The Many Faces of Role-Playing Games

Role-playing games have grown and evolved into a large number of forms in the last thirty years, spanning digital as well as non-digital media. They demonstrate a wide variety in the number of participants, style of play and the formal and informal systems that govern them. Despite this diversity players at least seem to think they know when something is a role-playing game. Yet there is no commonly accepted definition which both captures games generally accepted as role-playing games and distinguishes them from other, similar, games which begs the question, whether roleplaying games are united by anything more than a colloquial name. Additionally, research involving these games is hampered by lack of a widely accepted definition of what constitutes a roleplaying game, as it is then not even possible to clearly delineate the subject of such research. In this paper various example of role-playing game are examined in an attempt to identify the defining set of characteristics of these games. On that basis a definition for them is proposed which is hopefully more successful at separating roleplaying games from other, similar, game forms.

Take, as an example of both the similarities and differences, the current Dungeons and Dragons rule set and its embodiment in the Neverwinter Nights series of computer games.The two have many elements in common.They share a basic setting (the world of Neverwinter Nights being one of the published backgrounds of Dungeons and Dragons) and the mechanics of the digital game are a very accurate transfer of the non-digital rules into a digital form.They also differ in fundamental ways, the most obvious being the existence of a graphical interface for the computer game.Less obviously, Neverwinter Nights, in common with other digital role-playing games, is unbending in its application of the rules and outcomes.In a non-digital roleplaying game there is at least the opportunity for flexibility with (typically) a game master who can choose which rules to use when and whether the results of those rules are to be applied unaltered or moderated in some form.More subjectively there may be argument about whether the play of a given game involves actual "role-playing" or not.
The differences between role-playing game forms are not simply explained by the digital/non-digital divide.For example, "Pen and Paper" and "Live action role-play" are two categories often applied to non-digital role-playing games.The former usually consists of a small group sitting around a table, using verbal description for most of the game play, while the latter can consist of 1000's of players, using both verbal and physical enaction techniques of game play.These are not the only sub-categories that have been used in describing types of role-playing games, others include "freeform", "tabletop", "systemless" and "pervasive".Whether these are truly separate and distinct categories is debatable, but they all enjoy some colloquial use.
Matching this wide variety, researchers have approached role-playing games from a number of perspectives.Some have documented the history of one or more forms of role-playing games, for example Koster (2002), and Mason (MAS04).Mackay (2001) examined them from the performance point of view.Copier (2005) relates non-digital role-playing games to their place in the Dutch fantasy subculture and their relation to ritual.It is also worth noting the use she makes of role-playing games in discussing the concept of the "magic circle", an example of the study of roleplaying games being used to examine more general gaming concepts.Fine (1983) uses a participant observation approach to discuss them mainly in the context of the interactions between the players.Montola (2007) describes the application of role-playing games in pervasive gaming format.Tychsen et al. (2007) examine the changes in player enjoyment and engagement between some of the various forms of role-playing games.For Dormans (2006) they are an opportunity "to take some theoretical concepts and notions developed for computer games and use them to study roleplaying games".
Outside the academic sphere, some members of the role-playing community itself have attempted to analyse these games.Such self-examination has been extensive, for example in the internet venues of The Forge and RPGnet, but generally lacks connections to wider game theory.It is worth noting though, as Copier (2005, p.4) does that "Some researchers take part in both the academic and the player's discourse on RPG theory".Notable work originating from the role-playing community includes discussions on game play style, for example (Hetland 2004) and (Edwards 2001), and examination of the place of narrative and story-telling in role-playing games, for example (Henry 2003), (Kim 2003) and (Padol 1996), amongst other topics.
