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

Main Menu

Shared Imagined Space, Shared Text

Started by Victor Gijsbers, April 02, 2005, 03:58:09 PM

Previous topic - Next topic

Victor Gijsbers

Hi Sean, thanks for your interesting post. I'm not sure if I understand correctly what you are saying, so please correct me if I'm mistaken.

It seems to me that you are assuming that I have two main motivations:

1. I am concerned about the fact that system only restricts the public interactions, not the private contents of the players' minds.
2. I am concerned about the metaphysical or epistemological status of a shared image, and believe a shared text to be less problematic.

In reality, I do not much care about either; and in fact I do not believe that the SIT has any advantages over the SIS on either count. Now I may be overlooking the fact that the things I have identified as my concerns are, at bottom, really these two worries; if anyone pointed that out to me, I would drop all my theses.

Let us assume for a moment that there is both a Shared Imagined Text (here defined as the shared remembered list of accepted statements) and a Shared Imagined Space (here defined as the [class of] possible world all players believe the game world to consist in). We are not interested in metaphysics, we are interested in what is going on in the process of roleplaying. Then, what I would be claiming is that:

1. Sometimes, we add statements to the SIT without any change taking place in the SIS. This can be a meaningful and important element of roleplaying, and is in fact constrained by the system.
2. Sometimes, we add statements to the SIT and (through that) elements to the SIS, but whereas we accept that the SIT now has this element, the addition to the SIS is 'bracketed'. It can still be changed in the future, without invalidating our present agreement. [This is the case of the unreliable narrator.]
3. Sometimes, we are working with a lot of SIS's, even though there is always only a single SIT. For different SITs would be different games. [This is the case of inconsistent of explicitely ambiguous texts.]
4. Sometimes, our actions change the SIS in radical ways, destroying many elements and changing lots of others. Meanwhile, however, we are still simply adding lines to the SIT. So although anything about the SIS can be changed at any point of play (you can always add the lines"Here Frank stopped talking. The shaman looked him deep in the eyes, and remained silent for a long time. 'So that is how the story was told to you?', he ventured at last. 'The truth was very different. Let me tell you the real story.'" to any SIT), the SIT is solid, and continually added to.

I believe that these four points make the SIT more 'fundamental' than the SIS; not from a metaphysical or epistemological point of view, but simply as an underlying notion to understand the actual process of roleplaying. Would you agree with me that this is a valid concern, quite different from the two motivations I identified above? If so, have I correctly understood your post to be about those two motivations, or have I misunderstood it and does it address the real motivations which I have just explicated?

I'm not married to the terminology we are using here. If anyone has better word than 'text', by all means tell me. I could even be happy with the term Shared Imagined Space for what I am calling the SIT, but currently I need to define the SIS as I did above in order to keep the distinction I wish to draw clear.


Hi Victor,

The summary of your post seems solid to me, though I have two caveats.

1) The 'Shared Imagined Text': doesn't it break down into individual imagined texts? That is, not everyone hears everything, remembers everything the same way, interprets what was said the same way, etc. I would probably be willing to grant that there's more cohesion at the level of the shared text than at the level of the shared imagined space, as you're breaking those two things down,

If so, then making it 'more fundamental' seems a little misleading. Which leads me to

2) I'm not sure ultimately that the distinction between the 'text' and the 'imagined stuff' is a legitimate one, which is why I tried to go all Wittgensteinian on your ass in my last post. That is, 'lines' in the 'SIT' are, among other things, expressions of material of the participants' respective IIS', aimed at bringing others' IIS' more in line with theirs, and imposing/sharing their vision on the group construct, which we're here calling the SIT, but for which we can imagine a single abstract 'author' whose 'imagined space' the SIT expresses much as individual lines of it can be expressions of individual participants' IIS.

So you might say that on my view there's an inflected whole which includes several IISs and an SIT in which they get expressed and which makes a feedback loop into the individual IISs. If you want to say that in a certain sense the IITs and the SIS are 'reflections' of these two 'more fundamental' things, I might wind up agreeing with you after all. But since I think that as it's most often used 'SIS' doesn't mean what I just called 'SIS', the imagined space of some individual author-construct on the text produced by the whole group, but rather means this whole negotiated feedback system, I think there's an ambiguity here which is not necessarily helpful for dialogue at the Forge; especially because the point of the term is normally to call people's attention to this dynamic as the underlying stuff of role-playing, and not to anatomize it.

