Why is the Prioritization the Thing?

[marcus]

Hello there- first time poster to this forum.

When first reading Ron's various essays on GNS, I formed the view that Ron was saying that a good RPG should contain only one of the three GNS outlooks, to the exclusion of the other two. I then disagreed with this doctrine. In response, Ron suggested that I had misunderstood his model, and directed me to some further postings on this subject I had not read.

The most helpful of these was a posting of Ron's from the topic "Can a game have all 3 G/N/S Revisited" in this Forum. In that posting, Ron said (amongst other things) by way of commentary on an earlier posting:

You begin your breakdown by specifying that you are talking about "elements of role-playing games." This needs some fixing. GNS does not describe elements of role-playing games. It describes three possible first priorities of role-playing as an activity.

This is a big deal, because lately people seem to have begun repeating some insupportable phrases about "every game has elements of all three" and similar, which I have decided needs weeding.

From the above I glean that according to the GNS model, the key question is which of the three GNS goals has first priority in actual play. And, of course, within this framework it is indeed "unsupportable" to speak of games being Gamist, Simulationist and Narrativist together, because as a matter of definition only one of the three goals can have first priority (except in the very unlikely event of two being of precisely the same priority).

As Ron said later in the post I am quoting, "the prioritization is the thing" as far as his GNS model is concerned.

My question, however, is why exactly is it the prioritization of outlook that is the key point for analysis of play?

If there are three play goals- to simulate, to game, or to narrate (if this is terminology is acceptable, I have observed that terminological debates are an important feature of this forum)- then why is the most important question which of these goals is pursued to the greatest extent?

To explain what I mean, let's imagine a system whereby the tendencies to value a particular goal in play were assigned percentages representing the percentage of the players' total attention that was given to each. Under such a system, each of the following would, under the prioritization taxonomy of GNS, be rated as "Gamist":

Game-valuing 90%; Simulation-valuing 5%; Narrative-valuing 5%
Game-valuing 35%; Simulation-valuing 33%; Narrative-valuing 32%
Game-valuing 50%; Simulation-valuing 45%; Narrative-valuing 5%

In each case, the play is Gamist as game-valuing is more important than either of the other two goals. These three situations are, however, very different. In the first Game is valued almost to the total exclusion of the other two outlooks. In the second, all outlooks are valued almost to the same degree, but with only a slight preference for Game. In the third, there is little concern for Narrative, but Game and Simulation are of almost equal importance.  

It strikes me that there are very important differences between the above scenarios, which are lost by reducing all to the term "Gamist". Indeed, I would argue that it was more important to distinguish between these three situations than three types of play in which each goal was of fairly equal importance, but with a slightly greater emphasis on a different goal each time (such as G 35%, S 33%, N 32% (Gamist) v N 35%, S 33%, G 32% (Narrativist) v S 35%, G 33%, N 32% (Simulationist)).

By way of illustration, I will use an analogy. In design theory for Armoured Fighting Vehicles, there are three major performance goals- obtaining high firepower, obtaining high mobility, and obtaining high survivability. All these three goals are, to a degree, incompatible, with, for example, big guns making a big tank (big target = low survivability) and more weight (low mobility).

Now just say one adopted a taxonomy analagous to that in GNS and decided to classify all AFVs in accordance with which of these goals was to be the first priority. One thus has a "GMS" model, with Gunnist, Mobilist, and Survivalist designs. Yet how useful is this taxonomy? Certainly, there are some circumstances in battle which one is only really interested in one of these three matters (like sitting under an artillery barrage and praying one is in a Survivalist design), but generally speaking one hopes that the designer has made an attempt to facilitate each of the three goals, despite their partial incompatibility. Tell a general that he is to be given the latest Survivalist tank, he or she will legitimately ask "But can it move and shoot as well?".

I appreciate that argument from analogy is weakened by the possibility of challenge to the aptness of the analogy, but I take this risk in order to give a little concreteness to an issue that might otherwise seem of merely abstruse theoretical significance.

So, putting the question again, why should we be so concerned about which goal of play is the first priority, rather than to what extent each of the three goals in play are met? Why is RPG design so different from tank design?

Marcus

[Calithena]

Hi, Marcus.

These are questions a lot of people who encounter GNS ask, and they seem reasonable to me. I'll let the Forge Sages respond in more detail on behalf of the theory, but basically as I understand it things go like this. G/N/S are descriptions of what people look for in their (actual) gaming, which goals (or modes of Exploration) they are consciously or unconsciously focusing on when they want to have fun.

If you're an old-time, experienced role-player, odds are good that at least two and maybe all three of these modes will appeal to you, at least in different times and places. But consider, for example, players who pride themselves on both playing 'long political adventures where no dice are cast' and 'hackfests'. Players who say this are in groups which switch back and forth between a Nar or Sim-oriented LARP type mode often only loosely supported by the underlying mechanics, if at all (consider 'no dice are thrown') and a Sim or Gam-facilitating combat engine which typically is better supported by the system in question. Such groups are functionally Drifting from one mode to another - functionally, because at least some people are having fun. (More on this in a minute.)

If you're wedded to the traditional model of the long-running campaign - a model which is far less common in actual play than warrant it being the default assumption - then you might think "oh, to be a fun, fully rewarding game, you have to drift like this, so you can experience all different modes of exploration at different points".

