clehrich (clehrich) wrote,

  • Music:

Design and Theory

For once, I will post some remarks about design, in response to a post on Ben's LJ and the initiating Forge thread and a followup thread, which is simply the order in which I read them. In essence, there has been some discussion --- and I gather there has also been reaction, some of it violent, elsewhere on the web --- of half-baked games being published too soon, the Forge’s possible and actual participation in this situation and the potential for improvement, what the Forge is and should really be for, and any given designer’s own personal emphases with respect to play, design, publishing, the community, and so on.

If I wanted to comment on these threads directly, with reference to their original contexts and debates and personalities and whatnot, I’d do so at the Forge, breaking my self-imposed omerta (which is mostly a matter of lack of interest combined with lack of time). This is just my own weird take on the issues from what is perhaps a peculiar perspective.

What I will propose here is an alternative theoretical direction that may perhaps be of some service with reference to these difficulties. Please be aware that this post is VERY LONG.

At base, I think there are two quite different issues here. One is purely a matter of the practical realities of community interaction, and quite frankly I think there isn’t a lot to be done here in the short run. But the other question really comes down to the intersection of critical analysis and game design, and here it is quite possible to do something --- but I think the way it’s been approached has not been effective and isn’t likely to get more so by itself.

As you may know, I have no ambitions when it comes to design or publishing of RPGs. Whatever I have designed, like Shadows in the Fog or Lévi-Strauss in Amazonia, I have designed because I was playing with something for my own reasons. If someone wants to pay me a dollar for Shadows, thank you. If someone wants to buy Amazon books through the Shadows website so I get a kickback, thank you. But I really don’t care one way or another. Any such ambitions for me are purely professional and have nothing to do with RPGs.

Furthermore, I have fairly often insisted that theory is not necessarily tied to practice, in design or otherwise, and a lot of folks have given me flack about that. Sometimes that’s because the person doesn’t understand why anyone would want to theorize other than for practical results; sometimes it’s because the Forge is a design site, which I reluctantly have accepted --- one of several reasons I no longer post there; sometimes it’s because of other reasons, such as a denial that theory in any form is valuable for design, which misses the point. In any event, my interest in this set of discussions is largely exterior and theoretical, but I do think it gets at the role (or potential role) of theory in design, something I have usually downplayed.

Beyond this, I am increasingly of the opinion that there has been a mismatch here between what I have been talking about as “theory” and what most of my various interlocutors have meant, and that it is simply not a question of one side being right. If I’ve said otherwise in the past, I’ve been wrong about that.

At base, I have generally taken for granted that the primary reason to study and analyze RPGs is for much the same reasons one usually studies and analyzes ritual, myth, and the like, which is to understand them but not especially to construct them. I continue to believe that there is no necessary reason that these things should be linked, but I have tended to argue as though it were invalid to link them, which is not the case. Mostly this has been a question of rhetorical over-statement: many people have argued again and again that theory without practical result is not valid, which is equally wrong and wrongheaded. But at a distance, I can see that simply formulating the opposing pole does nothing but entrench prejudices on all sides, including my own.

Let me start with the problem actually proposed in these threads.

Social and Structural Problems

Ben and others have suggested that there are social issues complicating publication and criticism. In particular, a given designer’s track record has a significant impact on patterns of purchasing and support, the latter ranging from playtesting to moral support to whatever. Furthermore, the broad social pattern of being supportive without criticism --- “Man, that sounds awesome, so although I haven’t seen the game, much less played it, I think it’s great!” --- interferes.

