(This is a based on an earlier draft entry. I thought it'd be better to repost with edits than to edit the original.)

This goal of this post isn't to define RPGs. Rather it's to provide a categorical survey of defining characteristics, as suggested by various observers.

First note that all games except a very few have elements of freeform, and even those that don't still have a fiction; otherwise, they are merely (virtual) kinetic art. (By the first I mean, very few games tell you precisely what to do at every step. Chutes and Ladders does. Chess and Monopoly do not. By the second I mean that the game proposes a set of meanings--often the concept of "winning and losing"--that have no real impact outside the game.)

So what is an RPG? What distinguishes it from other games? Markus Montola has proposed some criteria for roleplaying. An earlier discussion on rec.games.frp.advocacy also comes to mind.

Lea Crowe wrote:

Specifically, I think, a wargame does not concern itself with "literary" issues: character, plot, mood, etc. In a wargame, the action is all. [1]

This may be a side point, but in the majority of (modern) wargames, action is delimited by the rules, rather than merely being guided by them. For example, you can only use the tactics of spying, seduction and assassination if there are rules for them -- you can't just come up with a "spy" unit, any more than you can arrange a Mafia hit on someone in Monopoly. [2]

Lea's first paragraph refers what I'll dub the thematic or aesthetic criterion. It's been a problem for RPGs for a long time. Arguably this criterion underlies GNS (in the sense that aesthetic goals are what GNS is about instead of formalism and procedures). It's also related to some of Markus Montola's criteria.

The second paragraph is the freeform procedure criterion. I think this is a weak criterion for RPGs but it's advanced frequently. Essentially it's the criterion which says that the vision of the world overrides formal rules, or rather the vision of the world and how it can be acted on by the players cannot be encapsulated in formal rules.

In response to Lea, I added:

--Wargames generally have unambiguously defined "victory conditions" as part of the rules. RPG's generally don't. [3]

--Wargames have well-defined conditions for when the game ends. RPG's generally don't. [4]

The third criterion, which I proposed, is the motivational criterion. The claim here is that RPGs do not provide clearcut purpose, within the game, to guide player "moves". Consider: tennis, within the game, is motivated by scoring points and winning the game/set/match. (The goal of winning the match might conceivably override the goal of winning the game. As e.g. if you make shots which your opponent can score on provided he or she exhausts herself. But there is a unitary goal guiding your strategy, tactics, and technique/skill.) Outside the game, tennis may be guided by things like trying to impress someone on the sidelines or not making your boss look bad. But the goal is in the fiction of the game and the metagame goals are achieved via the game fiction.

By this criterion, an RPG explicitly does not have a formal goal in the fiction. I think this is an essential criterion, but it is sometimes excluded (generally only provided freeform criterion is satisfied; otherwise you have a closed system that becomes boardgame-like).

The fourth criterion, above, is the endgame criterion, but I think it's been basically disproved. At most it's a special case of the third criterion. Nevertheless it's still expected by many that RPGs will either go on indefinitely, or end only when some non-formal condition is reached, such as general agreement that all the "story arcs" have been played out.

[The rest of this post remains highly sketchy. Sorry.]

Now compare: pictionary, charades, the imagine-a-journey game, etc.

added: note that I don't use the word role above. So let's read what Jonas Dagar has to say about "not an RPG" and Wittgenstein. I find it very exceedingly useful to imagine that "these games" are not being called "RPGs" or even "storytelling games". These terms imply motivation and may (1) constrict play and (2) give a designer an excuse not to really explain their game. Charades doesn't have that problem. Nor does Werewolf. So what are "these games"?

Also look at this rpg.net thread about Capes.

This is prompted by a current thread at theRPGsite, where I took Levi to task for bringing up "Lumpley Principle"-esque definitions.

Anyway, I'm going to make my own set of definitions, not as a bid to impose them on the rest of the world, nor as an attempt to "interpret" the essential meanings of terms as used in Forge/Story-Games circles. These definitions may be influenced by earlier ones but they're completely de novo. Also, they're just a stepping stone to talking about the underlying concepts as they relate to design and play, in less jargonistic fashion.

Here goes.

System: the formal rules governing the distribution of authority in an RPG, and the transformation of participant declarations into game-world "facts". Systems are concerned with explicit rights and procedures.

Paradigm or ethos: a collection of common or mutually-complementary understandings regarding the responsibilities of the participants and the purpose of play.