But a couple questions then arise. One is: what rules really SUPPORT this - not just allow it in the hands of good GMs or good players, but actually support it? Arguably, there are no rules that really offer support to all three modes of play. Another is: how many of your games actually fit the 'long-running campaign' model? Many, maybe even most role-players never experience this after college, and even in younger days there's plenty of switching back and forth.

OK, anyway, there are a lot of different claims going on here. The 'scientific' one (as I understand it) is that there are three identifiable modes of gaming behavior, which are all observable parts of gaming conduct. Separating out Sim from Nar and Gam both is the key here (or at least it was for me), and I think this is a really, really important distinction for people to get, even if they ultimately go back to a hybrid outlook.

The 'practical' one, which is what you are questioning, is twofold. It basically says:

1) Many cases of gaming unhappiness are cases where groups of gamers have conflicting GNS priorities in their play.

2) Being clear about what your priorities ("Creative Agenda", which needn't be cast at the outset in GNS terms) within your Exploration are will often allow you to have happier, less conflicted gaming.

3) Designing games with clear GNS priorities will therefore make for happier gaming.

3a: "But I enjoy all three modes!" Well, if you're running a long-term, old-school campaign with the same players over several years, then you'll have to come up with good drifting mechanisms, maybe meeting about 'what you want to do next' on a semiregular basis. But if you're like most gamers, where most of your games go from, say, 4-12 sessions, then this is much less of an issue. What do you want out of your gaming right now? Pick a game that facilitates it and then move on to the next one.

I'm far from an expert on this stuff so we'll see what others say when they chime in. But these are the answers I've come up with for myself (with the help of the Forge community) to the questions you're asking. So maybe we'll both get plowed over or maybe there's some kind of rapproachment in the offing, but this is where I'm at with what you're saying.

[pete_darby]

Time for my tuppenceworth...

Because designing an AFV isn't like Jazz.

Once you've designed a tank, it's out there being used. Sure, your end-users can (and will) tinker with it to maximize their priorities, but the latitude for doing so is fairly minimal.

Desiging an RPG is much more like writing a manual about how to design an AFV; you can lay down the law all you like, but in the end, your end-user can ignore every damn thing you say to achieve their ends.

To extend the analogy beyong breaking point, the model being discussed here came up as a result of a bunch of folks realising they've been using bits of mechanical instruction books for everything from AFV's to Ocean Liners to Jumbo jets to segways without knowing what came from where, or what exactly they were trying to build.

Once you've got an idea of what your trying to build, what's been done before that works with that, and what doesn't, don't you think it'll make design (and maintenance!) easier?

But it's not much like engineering at all. Because, at best, an RPG design is analogeous to that design manual, rather than the physical end product. The end product is an experience of rewarding role-play, a collaborative process rather than the physical product of a rulebook, sourcebook, etc.

The GNS distinctions are, afaics, very broad outlines of goals that are, ultimately, mutually exclusive. Note the word "ultimately" there, because many, if not most GNS conflicts can be negotiated amongst the group if it's socially functional. But the negotiation will be for what we will prioritize in this instance, to the detriment of the other agendas.

Now, the question is very apposite: why should we worry about classifying what agenda is prioritised, when the extent to which the goals are met is important? Well, what does it matter that a given system is equally good (or bad) at providing equal amounts of GNS, when the compromises necessary for that make it not-quite as good for any one of those agenda as another game? If I want to buy a sports car, do I want a fairly fast car with seating for a family of seven, or a hell on wheels two seater? While the seven seater may be fast, my agenda is "ULTIMATE SPEED!" An SUV may be fast, but I don't want to go off road. Now, if I'm looking for a car I can do the school run with, bet the shopping, AND go wild in the country on the weekends, the SUV will look better, depending on how well it would handle in those two instances. Because it's not a mixed priority or agenda, it's two separate agenda in different instances. Even if I wanted to go wild in the country with the kids, "go wild" and "with the kids" are opposing priorities (a quad bike is great for "go wild," less so for "with the kids"), and in making a choice on will be prioritised, while the other will inform the decision in a subsidiary role.

The model doesn't say "Choose a game to fit that one agenda," but does say "choose a game that fits your pattern of play, or be prepared to change one of them."

[M. J. Young]

This question seems to be tailor made for me.

[*]I wrote a game, published just before System Does Matter was published, which I still maintain provides support for all three modes;[*]I play in all three modes with some regularity; and[*]I am a staunch believer in and defender of the model (and, I hope, a contributor to the understanding thereof at a few points).[/list:u]
Years ago, when I first made the statement that I played with gamist, narrativist, and simulationist priorities fairly equally, Ron challenged that assertion, ultimately with the statement, but not at the same time. I think that was what sold me on the model, because he was absolutely right. The three priorities interfere with each other; at any particular point in play, you are going to pick one, and you're going to follow that.