On the other hand I think its anatomy is interesting. So on the one hand I think you 'get' what Ron, Vincent, etc. are talking about, at least insofar as I do. On the other hand, if you want to start breaking it down in more psychological and/or philosophical detail, that could be an interesting project too. But I wouldn't say, if you wound up doing the latter, that you'd be rejecting the idea of the SIS as it's actually used by some prominent Forgites, but trying to pick it apart in more detail, as Vincent has done using rather plainer language than we're aspiring to (and this is IMO to Vincent's credit) in his excellent lecture and a lot of the material on 'anyway'.

So I don't disagree with your 1-4 but I think they're only part of the picture and I'd want to talk about all that stuff as a finer analysis of the existing concept of SIS rather than as a replacement concept.

Another note: are you familiar with Kendall Walton's Mimesis as Make-Believe? I think that for a 'possible worlds' analysis of RPGs which connects it to art more generally, this is a good place to start, with interesting points of contact with your view.



P.S. I wouldn't want anyone to infer from my previous post that I regard metaphysics as boring or a waste of time, since I don't. But I do think that many times when metaphysical issues (which are under discussion here: we're asking 'what is the real nature of the SIS' IMO) come up in theoretical discussions they wind up being unhelpful or misleading, though not always. This discussion may actually turn out to be an exception.

Ron Edwards

It's helping me.

I've decided this is one of those discussions which won't benefit from my input, but contrary to my last post, has great value.

Thought I ought to say that and give credit to Victor.


Jonathan Walton

Hey Folks,

Coming to this discussion late, since I'm not reading the Forge regularly anymore.

Seems to me that a piece of what you're getting at, Victor, is a distinction between the communicated content of play (what gets negotiated between players) and the context of play (which informs players how they should interpret the content).  My chief interest in roleplaying theory is to come at these issues from the perspective of cross-cultural contact and aesthetics.

In roleplaying, both the content and the context are negotiated, and then you have individual imaginings, which are personal and non-negotiable (though they're obviously influenced by the outside, so maybe it would be beneficial to view them as negotiated too?).

What's been called "Shared Imaginary Space," I tend to think of, nowadays, as the historical context of the production and reception of play, something that is unique to each individual (based on their roleplaying and life experiences in the past) but continuously made more similar through interplayer contact before and during the game.  This is the "space" in which roleplaying events (or, the "text" of roleplaying) take place: the context in which they are to be understood, a negotiated context created and developed by the specific mix of players involved and the coming together of their individual experiences and expectations.  This is not really a "shared" context, but individual contexts that are temporarily "similarized" (what's a better word for this?), since interpretations are ultimately individual.

It's also important to note that the context that surrounds a given work undergoes adjustments while the process of production and reception goes on, due to the experience gained from creating or appreciating the work in question.

Note: this view of SIS as context is pretty dramatically different from Ron, who's said (correct me if I'm wrong, Ron) that he views SIS as synonymous with what he calls Exploration.  However, I tend to think that the "space" in SIS is meant not to represent content but a place in which content can exist, and to me this suggests context as the culpret.

These be my ponderings of late.

Edit: P.S.  I tend to think that the jargon term "Shared Imaginitive Space" has pretty much worn out its usefulness, simply because people use it to mean so many different things.  It's obviously getting at some important ideas, though, so it may be time to reapproach this whole concept and try to get a firmer grasp on what's going on.


Hey, Victor, :)

I'm grasping yuor distinction and it gells well with some problems I've come across in the past, when trying to explain this particular piece of theory. Cool stuff.

Quote from: Victor GijsbersIf anyone has better word than 'text', by all means tell me.
A number of alternatives occur to me: S. I. Log (or Playlog) and adding entries to it; S. I. Script (or Playscript or Transcript), and adding (real world) actions to it. Both of these have their own terminological flaws, though, so take with a grain of salt.

My favorite is to simply call it 'Gijsbers Space'.

Shared Gijbers Space - the virtual collection of all the player actions which have been accepted by the group as creating credible meaning towards constructing Individual or Shared Imagined Spaces.

Individual Gijbers Space - each player's perception of the underlying Gijbers Space, and the actual collection of actions each player uses to contruct his own Individual Imagined Space.

The Lumpley principle might then be rephrased thus: System (including but not limited to 'the rules') is defined as the means by which the group accepts individual actions as Entries in the Shared Gijsbers Space.

