game Master or gaming group stressed table top "roll play" over dramatic "role playing".

Times have changed since the early days of FRP gaming. Not because role-playing is not a clear cut gaming form, but rather because it provides such a dynamic background of possible modes of play that every group eventually develops their own unique styles of play. Styles of play which are all equally valid. Yet styles of play which confuse and confound the novice gamer.

So what, then, is this GNS Theory which people speak with such fervor of at The Forge? More to the point, what is this "GNS" thing?

At its core GNS is yet another incarnation of the ever present "roll vs. role gaming" debate, yet it is also much more. It is a model of Role-Playing Game Theory, at least as outlined by Ron Edwards in his essay: "GNS and Other Matters of Role-Playing Theory"(2)- where the GNS model is essentially summed up as the underlying premise native to basic role-playing methodology that refers specifically to the styles of actual play. The basic premise is also otherwise known as the "Threefold Model"(3).

In order to better understand the GNS Theory one has to understand the basic model, or rather the styles of play as defined by this model. These styles of play are:


Gamism (Gamist): That style of role-playing which stresses direct competition amongst players. Such games center upon a central "Strategy Profile" for game premise, usually defined by games in which goals or predefined victory conditions are part of the rules. In the early days of FRPG games this premise was archaically referred to as "roll playing" and typified by a style of play stressing the strict adherence to use of established game mechanics over wandering narrative.
This style of play is not limited to role-playing games, evidenced by the fact it is also a element of table top war games, traditional board games, and many card games. Of course as applied to FRP games it is a very distinctive style of play. One to which much is owed since it is gamist play, as developed within the role-playing environment, and it's rules oriented method of game play, which translates best into static CRPGs.


Simulationism (Simulationist): Where gamism relies upon the "Strategy Profile" to define a game simulationism relies more upon the interplay of “decision makers” in situational conflicts where “determined objectives” in relation to "Genre Labels", these being the categories which games may be sorted by, are the primary stakes. Thus simulationism is that style of game play typified by the assumption of predefined roles for the purposes of in-game exploration of roles which sometimes, though not always directly, fall within the pursuit of set objectives. Thus making this style of play, in part, typical of the sort of role-play that is closest to the methods used in LARP gaming.
However here at The Forge Mr. Edwards defines the Simulationism objective as being focused upon "exploration as the priority of play" in relation to "the internal logic and experiential consistency"(2) of the game world being explored. Thus this could apply evenly to all forms of role-playing, table top, live action, or even minimalist "Beer & Pretzels" game play in which the story may or may not be key.


Narrativism (Narrativist/Dramatist): That style of role-playing in which story telling, or rather the underlying narrative of the game, takes precedence over all other aspects of game play. At it's extreme Narrativism relies purely upon the interplay of decision making and chance in the form of choices depending from direct player narrative to provide structure and direction within the role-playing environment. While this method of role-playing has been variously described and defined over the years it must be noted that this method of play holds much in common with early parlor games of the "Murder Mystery" category. Also very close in nature to this style of gaming, though not directly related to it, are the improvisational exercises used by theatre troupes.
Visitors to The Forge will probably be aware of Mr. Edwards essay on the matter in which he described this method of play as being "expressed by the creation, via role-playing, of a story with a recognizable theme"(2) (and) as set within one of the classic milieus of the literary genre. These being easily typified by classic "Genre Labels" such as: Horror, Science Fiction, Fantasy, Romance, and etcetera.


Are such distinctions possible in a single game?

Yes and no. It would be erroneous to classify any FRP game as a "single game" since most rulebooks present the opportunity for variegated styles of play. Then again this process is perhaps best summed up as follows: "Whenever you play any kind of game, there is a type of role assumption involved."(4) This is certainly true of games involving strategy like Chess, Life, Snakes & Ladders, Clue, Ludo, Monopoly, and even Checkers. But this statement becomes profound when applied to any sort of true simulation game, be it a historical war game or fantasy role-playing game.

But aren't all games merely simulations of an abstract environment?

Yes.

However, as outlined above, the GNS model is concerned with "simulationism", meaning the roles taken on by the players, not the simulated environment itself. Just as "Gamism" is meant to denote the practice or conditionals set within the game by the external interaction of the gamers, as opposed to being directly related to the game itself. In other words "role" verses "roll" play, to use the old terminology.

Summation per Mr. Edwards: "Gamism .. includes victory and loss conditions .. that reflect on .. play strategies" which usually are defined prior to actual game play.

And what of Narrativism?

That's simple enough; this is for the story behind the game. Something that was sadly neglected in the "roll verses role gaming" argument, which centered primarily upon the styles of gaming which the players liked, as opposed to the actual game itself. Thus Narrativism should be see more as embracing more than merely the narrative or dramatic roles which the typical FRP spawn, but as rather encompassing all the underlying elements which form, bind, and otherwise cement a good story in place. In short, Narrativism is the Authorial Role of the game, the game environment, and how the player (and Referee) help to shape that story within the context of game play.