Multiverser does support all three GNS priorities, but it does so by employing three points to get there. One is that, as a game in which player characters move from universe to universe, the rules change--and I see this the more so, because the settings themselves are part of the rules (I know this may run into a huge debate about setting versus system, but I've been through it again and again, and the only answer that makes sense is that setting and system are completely integrated into each other even if you use a generic engine). If I drop a player character into a universe in which there are moral issues staring him in the face, I've changed the rules: we now have moral issues to resolve. The second is that the game responds to player choices rather smoothly, such that whether it goes gamist, narrativist, or simulationist is very much based on what choices the player makes for his character. I can drop a character into a world with tremendous moral issues, and he can embrace them or ignore them, however he prefers. The third is that player characters are independent, much of the time even isolated from each other, such that one can play gamist to his heart's content while another is delving difficult moral issues and a third is discovering an enchanting world. Thus although the game supports all three modes, at any moment any given player is going to be focused on the one of his choice.

Further, although Multiverser supports narrativist play, it doesn't force narrativist play. A game like Legends of Alyria or (I think) Sorcerer will make you face the issues; it's built so solidly into the engine that it can't be avoided.

A game supports multiple GNS goals by only three known possible ways: congruence, drift, and transition.

Congruence refers to those rare situations in which the same decision supports all three goals. I use two examples. The first is the Armageddon scenario, in which someone comes to your character and says, "the world will be destroyed unless you, and only you, save it; will you do it?" Whether your reaction is gamist ("Hot diggedy, wow, what a great opportunity to prove what we can do!"), simulationist ("This should be interesting; I wonder how it will work out?"), or narrativist ("That's a fascinating moral question: how far will I go in risking myself if the world is at stake--do I care if the world survives if I die?"), you're almost certainly going to agree to attempt to save the world. Similarly, if we set up a game in which all players are members of a squad in Viet Nam during the war, while you're traveling through the jungle you're going to do pretty much the same things whether you're fighting for survival and victory, exploring what it must have been like, or wrestling with the morality of war. (There will be moments when even this will break down, because not all decisions will be congruent.) Designing a game system to promote congruence is a dream, but I don't think anyone's done it.

Drift is what Multiverser facilitates: it lets the referee and the players move to whatever mode they want to play in the current situation. It encourages drift through universe change, in that sometimes moral issues are going to be grabby, and sometimes challenges are going to be in-your-face, and sometimes worlds are going to be plain fascinating (and conversely sometimes it will be difficult to scratch up a moral issue, or find a challenge, or see something worth exploring), but the core engine lets the players decide what they're going to prioritize in play, and the referee follows their lead.

I'm not sure whether transition has been successfully managed yet. Fang Langford's Scattershot was all about transition, in essence structured drift, but was never completed. This would require a rules system that changed to support the various modes, either in response to in-game events or by conscious player choice.

Your AFV design analogy fails for the reasons Pete observed; but maybe I can clarify that.

Obviously, if you're prioritizing movement, you have to make decisions that prioritize movement. If what you're trying to design is a vehicle that moves, you're going to go for very small guns, very light armor, and make other decisions that are counter to the other priorities. Sure, in designing such a vehicle, you're going to make compromises to attempt to balance the three; but in each specific case you're going to focus on whatever the priority is first.

But as Pete said, AFV design is not jazz; and that's an extremely good way to put that. We're not talking about designing a practical vehicle; we're talking about writing a song, or creating a movie. We're not talking about the tensions between armor, movement, and firepower. We're talking about the tensions between lullaby, dance tune, and action scene music. We're talking about three things which cannot all be prioritized at the same time. You cannot make an upbeat dance tune lullaby; it fails at one or the other, at the very least. You cannot score a movie and have it be something to which you can dance (assuming it's not Chorus Line or Flashdance or something), because the music has to support the scene. The more you try to combine these three objectives simultaneously, the more evident your failure at all of them becomes.

Now, what happens in some games is they exhibit incoherence in the design, trying to support gamist, narrativist, and simulationist preferences in the rules and expecting that players will sort it out. A good example of this is the overlay of rather narrativist (morally conflicting) alignment rules on the extremely gamist engine of OAD&D. Everything in the game other than that is built around kill monsters, get treasure sort of stuff. That's built around wrestle with the moral issues of the world. How do you reconcile these? Everyone does it differently. Some people completely ignore alignment altogether, and just go for the gold. Some minimize the importance of advancement and build stories around the conflicts created by alignment. Many seek some balance between the two, trying to work out how to pursue the gamist goals promoted by the core engine while acknowledging their commitments to specific moral and ethical principles (perhaps one of the reasons why chaotic evil is said to be a preferred alignment). The balance gets placed in different places, and each game group believes they are doing it "right"--but when players cross to other groups, suddenly there are conflicts about what is the "right" way to play. Many gaming groups have inherent problems because some are trying to deal with moral issues while others are going for the victory, or similar conflicts. Just about any game, particularly one with incoherent design, can be drifted to serve any goal; but if players aren't agreed on the goal, drift is going to create dysfunction, as they fight for control of "what we're going to do".

Congruence aside, it's not possible to make a decision that prioritizes all three goals as primary; even with congruence, this is not how decisions are made. That is, it may well be that the decision I made which improves my tactical position for the upcoming battle happens to be the same decision someone would make to realistically play a soldier in my position, or even what might have been chosen in dealing with the moral issue of how do I face a war I'm forced to fight; but I didn't make it for those reasons--I made it solely because it improved my position tactically, and therefore I was making a gamist decision. Similarly, in a longer adventure (not a campaign), it's not really possible for gamist, narrativist, and simulationist goals all to rise to the kinds of ratios of preference you suggest. We're not talking about a game in which 34% of the decisions were gamist. We're talking about a game in which 100% of the decisions which mattered to the players were gamist, and there were a lot of decisions that could be classified as simulationist or narrativist in some way that didn't matter.