While all these studies, and others, are obviously highly varied in their approach to role-playing games, it is notable that they generally take a circumspect and/or highly inclusive approach to defining what it is they are discussing.It is both interesting and understandable that many authors dealing with role-playing games shy away from the question of defining exactly what a role-playing game is.For example, while Copier (2005) offers some discussion of the forms of in which roleplaying games exist, the activities involved and the relation between role-playing games and well known definitions of games in general, she does not tackle the question of exactly what is a roleplaying game.Instead, the section in her paper entitled "Role-Playing Games" deals with their history, the demographics of Dutch players and the history of the study of role-playing games, without touching on exactly what a role-playing is.
Many authors that do address the question posit deliberately wide definitions.They may define the act of role-playing (as opposed to a role-playing game).Typical of these is (Henriksen 2002, p.44): "[role-play is] a media, where a person, through immersion into a role and the world of this role, is given the opportunity to participate in and interact with the contents of this world." A more extreme example is that of Pettersson (2006, p.101), for whom "roleplaying is the art of experience, and making a roleplaying game means creating experiences".As noted by Stenros and Hakkarainen (2003, p.61) many existing definitions of role-playing and role-playing games "have been largely normative, not descriptive".This is not to say that such efforts are without value.The roleplaying experience is undoubtedly one in which immersion, the assumption of a role and the involvement of the player are central.The player experience is, however, not the same as the activity in which they partake.Some authors have taken a more descriptive standpoint, for example (Stenros & Hakkarainen 2003, p.56): "A role-playing game is what is created in the interaction between players or between player(s) and gamemaster(s) within a specified diegetic framework.… [A] roleplaying game requires four things, a gamemaster, a player, interaction, and a diegetic framework."Again, this is a rather broad approach to the question.Many games not normally considered role-playing games are covered by it and similar definitions.This arises from the focus of the authors, which can be seen in their statement "We have created a model that includes all activities that we recognize as role-playing".Note that they refer to role-playing, not role-playing games.While this inclusivity is commendable when it comes to understanding the general activity of role-playing, it does not help in separating role-playing games from other game types.The same paper, for example, discusses the possibilities of role-playing in Risk and Monopoly, games not generally regarded as role-playing games.
It could even be argued that, given the extreme variety of form displayed by role-playing games, touched on above, and the possibilities for roleplaying outside of role-playing games, that a general definition can either not be arrived at or would be too vague to be useful.However, the extensive use made of the term "role-playing games" by these authors, and many others, implies that it refers to something and that a potentially identifiable object, the role-playing game, exists.Otherwise the use of the term could only be considered confusing at best.As discussed below these proposed definitions have significant shortcomings.
That a certain type of game exists which can be labelled "role-playing game" is implied by the widespread use of the term.Players appear to believe they know whether or not a game is a role-playing game.If the accuracy of such identifications is accepted then a definition of a role-playing game may perhaps be arrived at by analysis of the different forms in an attempt to identify the common features, if any.It is also important to note that role-playing exists outside role-playing games -in various social and cultural arenas, in education, training, etc.The broader activity of role-playing is not the topic being discussed here, instead what is examined is the group of games collectively, and colloquially, termed role-playing games.Whether any of these games involve actual "role-playing" is another question, as, in many cases, identifying roleplaying is extremely subjective and notoriously difficult to achieve agreement about.
The topic of this paper is role-playing games, not role-playing.Significant discussions exist of the role-playing activity, both in gaming and nongaming spheres.But as role-playing does not need a gaming context in which to exist, definitions of role-playing do not provide a conclusive answer to what constitutes a role-playing game.This paper addresses the latter question be examining the nature of those games, which provide the context for the role-playing that occurs within them.