I'd also add another principle: Entries added to the Shared Gijbers Space are automatically translated into the IGS by each player, according to his perceptiveness and level of engagement at the time the entry was added.

Those last two may have been stretching it a bit. :) Also, the above definitions may need tweaking by you, but hopefully, they demonstrate that my understading of your concept is accurate. Please correct me otherwise.



PS - "The best way to have something named after you is to create it and then not name it." - Donald Shell, author of the shell sorting algorithm (apocryphal). ;)
João Mendes
Lisbon, Portugal
Lisbon Gamer

Callan S.

Heya Ron,

I know you've left, but here's something for others to think about.

I think it's really missleading terminology to use those words that way. If a particular roleplay session consisted of everyone working on a block of clay, and you pointed to that block of clay and said "That's the shared imaginative space!" it'd really confuse me. If you said "That's the physical object/text that then goes on to inspire the imagination of each participant" then your getting me somewhere.

I think Victor is trying to seperate the actual real world object ('text') from what it inspires in terms of imagination, and to focus on that in terms of design. Keeping the real world text and imaginative space intertwined as a term is like intertwining player and PC into just the term 'PC'. Use 'PC' when you really want to just talk about player motivations, for example, and readers eyes will bleed. Likewise, if I want to talk about what actual, real world object/text was made (even if it was just a real world vibration in the air/voice) but I keep refering to SIS, readers are going to think I want to talk about what was imagined. Who cares what was imagined, when looking at the real world object that was made.
Philosopher Gamer

Victor Gijsbers

Just a quick post to let you know that I'm really excited about many of the suggestions above; answering them will take more time for thought and composition than I can give it tonight, though. (I don't want to rush it.) Sean: I wasn't familiar with the Mimesis as Make-Believe book, but I borrowed it from the library today and already read the first chapter. Interesting stuff, and it may have given me a stunning revelation about the use of 'props'. (Also needs some more thought.) Any chapters you'd say are particularly relevant for the present discussion?


Hi Victor,

First of all, I want to apologize for the negative attitude in my two previous posts. I think this stuff is really interesting too, now that I'm clear on where the conversation is going.

As far as MaMB goes, see if you can find a chapter based on the article "Fearing Fictions", which was in J. Phil in the late seventies. (If the chapter's like the article it should start with Charles in a movie theater being 'afraid' of an oozing green slime.) That article contains IMO the kernel of the book; although it presents itself as an explanation of why we feel things for fictional entities (which don't exist, etc.) the interesting stuff in that article is the way he develops the idea of fictional worlds as make-believe games. This fits together with your view of the SIS as SIT because Walton's interpretation of 'fictional worlds' more or less generates them by adding/subtracting propositions and rules for generating make-beleive propositions to the beliefs we already have about the real world.

When I teach "Fearing Fictions" I teach it alongside Gadamer's "Poetry and Mimesis", R.K. Elliott's "Aesthetic Theory and the Experience of Art", and sometimes Borges's "Narrative Art and Magic".  The phenomenon that all these articles touch on, which I call 'artistic identification' and which leads to what we call 'immersion' around here in RPGs, is one I consider centrally important to art. Walton's view provides at least the beginnings of an explanatory framework, though it needs a lot of filling out in terms of psychology and metaphysics both, and with respect to visual art and music I'm crusty and old-fashioned enough to think that mimesis is only part of the story, that there are 'purely formal' aspects to beauty in these arts that 'affect our senses' etc. Lot to discuss.

Victor Gijsbers

One thing that has become much clearer now is that Shared Imagined Space means different things for different people, but is used by at least some prominent people here as a term encompassing both what I called SIS and what I called SIT. So I would like to adopt Sean's suggestion that we are discussing the anatomy of the SIS, while keeping in mind that SIS may have been used (perhaps also in the Provisional Glossary) with something different in mind.

Second, JMendes' definitions of what he calls Shared and Individual Gijsbers Space are exactly what I was trying to get at with the notion of SIT. (It is certainly not my place to comment on the name he has given it! We'll see.) So I would propose to use those definitions in the remainder of this discussion, except for the fact that they contain clauses refering to the SIS and IIS. Let us delve somewhat deeper into that.