When it comes to it, you have to prioritize one. The others may or may not be along for the ride, but they are not relevant in any significant way to what we're doing, why we're doing it, and why this is fun.

I hope that helps.

--M. J. Young

[marcus]

Thanks for all your responses.

My near-namesake, M.J. Young, observed as follows:

Similarly, in a longer adventure (not a campaign), it's not really possible for gamist, narrativist, and simulationist goals all to rise to the kinds of ratios of preference you suggest. We're not talking about a game in which 34% of the decisions were gamist. We're talking about a game in which 100% of the decisions which mattered to the players were gamist, and there were a lot of decisions that could be classified as simulationist or narrativist in some way that didn't matter.

This is a factual statement about RPGing behaviour. If it were right, then I am certainly wrong. But do RPGers really act this way, making all decisions that matter in pursuit of a single GNS goal? This is quite a strong statement, even if one assumes everyone playing an an RPG all shares the same GNS priority (which can by no means be assured). Every single decision that matters made by every single player in a game session being in pursuit of the same goal? Sounds pretty unlikely, huh? Even if one looks at only a single player, I don't believe that there are many people out there who could muster such consistency of purpose.

My observations of the players I game with suggests to me that even if a player is more likely to make a decision in a given adventure/session on the basis of one GNS goal more frequently than for either of the other goals, in most sessions no goal will have a monopoly on the decision proceedure.

Ron has told me before that it is good to look at examples of actual play when examining GNS issues, and this is probably an appropriate time to do exactly that.  One of the favourite games played in my group at present is a game we refer to as "Superheroes" (the rules sprang out of Superhero 2044, with a Champions-style power costing system grafted on and most of the original rules changed to home-made mechanisms, but that's not important for present purposes). There have been several GMs for this game over the years, with myself being the principal (although not exclusive) GM at present.

With a campaign running some 20 years (although I have only been involved for the last 18), there is much shared joy at the complex campaign world that has grown up over this time. Some of this joy is manifest in the regular "newspapers" that are put together, each in fact styled as series of clippings from various publications of the game world, with each clipping written as much as possible in a style consistent with the editorial policy of the periodical in question, and in recent years often accomanied by photographs of the heroes in question in appropriate action dioramas. This might be thought of as a rather Simulationist thing to do, but the role of the newspaper is in fact (I would contend) to facilitate not only the goal of Simulation (building up the complexity of game world), but also Game, in that a major use for the newspaper is as a means of planting subtle clues for future adventures, as these can be easily hidden amongst the other challenge-irrelevant material. It is then a test of skill for players to separate out the wheat from the chaff, whilst simultaneously enjoying the stories for their own sake. The goal of Narrative is facilitated also to some extent, as moral issues can be emphasised to be adressed later in play. For example, the exploits of a particularly ruthless non-player hero feature from time to time to facilitate the narrative theme of "What makes a hero?", which regularly recurs in the campaign.

Overall, I would say that most decisions in Superheroes are in pursuit of the goal of Game. Players (myself included) like to have lots of fights against supervillains and similar, and will often spend considerable time in advance of a fight carefully working out their mutual strategies, and then enjoy spending their Experience point reward on character improvement. Yet I have often noticed players foregoing Game advantages for their character by a determiniation to stay "in character". For example, the player of "Brute", a super-strong and super-stupid character, will often delight in emphasising the character's stupidity at the expense of solving "the mission", even though at other points in the play of the very same adventure Brut will become highly mission-oriented. Other players will do similar by often making a point of playing out the flaws of their characters, whilst at other times in the same adventure they will down-play the flaw to facilitate the achievement of the mission.

Although the Narrative goal would be, in my opinion, the least frequently pursued of the three in Superheroes, at times moral and social issues do become quite important, particularly the "What makes a hero" issue I mentioned earlier. A situation the PCs often find themselves in is the dilemna of how hard to press a captured villain for information. In some instances, the players will have their characters take the Game-oriented course of pressing as hard as possible so as to obtain mission-important info. At other times, the PCs will take the Simulation-oriented course of acting according to the established personalities and priorities of their characters. On yet other occassions, the PCs will put aside both mission goals and previously established patterns of character behaviour to take a stand on the Narrative issue "Hurting prisoners is bad, heroes don't do that sort of thing".

I could go on with further descriptions taken from either this game or one of the many others I have played recently, but I've probably said enough for one post. My contention, which I say is based on actual observation of play and not merely a theoretical construct, is that players don't (or, at least, don't necessarily) act in a consistent way, always putting one GNS goal in front of another, even in the space of a single adventure. Although one goal may be favoured more than another taken overall, a player will not always make decisions in a given adventure based on that goal, but only sometimes. It is thus important to look at the tendencies of a player or group of players towards each of the three GNS goals, not simply the one that, summed over the entire adventure, is  pursued most frequently. For Superheroes, for example, simply calling the group "Gamist" would not be that helpful, whilst calling it, say, G: 45%, S: 35%, N: 20% would be more meaningful.


Marcus Young

[Valamir]

Marcus, a couple of points related to your example...in no particular order other than the one they occured to me in.