Even these definitions aren't so important in their particulars as they allow us to talk about these things separately in design and play. In design, they're fully distinguishable. A rule that says, "The GM may not declare a conflict without the agreement of the players" is formally the same as "the players may veto any conflict proposed by the GM". Both are part of the system. A "rule" that says "The GM should avoid killing player-characters unless they do something stupid" isn't a part of the system, because it doesn't formally address rights or procedures: it doesn't alter the fact that, presumably, the GM has ultimate say over life & death, or at least the right to over-rule the results of other procedures in the system. But it is an attempt to impart or explain a paradigm or ethos.

As a bit of an aside, an explicit paradigm or ethos may or may not be necessary. As I've argued in the past, many games do have a paradigm that guides play even though we tend not to be aware of it. Namely: winning and losing, concepts that are seemingly meaningless outside the "game-space", but which we allow ourselves to care about. The only exceptions to this are activities such as gambling and professional sports, which do have extrinsic outputs that clearly intrude on "real life". But most of the games we play are not of this nature.

Still other games operate on sub-cultural paradigms that barely need explaining to the initiated--and, in any case, can't be fully explained any more than other cultural activities, whose "purposes" and "language" are diverse, and constantly being transformed through use. For example, "going to a club to see a show" has so many possible functions, each understood in varying degrees by different subsets of the club attendees, that one ought to resort to a meta-paradigm of sociality, the idea of a "scene", if one wants to capture the "aboutness" of the activity. (I've never played a LARP, but I'm pretty sure this idea will ring a bell to those who have.) I think it's undeniable that tabletop can have the same quality. In fact most interactions between humans have this quality, but RPGs are one of those activities that can thrive on it. Furthermore there's a wide range between "using an RPG as a general excuse to hang out with friends" and "using an RPG as the focus for a particular mode of socializing". Even if one did seek a method to group the varies "modes" into categories, that would not in itself allow us to directly impart a specific mode.

Finally, some people wish to assert that certain paradigms are "natural" and don't need to be taught. Personally I think this is more likely to be true of "playing pretend" than various varieties of "telling a story", but that's neither here or there: I'm just including this possibility for the sake of completeness. If you can believe that dogs instinctively communicate with barks, growls, and whines--even if they've been separated from "dog culture" since weaning--then maybe it's possible that significant portions of human culture, or its "substrate", are innate and do not need to be taught.

Let's return from the digression. As I said, the distinction between system and paradigm as I define them is absolutely clear when it comes to the designer's job and the rules text itself. A system may be incomplete--for example, it may describe how to resolve combat, without instructing you how to tell if combat occurs: can anyone declare that it's started, or only the GM, or is there some set of objective conditions which automatically triggers combat? But that doesn't stop it from being a system. Implicitly the holes will have to be filled by a paradigm, such as "the group decides collectively based on common sense".

However, once we move to actual play, the system may or may not survive, but the paradigm goes through a complete transformation. It is no longer text, but action, and the difficulties I alluded to with regard to transmission of paradigms now applies much more widely (to virtually all RPGs, if not to all games), regardless of whether the designer made an effort in the "rules text" to impart a paradigm.

The importance of this observation can be seen by briefly returning to the concept of "System" that I've previously dubbed "LP maximalism". Under this concept, it's commonly been noted (usually as an epiphany) that "systemless" or "freeform" RPGs have infinitely complex "Systems" (LP sense) rather than simple ones. But the nature of paradigms in actual play reveals that this is a completely banal assertion: all RPGs work by means of, through, and indeed upon the paradigm, the web of social interactions and understandings, that guide play. A "systemless" game is only "complex" if it requires a drastic shift on the part of the observer: otherwise it's easy as pie.

Conversely, no matter how much or how little system (my sense) a game has, there are very few ways to avoid the complexity of social interaction. One is to sew up as much as possible under formal procedure, or to fall back on very well-worn paradigms like "win/lose". Either way, you impinge on the quality that distinguishes an RPG from a board game. (The effect varies from group to group: if you strongly buy into the notion that "you aren't really playing the game if you never roll the dice", to the point that you're always trying to hammer on the mechanics, then you're more likely to fall into this trap than if you take a light system as an invitation to apply it only when necessary, on top of your largely-freeform style of play. Viz.) Finally, you can pretend the complexity isn't there, either by appealing to naturalism (see "Brain Damage") or by culture-formation and identification.

See also Jim Henley's recent post about the different perspectives on rules, with a dash of polemic from Malcolm Sheppard (eyebeams) in the comments.

Ah, almost forgot: the next step should be to take all this and translate it back into English.

Nothing particularly profound here, just a snappy phrase for a common occurrence

Inspired by [Semi-Rant] Literalisation in RPGs at RPG.net.