As defined within the context of role-playing, Narrativism is that style of play usually referred to as "shared author fiction" or "narrative storytelling" in the sense that the premise and goal of this mode of play is to allow the game to evolve dynamically from dramatic player input. (As opposed to "railroading" the player characters within a fixed, often modular, environment whose story provides limited choices.) More simply, it is a style of play that affords players the opportunity to use their characters to expand the game beyond a mere set end-goal, often stressing co-operative group effort.

However, in the typical FRP game, the basic principles of the Pulp fiction genre also become a necessary integer of the formulae. At least in so far as there usually exists some form of conflict or conflict resolution. The difference is in the competitive dynamics involved in the story telling process, of which the players are often directly made a part, as mentioned previously. This involvement may be to the point where players are required to actively take part in narration of game events. Thus the game becomes a true role-playing experience.




1. From the introduction to "Fantasy Wargamming", page ix.

2. Mr. Edwards original article may be accessed in full here.

3. Related articles pertaining to this trefold model of role-playing can also be found at "Styles of Roleplaying" and may also may be found here "Role-Playing Games: Theory and Practice".

4. pg 20, “The Fantasy Roleplaying Gamer's Bible“.

5. Pg 25, “The Fantasy Roleplaying Gamer's Bible“.


Bibliography

"The Encyclopedia of Games"; ed. Brian Burns, Barnes & Noble, 1998; ISBN 0-7607-1025-2

"The Fantasy Roleplaying Gamer's Bible"; Sean Patrick Fannon, Game Codex, 1997. ISBN 0-9674429-0-7

"Fantasy Wargaming"; ed. Bruce Galloway, Stein and Day, 1982. ISBN 0-8128-2862-3

"Hoyle’s Games"; Lawrence H. Dawson, Wordsworth, 1994; ISBN 1-85326-316-8


Copyright © 2002 C. Demetrius Morgan

Kester,

just two thoughts for your consideration...

*I don't think it's necessary to mention the mathematical field of "game theory" at all. I think that is just going to create problems, because it can be argued that it does not apply to games-as-play.

It comes from the history of mathematics as applied to gambling and decisons relating to optimization of resources.

If an appeal to an existing academic discipline is intended, I think the fields of literature or drama would be closer matches.

*The GNS and Threefold models are not considered synonymous. Probably will have to narrow the discussion to be a bit tighter with respect to the GNS model.

Third thought:

Your summary of the history of gaming is jaw-droppingly terrific. Don't change a word, or I will have to hurt you.

Seriously, very nicely done, thank you for writing it.

Greetings MK Snyder,

The oscillating hum stuttering from the corner, where a tan fan spurts jets of distracting air into a dark room, distracts; like broken and decaying hieroglyphs. Why is the fan on? What strange words are being spoken through the static of the radio? Will they ever make pickle flavored ice cream for pregnant women desperate with craving?

Alas, none of those questions will be answered in this post. Interesting as they may sound, such topics follow a tangent beyond the parameters of this forum.

MK Snyder wrote: *I don't think it's necessary to mention the mathematical field of "game theory" at all. I think that is just going to create problems, because it can be argued that it does not apply to games-as-play.

MK Snyder wrote: * … Probably will have to narrow the discussion to be a bit tighter with respect to the GNS model.

MK Snyder wrote: Your summary of the history of gaming is jaw-droppingly terrific. Don't change a word, or I will have to hurt you.

Seriously, very nicely done, thank you for writing it.

Actually, I've said before, and will say again, Game Theory, while not in fact designed to address games as it's final product, is incrdibly important and useful to the game designer. Thanks for putting the comment in, Kester.

That said, how did you come across your definitions? They seem to match very early GNS theory, but seem to lack some of the later refinements. In particular, the association of Narrativism with freeform (tabletop, not LARP definition) and LARP ("Murder Mystery") seems particularly erroneous. Most "murder mysteries" are very Gamist in that the idea is to discover whodunit, and thus win the game. I've seen some very, very Simulationist freeform. The point is that I see no correllation, particularly,nor a need to explain where the impetus to play this way comes from outside of RPGS. You don't explain it for Sim, despite the fact that many wargamers play very Sim. This also points out the inadequacies of saying that Gamism is a wargaming thing.

I'd avoid the topic of likening these thigs to outside types of play altogether.

Mike

While the field known as "Game Theory" is of use to game designers, it's not what the GNS model is about.

GNS is concerned with human play choices, including large categories of choices that are not addressed by the discipline of Game Theory.

I think opening an encyclopedia article with game theory gives game theory far too high an emphasis. Certainly mentioning it with respect to the "gamist" class of play decisions would be, in my opinion, appropriate.