1) You mention that your particular brand of house rules wasn't important for the discussion.  Quite to the contrary.  House rules exist because a group wants to fix something in the foundational system to make it more to their likeing.  To adapt it to meet their own needs (around here we would say "to make it better facilitate their own Creative Agenda").  I'm not suggesting you need to brain dump your system here, but give some thought to the choices of rules you added, modified, discarded over the years.  They probably say alot about what your priorities as a group were at the time.

2) The idea of a Creative Agenda exists in its exclusive state for the course of a single "instance of play".  An Instance of Play is an intentionally undefined term that roughly translates to "how long it takes for a Creative Agenda to emerge during play".  This is often longer than a single session and certainly shorter than 18 years.  Therefor it is quite certain that over the course of nearly 2 decades your group has enjoyed likely all three Creative Agendas.  The exclusiveness clause simply means you weren't enjoying them all simultaneously.  

It also means that you generally cannot determine Creative Agenda from a single event during play.  Single play events are frequently used to illustrate concepts simply because they can be framed easily.  But in actual practical application they are insufficient.  See Walt Feitag's excellent thread on the dangers of focusing on "key decisions".


3) Assertaining a Creative Agenda from a play example generally requires a good bit of meta analysis.  In other words, what the characters did (or more pedantically, "what the players decided to have the characters do") is only a part of the story.  The social interaction among the players during the game is also important.  When a player has his character do something, how do the other players react to that.  Often times one can detect the Creative Agenda from what actions get cheered and what get jeered, or meet with disinterest.


4) You are not alone.  Just about every person here has gone through this same mental wrestling with the idea.  I myself had proposed an elaborate graph where the ratio of GNS modes in play at any given time could be mapped out.  Its practically one of the "standard stages of understanding" of the model.  The key is often to simply keep at it until one figures out what it actually says, rather than what one thought it said when one first read it based on one's own prior conceptions.

[Ian Charvill]

The thinkg here would be that GNS isn't about labelling the group but labelling the creative agenda of the group in a specific instance of play.  Labelling the group Gamist at all, extends the theory beyond it's specific purposes.

So you might be able to say, in 55% of all observed instances of play, this group displayed a gamist creative agenda but in a particular instance of play it wouldn't be valid to say the group was 55% gamist.

It's like in a 100 meters sprint.  Over a season, a particular runner might win 55% of the time.  However, in an individual race, only one runner can win - they can't win in that one race 55% of the time.

HTH

[pete_darby]

Well, I think you've missed an important point: we're not saying that one agenda will monopolise, far from it, but that one will dominate. I'd take issue with MJ's assertion of 100% that matter, and use the word instance instead of game, but the principal holds that in any given instance of play, one of the creative agenda will be dominant. Which is dominant is usually tacitly negotiated (or even dictated) socially, but at any one time, either one agenda is dominant, or the dominance is in the process of being negotiated.

I don't know about anyone else here, but I tend to use the terms "gamist", etc when referring to systems, players, etc rather than the agendas themselves, as shorthand for "having gamist tendencies" or "supporting gamist play." So to me, if I typify a player as "gamist", I'm saying she will tend to prioritize peer respect for actions above simulation or confronting moral dilemmas. But it's a tendency, and doesn't preclude her from getting into other modes of play. A gamist system is one that, through it's design, tends to support gamist play (tactical planning, character optimisation, whatever) above simulation or moral dilemmas.

The thing is, when a group is exhibiting a dominant agenda, and reinforcing it, it will tend to edge out the others. The creative agenda is ultimately an expression of the social contract applied to exploration of an imagined space. As such, while we can talk about gamist players, systems, etc, they are all just things with tendencies towards gamist play within the group.

NOw, like you say, what's the point in saying that a game, or session, or player is "gamist", without saying that they're 100, 75 or 51 % gamist? Well, firstly there's the impossibility of getting those numbers in. Then there's the question of what you'll do with that information.

Take your campaign as a fantastic example: my knee-jerk reaction is to typify it as a broadly simulationist game. The gamist propositions (optimisation of tactics, self-improvement through experience, seeking clues throughcombing through the paper) can also support sim play (exploration of systems, character and setting).  The narrative issues of "what makes a hero" can still be explorative rather than addressing premise. Crucially, you state that players will break established characetr to address moral questions, but even that gets couched in sim terms ("a hero wouldn't do that" is addressing a moral dilemma through established genre conventions).

As ron says, the crucial question is what do the players approve or get psyched about; by the looks of things, I'd think "being heroes" is probably the answer, as opposed to beating the bad guys or finding out what the price of heroism is, but I'm putting words in your mouth, and you know better than I.

So, if it's primarily a sim game, what then? Throw out all the gam and nar goodness? God, I hope not! But you're looking at GNS thinking that's what it's saying, but it isn't. It's saying that whatever system you're using, it will end up as a sim supporting system in this group, whatever else it does. From time to time, nar and gam play arises in this group: at those times, they're playing nar, gam, or whatever, and at those times the sim aspects of the system are deprecated in favour of the gam or sim parts.

But enough ranting: I think if you bear in mind that, when not talking about an actual instance of play, gamist, narrativist and simulationist are shorthand for preferences, tendencies and support for the given creative agenda, not pigeonholes and taxonomic defintions, the theory makes more sense.