It's the old situation where players are given a tool to exploit, and then the game basically tells them not to exploit it. Maybe the original one was the default assumption that all creatures had an XP value, so if you just needed a couple hundred XP to go up a level, you could go out and hunt your neighbors' dogs to make up the difference. "Why shouldn't I?" asks the player. "It's not in the spirit of the game" handwaves the GM (or the rulebook).

The old Runequest weapon-shuffle was similar. Fight with weapon A until you hit (earning an experience check) then switch to weapon B.

All I have to say about this here is: I dislike the solution that entails stigmatizing the player who refuses to go along with "the spirit of the game".

I also understand the concept of not trying to build additional rules to defend against this so-called "creeping Gamism". It's a fair concept. It's just not as satisfying as actually having the rules work. This is a bit of a problem I've had with DitV and The Roach. (A bit unfair to point to the latter, though, as I've only played it in a convention environment, without having read the rules cover to cover. And it was fun in spite of the rules difficulties.)

If I try to think of a rule that really does work, for a fairly difficult problem, the one that keeps coming to mind is Runequest's rules on sacrificing POW. Not that I can remember them very well, but particularly in RQ2, I just remember being struck at their elegance both as a game mechanic and for capturing the essence of (a conception of) bronze-age religion. Another rule I recall liking is the Dragonquest rule on ritual, which made it possible to cast powerful spells provided you spent enough time on them.

(This entry is a mess. It's an appeal for help locating a couple posts that aren't in Google Groups, along with some of the reason I'm looking for them.)

I know this is a shot in the dark, but does anyone out there have the original of the rec.games.frp.advocacy post by Mike Dalton which got quoted here?

It was part of a wide-ranging discussion of the use of randomizers in RPG design, titled "Philosophies of roleplaying and design". I keep returning to that thread--you can find many of today's debates prefigured and reconfigured in various ways. What interests me at the moment is how the arguments over dice/diceless really apply to the common trio of procedural methods:

1) Randomizers
2) Social mediation
3) Formal, deterministic rules

I'm not using the popular terms because, after reading Everway, I find them harder to keep track of than the above. Also, what I have above isn't exactly the same. And by "procedural method" I'm saying something a little broader than what people usually talk about. Here I mean: how you figure out what happens next, on anything from in-game events to who gets to talk next. Again, that idea is similar to various things that have been given a name, but here I'm using it in a slightly different manner, concentrating on the game text. (I.e., if the text is silent on why and how you get from the town to the dungeon, then the only way to use the text is to have someone make up the answers. It doesn't matter if it's one person using social authority, or the group operating democratically--either way, it's social mediation. Just as it is if the book tells you to make stuff up.)

In the cited thread, much of the controversy is over the claim that diced resolution caters to an adolescent need to deal with issues of power and authority. However, much of the discussion failed to recognize that diced vs. diceless (as people usually think of them) was only a subset of a larger set of dichotomies including "rigidity" and "dramatic effect". (An earlier post where I went into this in greater detail was also lost and only appears as quoted in a response.)

So for example, in addition to arguing over the significance of the fact that an RPG tells you to answer the "what happens" question by appeal to an external randomizer (instead of by answering the question "what do you want to have happen?"), we can also argue over the significance of using a formal, predictable process that temporarily centralizes authority, instead of operating through pure consensus.

Joshua BishopRoby has a good article on the difference between focused design and "toolkit" games. Look here. I'm writing a comment right now.

Thomas Robertson has an interesting piece on his blog where he takes up the System Matters angle once again to argue that "what your game is about" can be directly inferred from what's in the rules. He makes an analogy with tennis, which lacks any rules saying how high to throw the ball when you're serving; therefore, tennis isn't about how high you throw the ball.

I'm pretty sure I disagree (even though I've always believed that systems matter), but I don't want to get into a full-blown counter-reply. In outline, my argument includes the following points:

1) RPGs aren't tennis; we use the same general word for both activities ("game") but in many ways they are merely analogous. A tennis example may illustrate an RPG point but it isn't a proof.
1a) Nothing is essentially about any other thing. E.g., tennis is about hitting a ball over a net. It's also about exercise. It's also about mating. It's also about money.
2) RPG rules generally contain freeform procedures and the textual guidelines often focus on the freeform aspects. The texts do not prejudice the relative importance of freeform vs. mechanical procedures.
3) RPG texts aren't roleplaying. Actual play gives the lie to the notion that the texts, and particularly the mechanical procedures embedded in the texts, tell us what a game is about.
4) In conclusion, rules texts matter, but they're not all that matters, and in fact their importance can expand or contract depending on context and application.