I regard the question of how the SGS connects to the 'imagined world' (what I used to call SIS / IIS) as fundamental at this point. Drawing on Mimesis as Make-Believe, I suggest that the SGS can be viewed as a prop to be used in a game of make-believe. When reading a book, the text is the prop that we use in a game of make-believe: the sentences in the book make us imagine some fictious world, about which we can make statements. If we were writing a story by getting a group of people together, all of whom write one sentence and then pass on the paper to the next guy, the evolving text ('shared text') would be the prop to be used in this game of make-believe. (Here, then, there is a curious but not really mysterious co-evolution of the fictious world and the prop used in imagining it.)

Analogously, we can say that the SGS - not a text of sentences, but a space of actions which may include saying sentences, moving miniatures, dressing up and showing pictures - is the prop that is to be used in a game of make-believe. Then, the game of make-believe (which in pen&paper RPGs takes place mostly in our heads) in turn inspires us to add new actions to the SGS (in order to make them props the others have to use in their game of make-believe too). This is the feedback loop which Sean identified.

The nature of the game of make-believe is still mysterious. I argued, pointing to inconsistent fiction and unreliable narrators, that we cannot simply identify that which the text or SGS generates as a possible world. (My first posts were written under the impression that the SIS has to be such a possible world.) We may perhaps call it a fictious world, following MaMB, and leave its exact nature for later investigation. The reason to call it a world is the dedication of the players to see it as some sort of whole (even if it cannot be a consistent whole); if they saw parts of the SGS as being props for different and unconnected games of make-believe, we would say that they had started playing another roleplaying game.

(At the moment, I have no idea whether it is useful to speak about shared and individual fictious worlds.)

Now, my previous arguments come down to the fact that system is the means by which it gets decided which actions are added to the SGS, as per the Lumpley Principle; that roleplaying is the continuous offering of actions as possible additions to the SGS, and the rejection or acceptation of these actions; and that the changes which these actions can make in the 'possible world' can be additions of elements to this world, but can also be wildly different things, including radical transformations. I think this is still true when we replace the notion of 'possible world' with that of 'fictious world'.

Are you people still with me at this point?


João Mendes
Lisbon, Portugal
Lisbon Gamer


I'm not so happy.

The SIS has a number of valuable properties, IMO, that seem to me undermined by distinguishing between a "text" and a "space".  In the first instance, if the text is taken to be the physical cues and props and statements in the real world with which we construct an imaginary space, then the "text" seems to me indistinguishable from the physical act of play.

Secondly I'm not sure its, umm, ontologically valid.  I don't think the text, taken as actions and statements, can be meaningfully dinstinguished from the "image" it creates - i.e. the SIS.  Because the process of inferrence and projection into the SIS is so automatic that as soon as the act occurs the SIS changes - the observing human always and necessarily draws chains of cuasality and relationship out of every change, an the SIS is in some part made of those expectations and non-explicit projections.

Third, I have already argued that there are multiple Imahginary Spaces, but only one shared space.  The metaphor I favour is that of a shared document on a network - access rights are used to control who can change it, who can view it etc.  So for me, the SIS is necessarily only those common elements found in all the individual IS's; coordination of the IS's is a necessary prerequisite for the existance of a SIS.
Impeach the bomber boys:

"He who loves practice without theory is like the sailor who boards ship without a rudder and compass and never knows where he may cast."
- Leonardo da Vinci

Victor Gijsbers

Hi contracycle, (I'm sorry, I think I knew you first name but forgot it,)

Are you using SIS here purely as an image that is shared by the participants? That is how I orginally interpreted the term in the first post of this thread, as a 'space of events', a 'possible world'. However, it seems that many people, including Vincent and Ron, are using the term in a much broader way. That is why in my previous post I stopped calling the shared image of the fictional world the 'SIS'. Maybe you could expound a bit on the meaning you would like to attach to this term?

The "text", or "SGS" (I think we are still in need of a good descriptive name), is distinguishable from the physical act of play, because it is the (ordered) collection of those acts of play that are admitted by the playing group as meaningful contributions to the 'prop' that guides the shared imagining. The physical act of play consists of actions that are:
1. proposed additions to the "text"/"SGS", which get added
2. proposed additions to the "text"/"SGS", which are rejected
3. actions that are part of the process of adding and rejecting
4. external actions (such as passing the orange juice or announcing a break)
Only the actions in 1 are in the "text"/"SGS".

