GNS - does one preclude another?

[Ourph]

Models resemble the things they are modeling. An economic model predicts market behavior.

Perhaps there's a difference in nomenclature between different fields, but in the fields I'm familiar with (Psychology, Biology, Neuroscience) there is a clear distinction made between a model and a hypothesis/theory. Models are purely descriptive. Models describe the known interactions of their components, they do not predict future interactions. Future interactions based on the current model are the realm of hypotheses. If you submit a research manuscript to a neuroscience journal that claims your model "predicts" an outcome to a certain experiment, the action editor will write back to you and tell you to resubmit once you graduate from basic vocabulary school. (And, no, that's never happened to me before.... )

There are numerous claims made by Ron Edwards that can be compared to reality. It is "not predictive" only in the sense that it is a very bad model.

Ron Edward's personal opinions on gaming don't necessarily equate to predictions made by the Big Model. In addition, there is a distinction to be made between using a model as a basis for making predictions and the model itself being predictive. There is also a distinction between a model being diagnostic (diagnosis is just another name for description) and a model being predictive. Ron does claim that the Big Model is diagnostic.

 

[Umbran]

A scale model of the Eiffel tower resembles the Eiffel tower. All models have predictive power.

That is not correct. Whether or not a model has predictive power depends on how it resembles reality.

That scale model of the Eiffel tower, or a kid's plastic scale model airplane, does not necessarily resemble the Eiffel Tower except visually. Unless you want to claim "show how the real tower looks" as a prediction, such models have no predictive power.

In order to be predictive, there needs to be a mechanism for the model to take data of a circumstance and then present some result that predicts how reality will behave.

So, for example, with an atmospheric model, you feed the model some hypothetical or real data, and it spits out what tomorrow's weather will be. With a structural model you put the model under some load, and you watch how it performs, and you draw analogy to how a real structure will behave.

GNS lacks any place to hook in your data to get a prediction out.

 

[BenBrown]

That is not correct. Whether or not a model has predictive power depends on how it resembles reality.

That scale model of the Eiffel tower, or a kid's plastic scale model airplane, does not necessarily resemble the Eiffel Tower except visually. Unless you want to claim "show how the real tower looks" as a prediction, such models have no predictive power.

In order to be predictive, there needs to be a mechanism for the model to take data of a circumstance and then present some result that predicts how reality will behave.

So, for example, with an atmospheric model, you feed the model some hypothetical or real data, and it spits out what tomorrow's weather will be. With a structural model you put the model under some load, and you watch how it performs, and you draw analogy to how a real structure will behave.

Does the GNS model have such capacity? Or is it more like a 30cm tall model of the Eiffel Tower: pretty to look at, but not much use otherwise, except possibly as a paperweight.

 

[Umbran]

GNS does purport to predict many things. It purports to predict:
- That a given preference of play style can be satisfied by G-N-S elements
- That the definition of G-N-S are elements that constitute G, N, or S play

These are not predictions. They are assumptions the model requires.

- That it is more useful than other models for describing what happens in play and prescribing remedies to dysfunctional agendas

No, the model itself does not "predict" this. That the model is more useful does not actually fall out of the model itself.People who like the model assert it.

- That coherent play is better than incoherent play

Again, this is an assertion or assumption of the model, not something that comes from the model.

 

[Umbran]

Does the GNS model have such capacity? Or is it more like a 30cm tall model of the Eiffel Tower: pretty to look at, but not much use otherwise, except possibly as a paperweight.

To me, it is somewhat better than a paperweight. It is, at its best, a conceptual framework we can use for purposes of analysis and consideration. it does not in and of itself, predict anything about a particular game, player, or GM, but it does give us a point of view from which we can think about games, players, and GMs.

Consider, if you will, tarot cards. If you know the symbolism of the cards, and you lay them out, you can produce a picture of your past, present, and future. It is a vague picture, but a picture it is, and pictures are models.

Are the tarot cards themselves predictive? I will not make a claim on that score. What I will claim is that even for those who don't believe in their predictive power, the cards can provide some clarity and food for thought. The picture provided gives you hooks onto which you can tack your own reality. By forcing yourself to think in the terms of the cards' framework, you think a bit beyond your own usual mental paths and ruts - "out of the box" so to speak, and you may consider things you'd otherwise pass over, or consider giving higher or lower priority to real things than you would have without the cards.

That's what GNS (and GDS) theories are, as far as I am concerned. Frameworks to think from that may not be what we are habituated to - formalisms that may help shed light on what we are doing.

 

[resistor]

"those groups" don't exist. They're a figment of your statistical imagination. Those categories are too broad and ill-defined to be remotely empirical.

