News:

Forum changes: Editing of posts has been turned off until further notice.

Main Menu

Big Model as a Scientific Theory

Started by Wormwood, September 29, 2004, 12:32:00 PM

Previous topic - Next topic

Wormwood

Note: this is not a critique of accuracy. Rather it is an attempt to point out places where the structure and formalisms of the Big Model Theory cause difficulty in comprehension and application. It has been one of my goals for the past year to develop a more solid and scientific rephrasing of this theory.

The following are some major problems I see with how the Big Model works as a scientific theory. I welcome any comments and criticisms. I am confident that most of these can be removed, without sacreficing the utility of the theory.

1) The theory is based on statistical observations of play, but lacks a language to account and certify these statistics. This has shown itself most in discussions about what an instance of play is. What is needed here is a sense of statistical significance, hypothesis testing, and uncertainty.

2) There are several overloaded concepts, most notably that of social contract. On one hand it is the feedback loop which permits the dynamics of play, and on the other hand it is the superset of actions and events in play. This confounding means using this term must be done very carefully, to distinguish between static and dynamic contexts, and makes invalid notational abuse far to easy.

3) The theory is grounded in a top-down approach, assuming a gestault understanding of many complex events as the first step of the theory. This causes significant danger since human social perception is being used as an objective observer.  A bottom-up approach, for example, would permit a more direct accounting of observer bias.

4) The theory does not propose any experimental tests, and due to above points removes nearly any power that such tests could have, by enforcing a subjective basis for analysis.

5) The theory separates design from play, which causes inconsistencies in discussing the use of the theory for design applications and understanding drift.

6) The theory assumes non-existence of elements which are rare or artificially designed. This is best shown by the discussion of CA hybrids.  This is better handled by discussing relative populations and theoretical requirements of these elements.

7) The theory does not take into account itself and its effect on play, and in particular it lacks a definite way to do this without simply being intractibly self-referential.


Thank you for your time,

   -Mendel S.

clehrich

There was some discussion of this in a thread I started quite a while ago: More on jargon and models.

My contention there was that the Big Model is not scientific, but tries to be, and that making it so isn't worth doing.

Why do you think a scientific model, which would indeed revise the Big Model considerably, be useful?
Chris Lehrich

Jere

Quote from: clehrichThere was some discussion of this in a thread I started quite a while ago: More on jargon and models.

My contention there was that the Big Model is not scientific, but tries to be, and that making it so isn't worth doing.

Why do you think a scientific model, which would indeed revise the Big Model considerably, be useful?

To go further I think the "Big Model" would be mroe valuable if it was recognized, and develoepd, as a manifesto, which it is, rather than attmpting to legitimize itself as some "scientific" model.

Jeremiah

Ron Edwards

Hello,

I should point out that I have never made any such claim, whether for the Big Model, or any other RPG-based theorizing I've done.

Best,
Ron

clehrich

Two thoughts:
Quote from: Wormwood5) The theory separates design from play, which causes inconsistencies in discussing the use of the theory for design applications and understanding drift.
Wouldn't it be simpler to separate these entirely, unless and until it can be proven that they are identical?

Quote6) The theory assumes non-existence of elements which are rare or artificially designed. This is best shown by the discussion of CA hybrids.  This is better handled by discussing relative populations and theoretical requirements of these elements.
A polythetic classification system might help here, with some mild prioritization of elements.  Such a system might also be used statistically to find previously unrecognized conjunctions of seemingly independent factors.
Chris Lehrich

clehrich

Just to be clear:
Quote from: Ron EdwardsI should point out that I have never made any such claim, whether for the Big Model, or any other RPG-based theorizing I've done.
I assume the "claim" here is that the Big Model is a scientific model.  The point being that Ron has never said that it is scientific, and has no intention of making it so or claiming that it is so.

My own contention is that it is constructed in such a way that it does make some folks think it's scientific, or that it's supposed to be.  I think this is unfortunate.

Mendel, correct me if I'm wrong, thinks that it isn't scientific but could and should be made so.  This is not in disagreement with Ron.

Sorry, Mendel.  Back to your thread.
Chris Lehrich

Sean

I think the Big Model is a good example of what might be called a pre-scientific, or informal scientific theory. That is, it does aim to be descriptive, and whatever norms it prescribes are relative to certain goals that someone playing an RPG might have. It's not a 'manifesto'. A lot of problems that arise when people try to interpret the theory IMO are that they read it as a manifesto, based on their own preconceptions or things they've heard Ron or others say in other contexts, perhaps.