QuoteSecondly I'm not sure its, umm, ontologically valid. I don't think the text, taken as actions and statements, can be meaningfully dinstinguished from the "image" it creates - i.e. the SIS. Because the process of inferrence and projection into the SIS is so automatic that as soon as the act occurs the SIS changes - the observing human always and necessarily draws chains of causality and relationship out of every change, an the SIS is in some part made of those expectations and non-explicit projections.
I completely agree with you as concerns the latter half of your statement. The "text"/"SGS" and the "SIS"/"fictional world" are closely wedded to each other, psychologically. We cannot read a fictional text without conjuring, unintentionally, unreflectively, an image. But I disagree with the claim that this implies that a fictional text and the image it conjures forth in us cannot be distinguished. Even disregarding the fact that these two things are of ontologically different categories, the processes taking place in each in the act of writing are very different. (Or rather, can be very different.) Roleplaying is always the adding of elements, in an ordered fashion, to the "text"/"SGS", but the adding of these elements does not have to correspond to adding elements to the "SIS"/"fictional world". I have tried to argue this above, speaking about massive recontextualisations, about unreliable narrators, and about inconsistent "texts"/"SGS"s. So the distinction between the "text"/"SGS" and the "SIS"/"fictional world" seems at least analytically valid, because the processes that take place in them are often of radically different and can be analysed apart from each other.

I am not sure I understand your third objection, and how it applies to what has been written above. Could you explain it a bit further?


Hey, :)

Gareth, I'm happy with Victor's response to your first two points, and the first one in particular, about the distinction between the SGS and the physical act of play.

To address your third point, I'll quote myself and expand:
Quote from: IIndividual Gijbers Space - each player's perception of the underlying Gijbers Space, and the actual collection of actions each player uses to contruct his own Individual Imagined Space.
Entries added to the Shared Gijbers Space are automatically translated into the IGS by each player, according to his perceptiveness and level of engagement at the time the entry was added.
I agree with you  that coordination of the IISs is a necessary prerequisite for the existance of a SIS. We also seem to agree that the SIS is a conceptual entity. It exists only as the set intersect of the IISs.

I hold that the SGS is also a conceptual entity, but one which exists ony as the set union of the IGSs.

What I'm talking about here is the underlying process through which that coordination comes about, i.e., how those access rights and control mechanisms you mention are implemented.

So, the way I see it, the role-playing process might be modeled thus:

Act of Play -> SGS -> IGS -> IIS -> SIS -> feedback to Act of Play

In other words, your metaphor is still valid, I'm just taking it one step further.

Disclaimer: This is getting a into cognitive science and the like, and being an electrical engineer, that is so not the area for me to be butting into. Take the above with a grain of salt.


João Mendes
Lisbon, Portugal
Lisbon Gamer

Victor Gijsbers

Let me attempt to sketch a more detailed anatomy of playing a roleplaying game, focusing on two things: the feedback loop Sean identified, and the question on what steps in the process are influenced by the 'background' (game rules; genre expectations; and so forth). The attentive reader will see that I will not use the shared fictional world ("SIS"?), nor the IGS in my anatomy. This is not because I think they do not 'exist', or because I think they are uninteresting (if you wish to talk about the stories developed in RPGs, the shared fictional world will be your friend), but because I do not think they can figure in an enlightening way in this specific model.

There are three main 'spaces' in the feedback loop.

1. The proposed actions are all actions taken in the gaming situation that are interpreted by the gaming group as proposed for addition into the SGS. This excludes actions which constitute the process of selection, as well as actions that are clearly external to the game.

2. That which has been called by JMendes, for current lack of a non-ambiguous descriptive name, the Shared Gijsbers Space. This is the set of all actions that have been accepted by the gaming group as meaningful for the construction of the fictional world. I suggest, following Sean and the literature he pointed to, to see the SGS as a prop in a social game of make-believe. The contents of the SGS tell the players what they are to imagine.

3. The Individual Fictional World, or Individual Imagined Space, is the game world as imagined by the individual player. It contains, ideally, everything the players has been told to imagine by the SGS, as well as many minor personal additions and presuppositions, expectations, value judgements, and so forth.

Then, there are three dynamic processes connecting these three spaces.

1. The Process of Selection is the social process by which the proposed actions are either accepted and added to the SGS, or rejected and discarded.

2. The Process of Imagination is the individual imaginative process that transforms the contents of the SGS into the Individual Fictional World.

3. The Process of Creativity is the individual creative process that, starting from the contents of the Individual Fictional World, adds new actions to the space of proposed actions. The aim of this procedure is to have them added to the SGS, and thus influence the social game of make-believe that is roleplaying.