Hello,

Some of the following is a repeat of a message I sent to Chris (Kester) a while back. It represents my take on Game Theory relative to GNS. All comments from knowledgeable folks are welcome.

It's always struck me as a no-brainer. As an evolutionary biologist with a strong commitment to sociobiological topics, this stuff is so clear and basic to me that I can't imagine thinking in any other way.

In Gamist play, just treat the situation like any strategic game once you've isolated the variables of "success."

In Narrativist play, you have an interesting situation in which the Prisoners, so to speak, have no Dilemma because they are permitted to communicate (and thus play concerns Communicating About What); the Game-Theory strategy is then a matter of coping with the interjection of unpredictable things. That's why Fortune is turning out to be a very desirable feature of Narrativist design, counter-intuitively for many people.

In Simulationist play, things get sticky - to a large extent, keeping conflict of interest (or metagame "interest" at all) *out* of the picture is the big priority. Game Theory in this sense is largely a matter of avoiding unsuspected rewards for bringing those things in. I'd very much like for someone to analyze how it is that Gamist play can almost always "trump" Simulationist play (i.e. "invade" the system) whereas Narrativism generally cannot. I could do this, but frankly I don't have the time.

Best,
Ron

Kester wrote: So what is a role-playing game?

Ron Edwards wrote:
I'd very much like for someone to analyze how it is that Gamist play can almost always "trump" Simulationist play (i.e. "invade" the system) whereas Narrativism generally cannot. I could do this, but frankly I don't have the time.

I agree with Ron's points about the relationsip between Game Theory and RPG design. I especially liked the suggestion of using Game Theory to test for unintended reward systems contrary to stated game focus...hmm, that would be a form of "incoherence".

I'll still argue for some screen time in the entry for the other fields that are necessary to turn a game into a Role Playing Game (drama! creative writing!), just for balance.

Greetings All,

This is primarily a collection of responses, the quoted posters at a glance herein are: Mike Holmes and MK Snyder.

Mike Holmes wrote: Actually, I've said before, and will say again, Game Theory, while not in fact designed to address games as it's final product, is incrdibly important and useful to the game designer. Thanks for putting the comment in, Kester.

MK Snyder wrote: While the field known as "Game Theory" is of use to game designers, it's not what the GNS model is about.

MK Snyder wrote: I think opening an encyclopedia article with game theory gives game theory far too high an emphasis. Certainly mentioning it with respect to the "gamist" class of play decisions would be, in my opinion, appropriate.

Mike Holmes wrote: That said, how did you come across your definitions?

MK Snyder wrote: I'll still argue for some screen time in the entry for the other fields that are necessary to turn a game into a Role Playing Game (drama! creative writing!), just for balance.

Mike Holmes wrote: I'd avoid the topic of likening these thigs to outside types of play altogether.

Seth,

This being my first response to you let me just take the opportunity to say hello. (Hello! Nice to meet you.)

Seth L. Blumberg wrote: IMO, asking the Dreaded Question is a bad move here. "What is an RPG?" is a question that has stirred up enough debate to demand an encyclopedia entry of its own.

Seth L. Blumberg wrote: Also, this doesn't read like an encyclopedia entry. It reads like an argumentative essay, i.e., as though you are trying to convince the reader of something (though just what isn't immediately clear). Dryness is a virtue when you are trying to appear objective.

Seth L. Blumberg wrote: Lastly, GNS and Threefold are generally regarded as non-equivalent. It would be best to remove all mention of Threefold as other than a historical antecedent of GNS.

Kester wrote: (Hello! Nice to meet you.)

Greetings Seth,

Your post, and certain points made by Mike Holmes, got me to thinking. So I went in and did a bit of work on my initial post. I’ve begun reformatting it to look more like a proper encyclopedia entry. (Though in presenting my material in article format I think there might be room for one of those, too. So who volunteers to write one?) Of course this means I also had to cut out a lot of the *fluff*, hopefully not too much that others thought might be pertinent. But…

Seth L. Blumberg wrote: In an encyclopedia that does not limit itself strictly to Forge-related matters, the threefold and fourfold models deserve separate exposition, with particular attention to points of disagreement between those models and GNS. In an encyclopedia specific to the Forge, the threefold model is merely a historical footnote.

Just out of curiosity, I went looking through the index of my 1994 Britannica.

No entry for "role playing game."

The only entries for "role playing" were references to sociology ("employee training" and "psychodrama").

No entry for "Dungeons and Dragons" (even though it's got to be at least as culturally significant as the musical "Cats" or the Vonnegut novel "Cat's Cradle" both of which were in there when I flipped to a random page).

No sub-entry for "role playing" under "game (indoor)."

Very disappointing, I must say.

- Walt