There seems to be some disagreement about the social impact of a track record, but it strikes me that there is an ambiguity here: are we talking about the impact within the Forge, or are we talking about buyers at GenCon or elsewhere? From where I sit, at a distance from all this, I think there can be no question that track record does play a role; I think it would be bizarre if it did not. But at the same time from my couple of years of heavy posting and reading on the Forge, it did not strike me that games from well-established designers were privileged to the exclusion of others. Privileged, yes: no question but that a new design from Ron Edwards would get immediate and enormous attention, at every level, and a lot of “wow, that’s so cool” stuff. But on the other hand new people also received support and assistance in considerable measure; Mike Holmes, for example, was always very ready to read any game thoroughly and discuss its various points at great length, and he was not unique. All this seems to me more or less as it should be: why shouldn’t a designer who has produced excellent games get more time and focus? That way all the readers know they’re not wasting their time, or aren’t likely to be. There is a danger of a certain “wow, it’s Ron, so it must be genius” thing going on, but I didn’t see that especially strongly.

As to ordinary buyers at GenCon or something, the question appears to be whether selling them a game that isn’t really ready is picking their pockets. I have two issues about this. First, if someone is dumb enough to buy something sight unseen with no way of knowing whether it’s any good, I don’t have a lot of sympathy for him. Second, I cannot see how this practice hurts anyone in the long run except the designer himself: surely the result is that the relatively intelligent buyers do not in future buy from this designer, and the dummies learn nothing and buy whatever strikes their momentary fancy. The only real result in the short run is that there is a lot of garbage out there, which is bound to be true in any publication arena. If you want to sell and continue to sell, you need decent designs.

Now the question arises: what should the community --- specifically the Forge, which is all I can speak to and all Ben asked about --- do about it?

I think the first question to answer is why the Forge should do anything about it:

1) Because the Forge has a responsibility to the gaming community at large to improve the general standards of gaming and thus of published work

2) Because actually good games are in some danger of being obscured behind a mass of trash

As to #1, I can see that some may feel this way. I think Ron does, since I think it’s one of the central reasons he set up the Forge in the first place, and my sense is that a lot of the most visible and active long-term posters there feel this way. So what can they do? Frankly, I doubt there is very much the Forge can do as a community here. This is not, on the whole, the Forge’s fault. The same “wow, that’s so cool, you’re so wonderful, I haven’t seen your game but I love it” culture we see all over --- and not, of course, just in gaming --- has its counter-pole in the denunciations of “the Forge as a bunch of elitists” and so on: in essence, not only do such people think “constructive criticism” means supporting absolutely everything, but they also respond violently to anything that is actually critical as automatically destructive. If the Forge community wishes to improve things through constructive criticism, there is a necessary limitation, in that such criticism can only be received with any positive result by people who are willing to accept criticism. The rest will simply ignore real criticism and denounce it as elitism, thought-policing, and so on.

As to #2, there are in fact some things that could be done, but they take a lot of time and effort that somebody would have to shoulder over and above other responsibilities. One cannot ask Ron to do it, for example --- he’s got enough to do.

For example, one might revive the Reviews section of the Forge, or alternatively set up a new site for reviews elsewhere. Establish this site with a track record for hard-hitting but fair and very well-written and well thought-through critical reviews based on intensive reading and playtesting. Review all comers, both published games and beta-tests or whatever, so long as they are generally available.

But of course, somebody has to write the reviews, and somebody has to do the playtesting, and these somebodies have to write effectively and well, and also have few if any really strong biases (games that look like WoD or AD&D or whatever should be reviewed on their merits, not because they do or do not fit preconceptions about “Story Now” or whatever). Then this has to go on for a couple of years at a consistently high level. Eventually, the intelligent and knowledgeable gamer will know that this is a place to look before buying a game, and that the judgments are valuable. Designers will hope to get good reviews, and will publish with this in mind; they will also, one hopes, release beta-tests in hopes of getting buzz. And so on.

Will it happen? I rather doubt it: it’s a hell of a lot of work, and any positive results will take quite a long time to manifest. I can think of other possibilities in this line, but they are even less likely because even more time-consuming and difficult.

In sum, pragmatically, I doubt very much there is anything the Forge community can do to resolve these difficulties effectively, except to say --- as everyone keeps on saying --- that designers really ought to submit their games for tough constructive criticism at the Forge.