One other thing... MJ said "decisions that matter", which may make the argument circular, in that decisions that don't support the dominant agenda in an instance of play by definition don't matter. But that's an important distinction: in a session of play, think about the choices that made a difference to the play that night, what was behind those decision, and, crucially, how the rest of the group reacted to the way the decisions were made. That'll tell you what the dominant agenda was in each case.

The next question is... what do you want the theory to do for you? Becuase I think you think it's dictating something which it isn't.

PS: the joys of cross posting.

Skip the rest. What Ralph said.

That'll teach me to ramble all over my lunch break

[marcus]

You are not alone. Just about every person here has gone through this same mental wrestling with the idea.

Yes, I've heard this said before. It crosses my mind, however, that the fact that a theory is almost universally rejected by those people not subjected to intensive indoctrination in it is not a big recommendation for the theory...

As a way of stimulating thought and discussion about RPGs and their play, GNS theory does an excellent job- I have thought more about these things over the past week than at any time in 25 years of roleplaying. There is also a lot of useful terminology introduced which greatly facilitates RPG discussion (although I'd have to say that some terms, like "Step On Up", strike me as rather unwieldy). As a method of predicting what sets of rules or styles of play are likely to succeed in producing a felicitous result, however, I have grave doubts.

I have a further concern about GNS theory that it is really untestable, in that the definitions of the various outlooks are loose enough to allow many types of play to be classified in different ways to ensure consistency with the theory. For example, Pete (undoubtedly in perfectly good faith) interpreted apparent instances of Game-oriented and Narrative-Oriented in Superheroes as really being Sim play to support the idea that a game session might have been all Sim. Presumably another person might have interpreted the apparently Sim-oriented and Narrative-oriented play as crypto-Gamist. In such circumstances surely conclusions will be driven more by faith than by empiricism. Even if I deliberately set up a play session to test GNS by trying to facilitate, as far as possible, an equal mix of Sim, Game, and Narrative play and then reported the result, I fear that the result would be simply be brushed aside by the above process of GNS reclassification.

So as not to be too negative, however, I will ask this: What attempts have each of you made to attempt to verify the predictions of GNS theory through actual play, and what were the results?


Marcus Young

[pete_darby]

Marcus, to a great extent you're right: the model is, to a large, degree non-falsifiable, and it's supporters are filled with ranks of guys who will tend to say "You haven't understood it" to anyone who disagrees with it.

But you can say the same about most literary, artistic and cinematic theories. Because, like I said before, playing an rpg is more akin to art than science, and the model is one of collaborative creative endeavour rather than science.

It's an analytical, critical tool; expecting it to behave like a scientific theorem is unreasonable.

Now, like other critical tools, you can say it's nonsense, or basically not useful to you (which, amongst others, Bruce Baugh has said), but to try to demolish it as unscientific is futile.

Trying to turn it into a scientific theory, with psychological evidence and experimental predictions strikes me as trying to teach the proverbial pig to sing. It wastes your time and annoys the pig.

It works fine for me as a critical tool to apply to play, and games design by extension. Better than any other I've seen, in fact.

And as for "indoctrinating"... well, would you rather we let folks spin off discussions on the board into pits of oblivion of wrangling over the meaning of individual words, or get concensus on what the terms mean and enforce that consensus? Maybe I've emotionally loaded that last sentence too much, but I think my feelings on that are clear, if only because my early posts are full of stuff "agreeing" with the model, to be shot down by people who understood (and created) the model.

But, if it's got you thinking critically about play, even if you disagree with my imterpretation of your play... great! If you put out your own competing theory or modification to GNS that you champion against whatever we got going... well, you'll have a lot of company here. I disagree with Ron about the fundamental emotion of Sim play. Some days on the boards it looks like sim play is the "pure" form of RP, and the other two are weird offshoots, other days sim looks like a minor sect populated by poor fools who can't commit to either form of "proper roleplaying."

If we're singing from the same hymn sheet, we can barely agree what key we're singing in, let alone the tune.

[jdagna]

For example, Pete (undoubtedly in perfectly good faith) interpreted apparent instances of Game-oriented and Narrative-Oriented in Superheroes as really being Sim play to support the idea that a game session might have been all Sim. Presumably another person might have interpreted the apparently Sim-oriented and Narrative-oriented play as crypto-Gamist. In such circumstances surely conclusions will be driven more by faith than by empiricism. Even if I deliberately set up a play session to test GNS by trying to facilitate, as far as possible, an equal mix of Sim, Game, and Narrative play and then reported the result, I fear that the result would be simply be brushed aside by the above process of GNS reclassification.

I agree with most of your criticisms and concerns, and I think Pete followed up with mostly the same thing I would have responded with.  However, I thought I'd add my few cents in regards to this section, since its one that seems to come up a lot in regards to Sim play.

I'm convinced that one of the reasons these differences in interpretation occur results from hearing summaries over the web.  There are a lot of GNS clues that show up very clearly when you're there in person that don't get represented in a post (like a sigh or a grin at certain points).  

Additionally, offering alternative interpretations often helps people who are still learning the theory to understand the finer points.  For example, it's easy to think that anything that looks like a soap opera must be  Nar - you've got drama, morality, and it looks like a lot of games that you know are Nar.  However, it's possible to get that same game situation out of Gam or Sim play.  It depends on the priorities of the players.  If they're into the soap opera to address premise, it really was Nar.  If the soap opera is just a social competition to be won, it's Gam.  And if you're just faithfully recreating the soap opera genre, that would be Sim.