We can also recognise two further elements that influence these three processes. (I'm not too happy with the names I've given them. Furthermore, I recognise that they are very diverse, and that subdividing them will in fact be very fruitful. This is meant as a very preliminary exploration.)

1. The Individual Background, which contains the creative urges, judgements and preferences of the individual player.

2. The Shared Background, which contains the rules of the game, the setting information, the genre expectations, knowledge of the real world, social expectations, and - I wish to stress this - the Structures of Credibility.

The Individual Background influences the Process of Selection in a very clear way: the player wishes to have those actions added to the SGS that are in line with his/her personal preferences. Its influence on the Process of Imagination is more subtle, consisting mainly in the addition of value judgements and metaphorical interpretations to the fictional world. And it is often the main engine behind the Process of Creativity, by means of creatively exploring the possible ways the Individual Fictional World could be changed, and judging which of these ways are cool enough to be proposed.

The Shared Background also influences all the processes. The Process of Selection is arguably the most obvious, since it is here that the rules of the game come into play most often; furthermore, plausibility constraints, genre constraints, and the structures of credibility play a large part in the acception or rejection of proposed action. However, the Shared Background also features in the Process of Imagination (adding context and often value judgements), and in the Process of Creativity (which is heavily contrained by genre expectations, game rules, and the structures of credibility - you do not propose actions you know you cannot credibly make, and may not even think of them).

Now I'm going to say something that came as a revelation to me, but is probably old news to most Forgites. The game text, that which the game designer writes, influences all these processes significantly. It is of course obvious in the case of the Process of Selection: the vast majority of rules are rules for selecting certain statements. (Who gets to say what? How can we let the dice decide what actions will be added to the SGS? What background material can always be drawn upon? And so forth.) But even if we concentrate only on these structures of credibility that are defined in a game text, the influence on all the other processes is vast.

Suppose that the verbal statement "Judd the barbarian screams in rage and attacks the 100 orcs!" has just been added to the SGS. What are we to imagine? Certainly, that Judd the barbarian does these things - the rules having done their work, what we have to imagine is no longer influenced by them. Right? Wrong. Say that the rules are D&D, and say that all players can easily predict what actions the rules will allow to be added to the SGS next. Not "Judd is victorious!", but "Judd get 134 damage and dies messily". We know what is going to happen, because of the complex structures of credibility that D&D creates. (If any said that Judd were victorious, his statement would not be accepted.) Therefore, we know that Judd's action is foolish. A value judgement has been generated by the game text.

The influence of the game text on the Process of Creativity is probably even clearer. The very player-GM distinction has a huge phenomenological influence on the creative processes of the players; they tend to imagine only possible courses of action for their own characters, for instance. Or think of My Life with Master, where the fact that you can only choose between three kinds of scene has a major influence on the direction in which we apply our creativity.

This is where things start getting interesting. What is the relative influence of different game texts on different processes? What are the ways in which we can influence these processes? Can one design games that influence only one or two of them? if so, can this be done for all combinations? (Could you make a successful RPG that only influences the Process of Imagination, leaving all else to Social Contract and stuff?) And so on, and so forth.

I'd like to hear your thoughts on this model.


Hi Victor,

I think this is really good.

I've been thinking for a while that there's probably a publishable article in a serious philosophy-friendly aesthetics journal on this kind of thing. Specifically, I think RPGs are important for Walton's aesthetic theory for the following reason: he's got children's make-believe on one side, highly participatory and creative, and then fine art on the other side, where he does a convincing job showing how some of the main mechanisms of make-believe come into play, but it's for the most part only participatory in a very attenuated sense. I believe role-playing games are an important 'borderline' case that adds plausibility to Walton's theory. This is a 'win' both for Walton's view (who's probably the second most prominent living American philosopher of art, after Arthur Danto) and for RPGs being taken seriously as a significant art form.

Your picture is similar to what I've been working with though better explicitly fleshed out. Also, the insight that the non-rules game-text contributes fundamentally to the game's SB and 'therefore' the 'SIS' is actually really important. After reading "System Does Matter" I, and I suspect a lot of other people, came for a time to regard non-rules RPG text as so much hot air. (Ron never made this mistake of course, which is clear from the GNS articles.) There is an emphasis on "show me how the rules facilitate it" around here (what Eero calls "formalistic game design") but your anatomy shows clearly why non-rules make a central contribution as well.

Keep up the good work! Let me know if there's anything I can do to help along the way.