Theory and Design

I do think, however, that the potential connection between theory and design is somewhat underutilized and under-thought, and that in the long run it might have a positive impact on these publication issues.

The most common uses of theory in RPG design, as far as I can tell, appear to come in two flavors.

Analytical: I have a game that has some problems. I align the game design to my theoretical model, and I see how it all stacks up. Now I find that there are some mismatches between the model’s predictions about effective design and what I actually have in hand. I therefore revise to accord with the model.

Implementation: I have a great game idea, and some ideas about how to write it as a game. As I go along my thinking and designing work, however, I run across something I don’t know how to implement. I look at the established models to find a range of possibilities, hoping not only to find a little toolbox of options but also some sense of what the implications will be of each choice. I choose the option that seems to accord best with what I want both in immediate effect and in larger implications.

I have no objection to these methods. It looks to me as though they have worked fairly well over the years. I think that on the design side these things were primarily what Ron had in mind with GNS and then the Big Model, and I think that a fair number of designers have found those models useful for the purpose.

Personally, I find theory much more interesting for other reasons, but with one exception I will keep quiet about this here; lord knows I have yapped about it enough elsewhere.

There is however another potential use for theory, and it is not one I have seen a great deal in design. I think, furthermore, that this potential use, if developed extensively, might well have a positive effect on published game quality, because it might have a positive impact on the reception of Forge community-based constructive criticism. This is because it should have the effect of dislodging the ossified, even paralytic condition of the theory itself.

You will note, in the two applications already mentioned, that the theory itself is a fixed object. That is, the design is not expected to have any effect on the theory --- only the reverse. But this has the necessary effect of suggesting that the theory use constitutes all that it is possible, or at least useful, to know about gaming insofar as design is concerned. Logically, however, such designs cannot provide data about the validity of the theory as a universal statement; they can only provide data suggesting that the theory works for whatever less-than-universal ranges it covers. Hypothetically, it is possible that a theoretically perfect game that stinks might undermine the theory, but there are always enough non-theorized factors in any game design that I think it very unlikely this would happen. So the theory remains impregnable, but there is no data support it broadly. This leads to a difficulty: people who do not wish to accept the validity of this theoretical model have grounds to do so, whether or not they are fully aware of this.

Now there is of course analyzing theory abstracted from design, which I like and Ron hates and lots of others are somewhere on one end or the other or in between, but let’s stick to design at the moment.

Another possibility in design is to start with a theoretical problem or question and develop a game design as a test-case for it. I have not seen this done very much, though I admit I do not read every game, nor anything like. Yet such an approach has a distinguished history in a great many other fields.

To take one example I know passably well and is very closely parallel in historical usage, we might consider the relationship between theory and compositional practice in music; here I am thinking of what is often called Western art music, also known rather problematically as “classical music.”

Traditionally, theory serves several functions in such music. First, it provides a constructive tool for implementation of ideas, telling a composer how to set about turning an idea into actual music that expresses the idea; this is parallel to the “implementation” approach in gaming. Second, it grants analytical purchase on music, either to fix a composition that isn’t working or to examine other works and acquire knowledge for future compositions, as for example in Schoenberg’s writings on composition and harmony in which he constantly refers to examples from Beethoven’s string quartets; this is not unlike the “analytical” approach in gaming, but is quite a bit broader than has commonly been employed, not least because the theoretical resources in music are a great deal more thorough and precise, allowing one to analyze a very small part of music with a high level of technical refinement, which is not currently the case with gaming theory. And third, theory in music has provided ideas of its own. That is, musical works generally begin with an idea to be expressed, and quite often this idea comes from theory, in the sense that the composer sees an incompletely-explored area of theory and develops an idea for exploring and developing that area, or again the composer’s analysis of previous works prompts a corresponding but transformed idea for his own compositions, and so on. The results of such theory-generated compositional ideas have sometimes produced what are often called “school works,” i.e. works of interest primarily to theorists and composers, lending themselves to intensive analysis and to theoretical development. Yet in other cases, the works produced have been masterpieces: Wagner’s Ring, Schoenberg’s Moses und Aron and many others, Stravinsky’s Rite of Spring, and so on. Not all of these masterpieces have been modernist, either, though to be sure it’s more common in modernism: Beethoven’s late string quartets and especially the Grosse Fuge, Bach’s Art of the Fugue, arguably Mozart’s The Magic Flute.