And that's where prioritization comes back in.  A player in the Sim mode in this case might enjoy the challenge and might notice a theme developing, but neither are priorities if he focuses on recreating the genre and chooses to do that when he comes to a crossroads.  

By the way, I'd encourage you to try running a game that uses all three modes equally, especially if you don't tell the other players what your goal is.  My guess is that you'll confirm the theory by sowing mass confusion and wind up with a train wreck of a session as expectations and priroties shift and collide.  (Either that, or your players will doggedly stick to whichever one you started with and wonder what's wrong with you).  In my mind, that's all the verification the system needs... because whether G, N or S are really THE modes, the fact that modes of play exist to represent different player priorities is a pretty well-established fact, as are the conflicts that can result from lacking a language to discuss those differing priorities.

[Mike Holmes]

Wow, well said, Justin. I think you've just created the test of the theory. It would require that it be applied blind in order that bias not play a part (I can't be the tester), but I think the outcome would be obvious. Then again that might just be the bias talking again.

Mike

[greyorm]

Marcus,

...subjected to intensive indoctrination in it...

What are you saying here appears to be that unless a reader/student can grasp something immediately, it must be false and only comprehensible through brainwashing?

Consider quantum physics in that light: Those damn quantum cultists pitching their ridiculous, convoluted theories. Doesn't say much for the theory that it requires years of study to understand.

Please forgive any perceived snippiness that might come out of that statement, it's just I found the "intensive indoctrination" bit rather an insulting implication to so casually ascribe to a group of individuals you barely know.

There are fine points to the theory, as there are with any theory, that are only processed by the student after they've learned to remember to apply them...in other words, it takes a while to get it all down within the framework, but I hardly consider that "indoctrination." After all, Gods forbid something require a bit of work on the part of the reader in order to grasp it!

Moving on...

Justin hit most of the points I would have pretty well. I do have a couple of points to add from my own perspective. Take them as you will.

I have a further concern about GNS theory that it is really untestable, in that the definitions of the various outlooks are loose enough to allow many types of play to be classified in different ways

All I have to say is that it never has been for me. The differences in seemingly similar style, such as that between "Exploration of Situation" and "Narrativism" have been quite obvious from play, for the same reasons Justin brings up about prioritization.

Reading over your commentary in various places on the boards thus far suggests strongly to me that the problem is not with the theory being too "loose" with definitions, but with your own understanding of what the terms actually refer to causing some small measure of problem.

Remember, the terms are non-intuitive, they're placeholders for ideas, rather than meaning exactly what they appear to be. I am led to perceive you are still approaching the terms as the terms, rather than as their defintions.

FrEx, your approach to the issue of the "newsletter" created for your game and how it is all three modes at once, and your mislabeling/misuse of the G, N, & S terms for specific activities/behaviors which are not part of those styles (particularly "what this is" in specific situations based on non-GNS criteria) are telling.

To expand upon the newsletter example: it doesn't matter what it COULD be used for, it matters what the players actually do with it, and that's where prioritization comes in.

This, along with what behaviors actually go along with the three modes, are some of those fine points often not immediately apparent. IE: Yes, you're doing THIS, but why is the question we're looking at, rather than what category is this?

As a method of predicting what sets of rules or styles of play are likely to succeed in producing a felicitous result, however, I have grave doubts.

Well, it worked for me. Which is all that really matters when you get down to it.

[M. J. Young]

I want to thank the Reverend for reminding me that I wanted to comment on that newspaper thing. He's pretty much nailed it: it isn't that you have an in-game newpaper, it's why.

In nearly every game I play, there's someone (frequently me) writing something that stands as a record of what's happening in the game, or in the world. Making it part of the world (I've done journals, reports to superiors, letters to loved ones, restrospective history books, and more) doesn't do more than create another detail in the world. However, in reading what is written you can often discover much about what it is the players are prioritizing.

One thing that we see (and do) a lot of here is the sort of Monday morning quarterbacking of trying to figure out what someone else's playstyle is from their descriptions. There's a lot of disagreement in these threads when that happens. For example, I was thinking that you were probably playing gamist from your description, but when I saw Pete suggest you were simulationist I realized 1) that trying to analyze your play style from the available information required filling in a lot of gaps and 2) that such a discussion at this point would only cloud the issue. It seems like you might be playing in any mode, but ultimately you're probably actually playing in one.

I'm also beginning to think that familiarity with the theory colors the telling of the tale. Let me explain what I mean.

I can tell some great game stories from before GNS was published, stories which I've told many times before, in which whatever it was that was really exciting about the game at the time comes through clearly, and we see that this game was gamist, that one narrativist, the other simulationist. However, when I read accounts of people's play styles here, they're always laced with, "I thought we might be playing gamist because...but then this part seems narrativist because...and of course there's all this going on that fits with simulationism, because...." The stories people tell here aren't what excited them about the game; they're efforts to marshall evidence that supports each of the three creative agendae and try to figure out how much of each was actually included.