I'd like to look at the tennis example a bit more closely. Although I don't know the exact rules of tournament-level tennis, I have a basic grasp of the game. Now, as do all games to some degree, tennis requires a psycho-social context, a willingness to obey the rules and ascribe importance to them. If we play tennis and I refuse to hand the ball over (we're poor, we only have one ball) after hitting it into the net, does the game go on? What if on my serve, I stand there for an hour while I wait for you to let your guard down? What if we play for an hour and then I declare that you've forfeited because I called a fault under my breath on your first serve, yet you played on without re-serving?

Obviously these examples all seem ridiculous, like something out of a Monty Python sketch. We know how to play tennis, and anyone who does those things simply isn't playing the game. Well, if that's the case, then for all intents and purposes, we could posit a form of "strict tennis" to force jerks like me to play the game properly. Just put a bunch of killer robots around the court, and if I do something idiotic like not handing the ball over, they'll come and take the ball from me. A somewhat more real-life example would be to stop playing tennis and play a video game instead: all the rules and procedures of the game are physically embedded in the software and hardware; as long as the players agree not to jump "out of context" by, say, reaching over and tickling their opponent, the "point" of the game will reliably be manifested in actual play.

Except for one thing, and this applies to both "strict tennis" and any video game: it is not sufficient to go through the procedures. The players must also buy into the fiction of the game: i.e., they must understand what "winning" and "losing" are. "Winning" is what you want to do, what you're supposed to try to do; "Losing" is what you want to avoid doing. For some games, such as Candyland and Chutes and Ladders, "trying" is irrelevant to the play of the game, but "winning and losing" are still key to enjoying the game. Without that fiction, they're just rather boring stochastic processes--watching Conway's Game of Life would probably be a more intellectually and aesthetically fulfilling activity. (A similar comparison could be made between playing Pachinko and watching a kinetic sculpture.) In most games, though, the fiction of the game not only provides ultimate value but it also motivates the activity within the game.

At this point, the problem with determining what an RPG is "about" based purely on the rules structures should be manifest. Some RPGs, such as My Life with Master tell you explicitly what the outcomes of their procedures "mean" (to some extent through conflict resolution, and very clearly through Endgame). Many or most rules texts do not. Their mechanical procedures circumscribe certain activities (like jumping over a fence) and even in some cases provide small-m meaning (like the development of traits in Dogs in the Vineyard), but the ultimate Meaning of the game is not embedded in the mechanical structure. Instead, it's found in some interaction between the rules, color text, guidelines, and the social context of the game.

In short, the trope about many RPGs being like life and the rules being the physics of the world is true. The mechanical procedures of the game don't tell you why you're playing the game or how to play it; they only tell you, in limited ways, how to get what you want.

So, system (parsed here as "rules text") doesn't matter? No--see point 4 above, and compare the varying degrees of success of historical efforts at social and political engineering.

Jim Henley has a habit of saying stuff that blows my mind. Over on John Kim's LJ, in a post about the controversies over diceless, GM-adjudicated play, several of us got into a side discussion about the fact that all functional gaming is built on trust, and how strange it is to see rules, in a way, as an alternative to trust. This reminded me of a post on the Forge which opined that Polaris is neat because you can use to the rules on conflict resolution to resolve conflicts over the rules themselves. Of course that only displaces the problem by one level: unless you're willing to depend, ultimately, on trust, you risk getting caught up in an infinite regress.

Jim responded, "There's one other wrinkle: you have to enjoy using the rules to resolve rules disputes."

Deep, man.

See, the same thing applies to resolving rules disputes via discussion. I see a connection to Chris Lehrich's ideas about bricolage ("tinkering") as a mode of discourse within games, and the way that traditional gaming has tended to treat its own rules systems as objects for tinkering. (The most developed version of these ideas which I've seen is at Chris's "personal" LJ, here with followup here.

IOW, in much of traditional/classic gaming, both the rules and the in-game action are approached through a tinkering mindset which the participants enjoy, or have trained themselves to enjoy. On the other hand if you enjoy the mode of discourse offered by Polaris, it becomes natural to apply it recursively to the rules themselves.

What I think comes out of this is that the actual practice of playing a game (not the rules text per se, but the procedures and behaviors engaged in by the participants) have the effect of checking each other and building trust--assuming the game is successful. There may be a connection here to signaling (signalling) games, in that certain behaviors can be taken as proxies for the statement of one's honest intentions. (I wouldn't push that connection, though, without a better understanding of signaling games and their applications.)

And taking this a step further into speculation-land, we might look at some forms of dysfunction as a breakdown in signaling: we're playing a game that, to be enjoyed properly, requires oblique expression of our interests. But if either of us is inept at sending/receiving/acting on social signals, what happens? (Cf: a bad date.)