I suggest that a move like this can and should be made in gaming design. In the short run, there are two obvious positive developments. First, it will produce new and innovative designs. Obviously this is not the only way to get such designs, but it is one legitimate way. Second, it will necessarily produce theoretical refinement and development, undermining what I see as the current paralysis in theory. For one thing, such a move must entail enormously greater analytical precision in theory, because otherwise there is no way to break down other works and understand their full functional (and dysfunctional) structures.

At the social level, getting all the way back to the threads that started this excursus, I think that a theoretical dynamism currently lacking will work to undermine the current perception of the Forge as a place where only certain kinds of games are supported. The question is not (for me) whether this perception is justified; the fact remains that it exists. To my mind, the proper worry is that some designers with interesting ideas may come to believe this perception and thus not contribute to the Forge as a critical community for game development and design. These designers will thus turn to other resources for criticism, and it appears that most of the other readily-available resources are not critical at all. Thus someone who began with interesting ideas ends up producing shoddy work. Practically speaking, of course, the changes of perception will take quite some time to manifest, but I think one has to think in the long term if one is interested in broad social change of any sort, even in a tiny community like this.

As to how exactly to go about such developments, that’s going to take a while to explain, even preliminarily. I will say that the only two games I have actually designed did work from something of this perspective, though I was not entirely conscious of it during the process.

With Shadows in the Fog, I had on the one hand a very clear sense of setting and general feel, but I also wanted to design a game that could not readily be boxed up as G, N, or S. Furthermore I wanted a basic mechanic that veered outside the norms of resolution mechanics, and would carry its own connotations intrinsically. Thus my various experiments with Tarot mechanics. Finally, I wanted a game that would take some rather fuzzy notions of bricolage in gaming and make them concrete throughout.

With Lévi-Strauss in Amazonia, the initial prompt was simply Eero Tuovinen suggesting that I write the game based on My Life With Master. As the design began to develop, however, I got interested in whether it might be possible to make setting construction an intrinsic part of the play of the game without making the elements of setting meaningful in a fixed way. Specifically, the tribe and its pattern of life would slowly evolve in a complex, organic fashion, but at the end of the day none of the players --- including the GM --- would entirely understand this tribe; on the contrary, everyone would be faced with an increasingly thoroughly-viewed tribal life and a growing sense that a great deal of what was going on was comprehensible but not yet comprehended. This would produce a tearing-apart sensation in the players, the half-breed native informants, in that they would have to choose between a lifestyle they could not comprehend but to which they were bound (the tribe), or a lifestyle that rejected them and their tribal roots but which was at least comprehensible (the whites).

In any event, I am of the opinion that this sort of design development would have a very positive impact in the long run, and might also produce some very peculiar and interesting games. Furthermore, it would help make theory change and develop in useful ways.

But how does one go about it?

Analytical Tools

The first point, I think, is that the analytical tools currently available simply are not all that good; properly, they are not especially precise, and they have wound up into them an awful lot of assumptions. That is not entirely avoidable from the outset, but as things stand there is also no mechanism for self-correction.

The most thorough (which is not necessarily to say best) model I know of currently is the “Big Model” (hereafter BM), which as you know has all sorts of little fiddly bits and various levels and whatnot. Now let’s suppose we apply it for purely analytical purposes to something like AD&D or Rolemaster or Champions. If you think about it for a minute, I think you will see that one can only take such analysis so far. For example, there is within this model very little means for differentiating among different editions of Champions, despite some enormous differences. Again, how closely can one evaluate differences in character design mechanisms between the various intersecting class, race, and whatnot systems embedded in AD&D? So the model is at the moment rather coarse, a blunt instrument.