That's why I say you have to focus on the decisions that mattered to the players. When they create those newspapers, are the excited about how realistic these seem, with classified sections, comics pages, a Bridge column, and full page ads for department stores, along with national and regional news and an editorials section, not to mention sports, fashion, and entertainment? That may be a little extreme, but there are players who would very much enjoy just creating that newspaper, like model railroaders setting up electric lights that turn on and off on a timer set to the day/night cycle of their lines, or model shipbuilders carefully painting faces on the backs of each of the four hundred cannon on the huge galleon they're constructing because some spanish ships had these visages. Or are your newspapermen thrilled by the idea of something that looks like a newspaper hiding clues for them to ferret out of the ordinary news, so they can get a jump on the enemy? Or is it perhaps rather that the editorial slant of the paper puts the player characters in sharp contrast to the opinions expressed, such that they become the core of the issues being raised here, people who think they're doing the right thing who are being characterized publicly as villains? There are a lot of ways in which the careful meticulous construction of a newpaper can enhance gamist, narrativist, or simulationist goals. The question is, what is it that really excites the players about this? Sure, the paper might support all three agendae; but to which of these are the players actually responding?

I think that should help.

Oh, and the Rev is right--your comment about indoctrination being necessary to understand, even if meant in a joking manner, was a bit harsh. I believe that Relativity is a sound theory. I get letters from people frequently who want to argue that Einstein was wrong, but I generally write back and tell them that they should throw out their laser-driven CD player because lasers can't possibly work, and get the government to stop wasting money on particle accelerators since the time dilation effects the create don't happen, and work to stop the massive propaganda campaign that the Japanese have mounted to persuade the world that we dropped atomic bombs on them, since these are patently absurd. There is much about the model that is not intuitive, and a lot of people who objected to it and who still object to it for that reason. However, the combination of clarifying that the model and what you understand the model to be are not exactly the same thing with real evidence that the theory is able to predict outcomes (and there is such evidence for GNS--Legends of Alyria is a phenomenal example of the theory in application) usually eventually gets people to recognize the value of the model.

--M. J. Young

--M. J. Young

[marcus]

First of all, I apologise to those I offended with the term "intensive indoctrination" in my earlier post. I was not using that term in its perjorative sense, but rather in the neutral and literal sense of "thoroughly steeped in the doctrine" of GNS. My point (with which you may all disagree) was that if almost all role-players when first hearing the theory respond "that's wrong" and it takes weeks, months or years of clarification, exemplification and assertion of the theory through this forum to talk them round, then I would say that is a danger sign for the theory. Even persuading someone of quantum theory would be an easier job than this- the results of the two-slit diffraction experiment should be enough to underline the case.

Secondly, the responses to my last post on the question of verification of the theory might be summarised (I hope not too innaccurately) as two suggestions how the theory could be verified in future from observation, one claim that it is futile to try to verify it as it isn't a theory that makes predictions, and one statement that the theory has already been verified "because it works for me". These responses do not exactly fill me with confidence that the theory has ever been put to the test in an empirical fashion.

When I talk about testing the theory empirically, I am only referring to its predictions as to what will make for good gaming. The taxonomic and terminological aspects of GNS- how it classifies games and gaming behaviours and the new terms it has created to facilitate discussion- are not, of course, amenable to empirical testing as they make no statements about the world but are merely a set of definitions. A person can, of course, define words as he sees fit- there is no truth or falsity in a definition. The only tests in this case are internal consistency and usefulness. GNS taxonomy and terminology appears to me to be both internally consistent and useful (although the difficulty of determining into which category a given behaviour falls limits this usefulness to a degree), but such may be the case with many other such systems. If GNS is to be more than a set of terms and a school of game design claiming no priority over any other school, it must show that games designed ignoring its dictates will inevitably be poorer, and games observing its dictates are inevitably richer.

My final comment follows up on Pete's opening words of his last post:

Marcus, to a great extent you're right: the model is, to a large, degree non-falsifiable, and it's supporters are filled with ranks of guys who will tend to say "You haven't understood it" to anyone who disagrees with it.

This a real problem I have encountered when discussing this topic. There is a sense that only those that have already fully adopted GNS theory are qualified to discuss it. This practice makes GNS theory all but invulnerable to criticism.

If I do not understand GNS theory, it is not through want of trying. Prior to starting this thread, I read Ron's original "System Does Matter" essay, his 1991 GNS essay, and his essays on Simulationism and Gamism. There were then about half a dozen threads on this forum to which Ron directed me (including the recent "This is it" rephrasing of the model)- I read them all, right through, and several others beside. All this took more than six hours over the space of several days. Further, I would like to think that I am reasonably literate man. As a commercial lawyer I am used to construing the most complex of definitions and the most turgid of texts. I am also well familiar with roleplaying, it having been a major hobby of mine over 25 continuous years, during which I have seriously played close to 50 different systems, read but not played perhaps 20 more, and even designed a couple of systems myself (admittedly only for in-house play). Now I know that some of you are published RPG authors, and may very well have experience in gaming far greater in both quantity and quality than my own, as well as being further qualified by years of discussion of RPG design theory, and that I may in truth be a minnow amongst a pod of whales, but my point here is that my background should at least permit me to gain a reasonable degree of comprehension of the words I have read in my recent labours.

In short, I would suggest that if a literate long-time role-player like myself, after 6 hours of solidly reading all the major essays and all the key discussion threads, has so thoroughly misunderstood GNS theory, then I would respectfully suggest that perhaps it might be stated a little more clearly.


Marcus