Mendel, if you can carry out any part of that research program, more power to you. In particular, it would be hell of cool if we could get people to really analyze transcripts of games they were present and taking notes at, as spectators or, failing that, players, with several different groups, and look at the psychodynamics of the group interaction. This could be done both on a social-transactional level and on a level of what the participants in the game are looking to get out of the game.

There are sociology departments out there which would regard such an effort as a legitimate PhD thesis project. I'd love to see someone do this. I know there's at least one person out there working on the sociology of RPGs (a player in MAR Barker's long-running Thursday Night Group, in fact), but I don't know if he takes an interest in Ron's work or not.

It's a big project, like most serious work. But saying 'it's not science' or 'it should be science' or 'it is science' or 'if it's not science, and it's not, it must be a practical manifesto' are all equally, well, sort of useless from a serious intellectual standpoint. It's a seemingly coherent set of ideas with some empirical basis and some guesswork that's not obviously false and seems to illuminate some aspects of play. There are a lot of theories like this in the human sciences, actually.

If you can extend or revise these ideas, or put them on a firmer basis, lots of people who take an interest in RPG theory will be happy about it. Good luck!

Wormwood

The major advantages of a scientific theory is the ability to verify results via experiment and also to provide an underlying explaination for these results. To reiterate what Chris said, I'm not claiming the Big Model is meant to be a scientific theory, rather I'm looking for feedback on what distances the model from such a theory.

Chris,

(5) I assume you mean design and play? I think that keeping those two at a distance actually encourages the attitude that design is both rarefied and somehow more authoritative than play. I think far more game design is done overtly or covertly within the context of play than on its own. Drift also ties strongly with game design, for example, most of Ron's Fantasy Heartbreakers can be explained as codification of drift in some variant of D&D.

(6) A scientific theory should be able to distinguish between unobserved due to rarity or difficulty and unobserved due to impossibility.  What you describe is exactly what can be done with such a theory.


Sean,

I'd probably call it a heuristical model, acting as mental shortcut to cut through the general clutter of RPGs.  As a valid heuristic it isn't required to explain why, or be able to be invalidated via experiment. Any experiment or explaination which contradicts it is perfectly valid, it simply falls outside of the scope of a heuristical model.

The first real step to bringing the Big Model further along is to clarify where the heuristic breaks down. I would suggest that some of my points above are a few of those places.

As far as putting things on a firmer basis, that is the next step.


Thanks,

  -Mendel

clehrich

Quote from: Wormwood(5) I assume you mean design and play? I think that keeping those two at a distance actually encourages the attitude that design is both rarefied and somehow more authoritative than play. I think far more game design is done overtly or covertly within the context of play than on its own. Drift also ties strongly with game design, for example, most of Ron's Fantasy Heartbreakers can be explained as codification of drift in some variant of D&D.
I think I get this.  You're saying that things like drift and other manipulations of game structure that happen during play can and should be read as forms of design.  Since there is no clear ground upon which to cut these off from play in a broader sense, it is dangerous to assume that design that happens to occur outside of game-time is somehow a special, different mode.  Consequently the model should begin with play and assume that everything under analysis is some mode of play.  Is that right?  Actually, Ron hinted at this somewhere or other where he talked about reading rules as part of play, but I don't know that anyone's taken it quite to this extreme.

Boy, that's slick.  I like it!  I'm not at all sure how this is going to work, since I have a suspicion that it's not going to come out looking at all like the Big Model, but it sounds very promising.  Kinky!
Chris Lehrich

ErrathofKosh

I began designing games in lieu of playing when I lost my group and got tired of reading my gamebooks and making new characters.  So, perhaps, design is an extension of play... Players design characters, GM's design settings and situations; those are considered play, so why wouldn't the game designer designing system be play?

But, back to the subject at hand...  (A little commentary, please ignore if it goes too far off-topic.)

Cheers
Jonathan
Cheers,
Jonathan

timfire

Quote from: ErrathofKoshPlayers design characters, GM's design settings and situations; those are considered play, so why wouldn't the game designer designing system be play?
Jonathon, I started another thread to discuss that idea over in the Theory forum.
--Timothy Walters Kleinert

Wormwood

Jon,

The relationship between design and play is more a matter of nearly all play includes design (overtly or covertly), rather than design including play. I'll post other comments on the split thread, just want to distinguish the two ideas.

 -Mendel S.