Now so long as we recognize it as a useful step and not a completed tool, that’s perfectly fine. This is part of why I asked recently what you all thought about GNS/Big Model and its utility past, present and future. What I notice is that many people said they did not care whether the model changed. That wasn’t my question: I asked whether you thought it would or should change. The would part was intended to get at whether the current institutional framework for this model can allow it to change significantly in a relatively short period. The should part was intended to get at whether the model constitutes a useful basis for future work, or whether the whole thing should be scrapped. I took it for granted that some useful basis is wanted; I only wanted to know whether you thought GNS/Big Model was workable as such a basis. In any event, I’m going for the moment to presume BM as a workable foundation which may well in the end have almost no similarity to where we end up ten years from now. I just don’t see the point in reinventing the wheel.

So let’s suppose we use this model as a foundation for the sort of thing I described in music theory. Since the model isn’t very effective for things like AD&D, let’s start with successful games close to home, like Sorcerer or MLWM, where presumably the model has some reasonable grip. So we sit down to do some analysis, asking....

And then it hit me. You may be interested to know that I was sitting in the RMV waiting forever to find out why they didn’t mail me my car registration, taking notes about these issues on a little pad. I was thinking along --- actually I had gotten a hell of a lot farther than this --- and I imagined sitting down to compose a close BM-based analysis of Sorcerer, and I imagined the sort of analytical language in which articles might be composed, and I thought, “What the hell would the question be?”

“Why didn’t it work?” is only useful if it doesn’t, but the whole point of choosing successful examples is that they mostly do work. So that’s not a great starting-point.

“Why did it work?” is totally useless. Why did what work? For what? And in fact, the model is not very specific about this. “It was fun.” Okay, but how? Are we sure that all --- or even most --- factors have been accounted for here? Do we have a sufficiently clear definition of “fun,” or in fact any at all?

This is where the primary breakdown of theory for design lies, at present, in RPG theory. We simply have an insufficiently clear framework in which to express what it is we want to do other than grand-scale concepts like fun and coherence.

Let’s suppose we take a model from music theory that may actually work here. What I will argue is that it immediately points to a very serious problem in thinking about what theory actually is with respect to design and play. Fortunately, this problem is not in any sense unique to RPG theory.

The funny little field of musical semiology, of which probably the biggest name is Jean-Jacques Nattiez, works to understand how music works (mostly but not at all exclusively Western art music) through examination of the various little bits and pieces as signs. Nattiez is quite famous for arguing for a partition of semiotic analysis into three levels, all intersecting but also at least heuristically discrete.

He gives a nice example, from his teacher Jean Molino. You have an old parlor game that starts sort of like Mad Libs on a stripped-down level. Everyone writes down several random nouns on little slips of paper and puts them in a hat. Then you create a statement of the form, “A is to B as X is to Y,” and for each of the four terms you draw a slip out of a hat:

    The toothbrush is to God as a wallet is to Finns.

The game is to come up with some way to make coherent sense of the statement. Fun, huh?

So Nattiez (from Molino) derives the following points here:
  1. There is no intentional sense in the statement, since nobody intentionally put it together in this fashion.
  2. There is apprehensible structure in the statement, in the sense that we can, with enough examples, derive a constant structure.
  3. It is possible to develop interpretive meaning from such statements despite the lack of intentional sense, probably because of the compatibility of the structure with normative rules of our language.
  4. With sufficient examples and analysis, it is possible not only to derive the structure from such statements but also at least a likely system of rules by which they were generated.
This leads to the tripartition, which has some horribly ugly terminology:

1) The poietic dimension, which is the process of creation by which the work was produced, and which may or may not include things like intentionality but certainly includes some definable and analyzable rules and structures arising from the history of production processes and the inherent limitations of the materials used;

2) The esthesic dimension, which is the process by which someone makes meaning out of a work, very often understanding him- or herself to be receiving meaning, but in fact necessarily creating an interpretation; and

3) The material trace, the symbolic form actually embodied physically and materially, which arises from the poietic dimension and is encountered in the esthesic. Nattiez is quite famous for insisting that analysis of the trace, which he calls “analysis of the neutral level,” is not only essential but also continually neglected in most musicological analysis. (Nattiez 1990)

Now I didn’t bring this up in order to introduce some horrible new terms for the hell of it, and in fact I don’t think we really need to live with these; even Nattiez admits they are somewhat “barbarous.” But let’s think about RPGs for a minute in this context, and see what happens.