Victor, whose posts at the Forge I've enjoyed (and referenced in this LJ), has started a blog: The Gaming Philosopher.

I think the discussion has died down, and that also includes my enthusiasm for writing at length about the LP and the SIS. I was originally going to write something about how the depiction of gaming under the LP is intellectually descended from Enlightenment views on individual autonomy and natural law, but I think I'll just throw out that reference and anyone who wants to can run with it. (I do note that a couple of the comments over in Malcolm's LJ picked up on a similar theme of libertarianism in gamer-geek culture.)

The upshot of the spillover in lordsmerf's and adamdray's LJ's is that some people view the LP expansively in a way that turns System into a combination of the Social Contract and Exploration levels of the Big Model. Social context and implicit influences are part of System. I call this "LP-maximalism." Some people also view SIS as the intersection of all the beliefs held by the players about the game-world, regardless of whether those beliefs have been explicitly communicated. In other words, the term "shared" in SIS is construed as "common".

Against these views, there's the idea that the System only consists of the procedures (Techniques) evidenced in actual game play actions (Ephemera). And the formal definition of SIS is only concerned with elements of the game-world which have been "shared" in the sense of "communicated" or "transmitted" within the context of the game.

Personally, I hold the latter views, and based on the conversation in the Forge thread Shared Imagined Space, Shared Text, as well as the Glossary definitions of System, Ephemera, Techniques, and Shared Imagined Space, I think the weight of Forge authority is on my side (which came as a bit of a surprise).

This isn't to say that social context, psychology of the players, etc., aren't important areas of analysis, nor that the idea of creating a "common" imaginary world among the players is delusional. Just that it's more profitable to separate these social and aesthetic concepts from formal issues of how games are played.

An excellent example can be found in Markus Montola & Mika Loponen's article "A Semiotic View on Diegesis Construction", which can be found in PDF form in Beyond Role and Play. They build on Montola's earlier argument against the objective existence of a single diegesis, or shared ("common") imaginary space among players, in favor of multiple subjective diegeses. In "A Semiotic View on Diegesis Construction", he demonstrates how the subjective diegeses of the various players interact semiotically, often with the goal of making them equifinal, that is, "similar enough to cause indistinguishable consequences". (When I write this, I'm reminded of some of the language in La Ludisto's Interactive Model about "reconcile-and-develop".)

I would also point to works on ritual in RPGs, such as Chris Lehrich's article and posts at the Forge, as well as this thread. These may or may not touch on System, but they're really about the psycho-social and aesthetic elements of roleplaying, including how they interact with System.

eyebeams has a lively discussion going on over in his journal about the Lumpley Principle ("System (including but not limited to 'the rules') is defined as the means by which the group agrees to imagined events during play").

I'll just say this. I think that the reason the Lumpley Principle was originally embraced at the Forge is that it's the perfect rejoinder in a standard argument over whether "system matters".

I remember the old flamewars on rec.games.frp (before .advocacy) where someone (often me) would rake D&D over the coals only to be told that it was a perfectly good system because none of the rules were set in stone: they were just guidelines which could be modified or ignored in those situations where they'd produce nonsensical or dissatisfying results. And complaints about D&D fostering munchkinism or encouraging repetitive dungeon crawls would be answered by someone saying there's nothing in the game that forces you to make monster-hunting the central focus--with the right GM and players, you could play any kind of campaign.

In my view the LP answers the "guidelines" response by saying that when you introduce house rules and conventions to "fix" a game system, you're really creating a new system. "AD&D1e doesn't suck because you can always say that a crossbow bolt at short range will force a save vs. death" isn't an argument for AD&D; it's just shifting the terms of argument. For the "focus" comment, the LP answers that, at best, the lack of a clear focus means that a game is incomplete as a system. At worst, trying to focus on, say, romance and social climbing will require you to struggle against the tide of a game system whose text and procedures are oriented around fighting monsters.

Where the LP starts to get into trouble, though, is "including but not limited to 'the rules.'" I'll be following the comments over on eyebeams' journal, and if I can organize my thoughts, I'll try another post. Comments welcome here, too.

This Forge thread might be worth reading: Beyond Credibility (note the actual title is misspelled, so you might want to bookmark it for future reference).

[Edited to add] The LP also gets into trouble by talking about "agreeing to the imagined events during play". This begs the question of whether "agreement" (and "negotiation", a term often used in conjunction with the LP) is really an accurate way to characterize the way imagined events happen in roleplaying. I hope to say more on this, too.