So, your response to statistical evidence showing that there are "clusters" of preference that can be loosely described by the above descriptors is to stick your head in the sand and yell "nuh-uh!"?

------

I'm going to refer you Malcolm Gladwell's excellent talk on
what we can learn from spaghetti sauce.

TLDR version: Way back when, American pasta sauce companies made two kinds of sauce: basic, and spicy. One of the companies was concerned, because their sauce wasn't selling as well, even though focus-testing showed that it had all the desired qualities of a basic pasta sauce.

So, they hired a consultant, who did extensive surveying and clustering analysis, and identified a third preference group that never showed up in focus groups, and that people were identify through self-introspection: extra chunky. The company took his advice, introduced an extra chunky variety, and sales took off.

 

[nedjer]

So, your response to statistical evidence showing that there are "clusters" of preference that can be loosely described by the above descriptors is to stick your head in the sand and yell "nuh-uh!"?

At the point when you call something "loosely described" science I'm hoping like hell you don't work in the nuclear industry.

 

[nedjer]

Perhaps there's a difference in nomenclature between different fields, but in the fields I'm familiar with (Psychology, Biology, Neuroscience) there is a clear distinction made between a model and a hypothesis/theory. Models are purely descriptive. Models describe the known interactions of their components, they do not predict future interactions. Future interactions based on the current model are the realm of hypotheses. If you submit a research manuscript to a neuroscience journal that claims your model "predicts" an outcome to a certain experiment, the action editor will write back to you and tell you to resubmit once you graduate from basic vocabulary school. (And, no, that's never happened to me before.... )


Ron Edward's personal opinions on gaming don't necessarily equate to predictions made by the Big Model. In addition, there is a distinction to be made between using a model as a basis for making predictions and the model itself being predictive. There is also a distinction between a model being diagnostic (diagnosis is just another name for description) and a model being predictive. Ron does claim that the Big Model is diagnostic.

A very useful distinction.

 

[Umbran]

At the point when you call something "loosely described" science I'm hoping like hell you don't work in the nuclear industry.

Yeah, well this isn't the nuclear industry. We are talking about something that's far more akin to sociology than physics. People are not fermions, and cannot be described so precisely.

 

[pemerton]

I think that the Forge's Big Model (which includes an analysis of GNS as creative agendas) is more useful for gaming than are Tarot cards for life-coaching. (But then, maybe I'm being a little bit hard on the cards - I've got less experience of them than of the Forge.)

Useful things that I have got from Edwards's GNS essays, and associated Forge threads/blogs:

*A better understanding of the aesthetics, rationale and limitations of Rolemaster (my RPG system of choice for many years);

*A better understanding of my dislike of AD&D 2nd ed play, and similar 90s playstyles, and how I can avoid my game turning into that sort of play (this is the subject matter of Edwards's notorious "brain damage" comments);

*A better understanding of the significance of character creation, and the missing elements of character creation (especially relationships) in many traditional fantasy RPGs - it was not until reading Edwards that I consciously appreciated that party integration and the interaction of PC goals with the gameworld and with one another are a huge part of traditional fantasy RPG play - far more important, in my own play experience, than hiring torchbearers or buying songbirds - and yet the typical traditional rulebook has nothing to say about it for either the GM or the players;

*A better understanding of the difference between "simulationist" (physics of the gameworld) and metagame action resolution mechanics, and the way they work in various situations;

*An intelligent introduction to modern fantasy RPGs like The Dying Earth, The Riddle of Steel, HeroWars and Burning Wheel;

*Better understanding of the rationale and play of games that I've only read the rules of and will probably never play, like Nicotine Girls and The World, the Flesh and the Devil - even though I've not played these games, understanding what they do and how they do it has helped me in GMing more traditional games;

*If it was not for the Forge, I probably couldn't have made the transition from GMing Rolemaster to GMing D&D 4e - it was the Forge that gave me the necessary conceptual tools for understanding what 4e is about, and the rationale of its various metagame mechanics (hit points, action points, encounters/dailies, etc).​

As for the model/theory/prediction thing, I think importing analytical notions from the natural sciences into the humanities is not all that helpful. As an example: the apparatus of Weberian social theory, or the later critical theory that incorporates Weberian ideas, is not predictive. Nor is it descriptive, in any straightforward sense. It is interpretive, in the sense that it illuminates aspects of our history and our social formations, drawing connections between them that hitherto we had missed, and in so doing helps us better understand contemporary social phenomena and trends. The Big Model for RPGs is similar - it offers an interpretation, and therefore a deepened understanding, of our RPGing practices.