We can clearly see the poietic dimension, which is (preliminarily) all the processes and methods that go into design, some of them highly conscious and many of them not. Clearly this process can be analyzed, and it can be helpful here to have author descriptions of the process. In many cases we have multiple editions (ashcans, betas, multiple final editions, etc.) and can trace the process; in some cases too we have playtest data that led directly to revisions, and so on. At the same time, it is extremely problematic to divorce the actual play manifestation of “the game” from the poietic, since the process leading to performance is also part of the poietic dimension, being constructive and constitutive of the actual material trace, which must include the performance itself.

We can equally clearly see the esthesic dimension, which is what happens when a group picks up the game and plays it. It is obvious enough that a group will necessarily inject its own interests and foci into the actual play, but that is expected: remember, the esthesic is not a question of receiving a message but of creating meaning in a dynamic relationship with the passive trace. Yet it is equally apparent that to some important degree the group will generally either understand itself to be “playing the game” or not doing so. More to the point, the ultimate “meaning” of the game is going to be markedly different from player to player, and in fact may be undefinable in a general sense. That is, it turns out once again that the performance exists simultaneously at three levels, just as does the material object that is the game book.

The natural tendency is to compare the interpretations and results in the former instance with the things we have found in the simple poietic dimension: the designer wants to know whether his intentions have in fact carried forth into the end-result. That is a legitimate question, certainly, and an essential one for the designer. But there is a tendency to presume that the process therefore goes idea -> design -> production -> reception -> play. Things are far more complicated than this.

What is missing here is that the neutral trace is distinct and stands sharply in between. Furthermore there is ultimately no way to demonstrate a passably full connection between the poietic and the esthesic via the neutral trace. To do so, we would need to create some sort of metric that evaluated the poietic independently of intent and compared this to the actual play result independently of understood reception. But this is not at all likely to work; certainly there is no parallel for it working in any other art form, so it would be bizarre to assume it here.

What we have without debate is analysis of the neutral level, and this can perhaps be compared systematically with one or the other side. It is possible in some cases that the two sides will align in some measure, but a theory of how gaming design works cannot presume this as normal.

Which takes us back to analytical tools.

If we analyze a game like Sorcerer at the neutral level, we have to analyze what is actually present in the book. But that is not all: in fact, as noted before, the performance of a group at some moment is also analyzable at the neutral level. If you think of music, for example, would we say that the composer’s “work” is the notated score, or the performance, or what is heard by a listener? Surely all of these. Just so, Sorcerer as the result of a poietic process is “the game.” But analyzed in terms of an esthesic process it seems clear that Ron Edwards drops out of the equation: we players have no direct access to him, and we begin our circular interaction with the game at the level of a physical trace in book form; nevertheless the game as it is encountered by us includes not only the book but also what happened in the room Saturday evening. So what of the neutral level?

On the one hand, there is the problem that game sessions are not notated thoroughly, so we cannot be confident that our data adequately reflect the lived reality. But in fact, the same is true with music: the notation may be very clear (although it isn’t necessarily so), but the esthesic process of meaning-construction is not equivalent to the notation. Just so, we might think that because the book in front of us does not fully reflect everything that Ron went through in writing the game, it is inadequate, but the same goes even for the poietic: Ron cannot tell us everything he went through, and even if he could this would not take into account the performative dimensions that went into the specific moment that was Saturday’s game. So if the neutral level seems an inadequate level of analysis, all that is revealed is that all levels are inadequate to the same degree.

On the other hand, the neutral level can at least be taken to be discrete and bounded. It simply is not necessary to know what the intentions were, although these may be useful clues. It is also not necessary to have a complete transcript of everything in the game session, although obviously the more detail the better. But neutral level analysis is quite possible with only the various books and some descriptions of play sessions, however partial and unstructured. Furthermore, it is at least possible --- if not yet practicable because we have insufficient theoretical depth --- to perform adequate neutral level analysis from the book alone. Indeed, this would be the ultimate hope for the designer and the consumer: simply by reading the text very closely, to know what was most likely to happen in the game session.  This will never be sufficient as a full analysis of the neutral level, because it omits the performed moment, but on the other hand the performance is rather weakly bound to the textual trace, and so it is legitimate for practicality’s sake to seek some grounding in the text.

So what is the ultimate conclusion here? I submit that the first thing to analyze is the text of a game, and that the questions to be asked should indeed preliminarily come from the poietic and esthesic dimensions, mainly positive in the first case (“why/how does this game structure achieve the intended result?”) and mainly negative in the second (“why/how did this game not produce fun in this moment?”). Ultimately, with sufficient knowledge, the hope is to discard this.

Thus the first procedure, analytically, is to examine games close to BM bases that are established as successful, and to ask how specifically particular stated intentions of the designer are manifest in the game in all its parts. This last is crucial: we cannot rely on the designer to know where the concepts are present, nor indeed to know how best to break apart the work.

Tool Development

What I think will happen, when we come to a very close analysis of every piece of a game like Sorcerer or MLwM, is that we will find an inability to describe certain things sufficiently. More to the point, we will be unable to do so without presuming the total accuracy and completeness of the BM. Since the BM presumes from the outset a number of relations between design and play that must remain absolutely open (but not in any sense un-analyzed) within neutral level analysis, notably things like Creative Agenda, these concepts cannot be brought to bear as solutions to any descriptive problem. In fact, one hopes rather that thorough description can lead to a more effective analytical description of such concepts.

So where we run into descriptive weaknesses, the model will need to develop new grounds of sophistication. This will unfortunately tend, I think, to produce a lot of new terminology. Fortunately, most people who generate the really ugly terms tend to be thinking about very large meta-concepts, not fine details. In point of fact, I think it is very likely that most of the necessary terms can be borrowed directly from well-known games and their internal glossaries. There is already a very large technical vocabulary extant through such games, and no reason whatever to invent new words to do old work.

The next problem, I think, is going to be that such close analysis is going to reveal some variations and differences in relations between structures within games, whereas the BM is fairly simple in a hierarchical sequence of nested boxes. This simplicity is going to have to shatter, but there is no reason to do it until a particular example comes to hand. The question is going to be whether a given nesting is indeed normal, and if so, what the implications of violation tend to be.

The next stage, let’s call it Stage 2, is going to mean moving beyond the familiar. We simply have got to be able to analyze established, successful games that are not in any sense Forge-based or -similar, and given that we have no grounds for evaluating “success” apart from undefined esthesic notions of “fun” and a lot of anecdotal evidence of no real statistical worth, there is no way to know in advance that something like V:tM is not successful. Neutral-level analysis of this game will also be necessary, and of AD&D (various forms), and so on. These analyses are going to be much, much larger, because these books are very long and have a great many material qualities that are unusual (not to say absent) in independent games as a result of relatively high production values. As with Forge-based games, also, close analysis will require testimony from actual play sessions, which must be selected with fair impartiality: we cannot, that is, choose all success-stories with Sorcerer and all “I considered getting out of gaming, it was so bad” stories for V:tM.  In addition, that testimony will have to be considered in light of a much larger range of play patterns and social contexts than is generally the case with major Forge-based games, given the much wider audience.

I suspect that by the time we’ve gotten that far, the analytical model in use is going to resemble the current BM in much the way milk resembles beef: there’s a cow behind them both, but that’s about all.

Now we will be ready for serious experimentation, because now we will be able to generate complex and clearly-formulated questions on the basis of analysis. For example, in some analysis it has been postulated that a fortune-based relationship between structure A and structure B will have the following implications. So now we set out to design a game in which such a relationship takes a central position and see what happens. But this need not be a “school work” or pure experiment: it is quite possible to accept the postulate and think, “that is exactly the sort of implication I want operative in my game, because it sounds just like this thing I adore in my favorite TV show,” and on this basis to design toward this goal. The resultant design manifests a physical trace which is subject to further neutral-level analysis, leading to refinement of the analytical theory, and so on ad infinitum.

The other thing we will have at this point is the possibility of an effective theory-based “design strategy” approach, much the same way you can learn to write sonatas or symphonies by taking classes. The classes give you theory, which you then implement in small ways, and then your implementations receive analytical criticism based on the theory, and so on. It simply isn’t a black box, and nor should RPG design be one. At the same time, it is worth recalling that the development of such a theory-founded rigorous design system is bound to lead to people denouncing the thought police again. I could give musical examples of why this is stupid and ignorant, but allow me to point to an obvious one in prose fiction: if you do not learn the technical skills of grammar and prose mechanics, all the scintillating wonderfulness in your head will not come out as successful fiction. Just so, what we are looking to generate here is a prescriptive system of theory; once it has been mastered fairly thoroughly, a given designer may of course wish to move beyond it, entailing alterations to the theory, and so on.

The End Result

Certainly, what I’ve written up here is rather Quixotic, and it is not terribly likely to come to fruition. But it’s worth doing: I think it (or something very much like) is probably the only really rational way to move forward with a fruitful relationship between theory and design. This is after all the sort of thing engineers do, as well as composers. If it’s all just a hobby, something you do as a pastime, then obviously there is no reason to put this much work into it. But I hear a lot of people saying otherwise.

I suppose the real point here is that I see a lot of discussion and not a lot of developing methods for analysis. Everything is based on a lot of pop psychology and personal experience-based rules of thumb. The BM is very much a creation from such a basis, and that is not unreasonable, but at some point forward motion is going to require getting out of rules of thumb. It is going to require not a lot of brilliant designs but rather development of ways to analyze brilliant designs. It is all very well to call for a spirit of critical analysis, but at the moment I see no clear way to go about it, and I don’t mean to call for it: I mean that I know of no way to do such analysis that would actually take into account even the surface technicalities of mechanics. Consider the fact that lots of people have worked out ways to explain how odds work with various kinds and combinations of dice in various rolling systems, but that we have no general model that incorporates these things at all smoothly or coherently.

So my suggestion is that people start by sitting down with games they admire and analyzing them, step by step, piece by piece, figuring out as best they can how the various bits and pieces actually work. Try not to make too many assumptions about what is and is not a “real” piece; look at what’s there. If you have to coin new terms to describe something, rack your brains and ask whether you have seen something similar or even identical in some well-known game --- there are few new things under the sun --- in which case call it by that game’s term (and if that’s not clear, call it “GameName Term,” as in “AD&D Character Level”). As soon as you have something you think works pretty well, post it somewhere and suggest that people not reply publicly except in the form of additional critical analyses of the same game or other games, in the former case looking to deepen (and quite likely challenge) your analysis, and in the latter to expand it in other directions. When we have a good library of stuff like this, we will also have a growing methodology and terminology that works quite well, and much of what we thought we knew will turn out to be hopelessly naive.

My suggestions for a better future.

Nattiez, Jean-Jacques. 1990. Music and Discourse: Toward a Semiology of Music. Trans. Carolyn Abbate. Princeton: Princeton University Press. [He's written many others, and Eero Tarasti and David Lidov are good too, but this book is nicely comprehensive and in English, as well as coherently organized and available.]
Tags: design, semiotics
  • Post a new comment


    default userpic
    When you submit the form an invisible reCAPTCHA check will be performed.
    You must follow the Privacy Policy and Google Terms of use.