EXISTING DEFINITIONS OF ROLE-PLAYING AND ROLE-PLAYING GAMES
As noted above there have been a number of attempts at defining role-playing and/or roleplaying games.While they are, understandably, varied, they can be (roughly) divided into two broad categories -those which focus on the process and experience of role-playing and those which include descriptive elements about the game and game-play itself.These, to an extent, correspond to the normative and descriptive categories identified by Stenros and Hakkarainen (2003), although some of the definitions contain elements of both and their placement here into one of the two categories may be considered arbitrary.Such placement is not intended to be definitive, but instead a means of discussing current efforts at definitions.While space prohibits an exhaustive examination of the definitions offered in the literature, it is worth reviewing a (hopefully) representative sample.

Process and Experience Based Definitions
One of the earliest definitions of role-playing, from a time when even many role-playing games themselves did not address the question, is that of Lortz (1979), who defines a role-playing game as "any game which allows a number of players to assume the roles of imaginary characters and operate with some degree of freedom in an imaginary environment".The emphasis here on players assuming roles and the freedom with which they interact with the game world is a theme that later writers would return to in more detail.While it places the player and their experience at the centre of the role-playing activity it only conveys a sketchy impression at best of the games themselves.
This tendency can also be seen in (Padol 1996) where a role-playing game is defined as one that "Allows people to become simultaneously both the artists who create a story and the audience who watches the story unfold.This story has the potential to become a personal myth, shaped to meet the needs of its creators."This is an attractive definition, at least to those who wish their characters to experience interesting stories (which is at the very least a significant minority of roleplayers) and conveys a useful perspective on the role-playing experience.Its emphasis on story, which is clearly seen by the author as something created in the role-playing activity, but not as the whole of that activity itself, does not tell us a great deal about the context in which those stories are created.Read literally, any game with story elements where the participants have some input into the unfolding of that story, could be said to fit within this definition.It is doubtful that such a simple-minded interpretation is intended, and instead there appears to be an implicit assumption about what a role-playing game is, the definition telling us more about what happens in such a game.Its usefulness in separating role-playing games from other game types, on a structural or descriptive basis, is therefore limited.
A more general definition, without the emphasis on story, is that of Henriksen (2002), quoted above.Again, this sidesteps the question of the means by which and limitations upon the interaction that the players have with the game world occurs.This should not be seen as a particular criticism, as roleplaying, not role-playing games, are being defined.Role-playing can, and does, occur outside of roleplaying games.¹In fact, one of the earliest definitions of role-playing pre-dates role-playing games by about two decades, that of Mann (1956, p.

227):
"A role-playing situation is here defined as a situation in which an individual is explicitly asked to take a role not normally his own, or if his own in a setting not normal for the enactment of the role."As these definitions are not directly addressing role-playing games it is understandable they have limited utility in identifying such games.
One similar definition which does mention games is that of Montola (2007, p.179).Again the attempt is not to define role-playing games, but to offer a definition of the role-playing activity (italics as in original): "I see roleplaying as an interactive process of defining and re-defining an imaginary game world, done by a group of participants according to a recognised structure of power.One or more or participants are players, who portray anthropomorphic characters that delimit the players' power to define."While it is almost certainly unfair, given its stated intention, we can examine how useful this definition is in explicitly categorising role-playing games; the word "game" is, after all, included.An "interactive process of defining and re-defining an imaginary game world" could apply to any game, as any game, even the most abstract, has a game world which the participants alter through their game play.The phrase "recognised structure of power" is likely meant to refer to the game master function and the variety of forms that can take, but does not define how power within the game is structured or how it is recognised or indeed whether the power structure may or may not be egalitarian.It should also be noted that software and a player could be considered to form a group of participants, with a power structure, so this covers all digital games.This definition could then cover a range of digital games, for example first person shooters and three-dimensional platform games, as well as board games such as Talisman and Squad Leader which represent individual characters within the game.It is not likely that this is actually intended and again this definition has much more to say about the role-playing process than roleplaying games.
There are other definitions which fit broadly within the category discussed here.For example, that of Pettersson (2006) (quoted above), Pohjola (2004, p. 89): "Role-playing is immediated character immersion", Pohjola (2003, p.34): "Role-playing is immersion to an outside consciousness ("a character") and interacting with its surroundings", Mäkelä et al. (2005, p.207) "role-playing is defined International Journal of Role-Playing -Issue 1 ¹ Indeed, it is debatable whether a role-playing game has to involve role-playing at all. as any act in which an imaginary reality is concurrently created, added to and observed" and Edwards (2001), which discuss the requirements for the role-playing activity rather than the definition of a role-playing game.
It should be noted that in all these definitions it is role-playing, not role-playing games, which occupies the central position.It is understandable then that, while valuable within their chosen scope, they are less useful when it comes to identifying role-playing games as a separate category.As noted in several of them the act of role-playing may occur in a wide range of venues, including games not recognised as role-playing games and even outside a gaming context altogether.
As the focus of the current paper is to arrive at a definition that can identify role-playing games as a category within the broad spectrum of games, not on the role-playing experience, we need to consider other approaches.

Descriptive Based Definitions
There have been fewer attempts at descriptive definitions of role-playing games then those of the type discussed above.However, a number do exist.Dormans (2006) gives a definition of role-playing games by categorising them into four types, penand-paper, live-action, computer and massively multiplayer, ostensibly on the basis of "medium, means and scale".However, on examination the differences between the categories offered are actually only on medium (pen-and-paper, liveaction and computer) and scale (pen-and-paper and computer versus live-action and massively multiplayer).It would be difficult to argue against the proposition that many examples of role-playing games fall into one or another of these categories.However, the proposition that these categories are sufficient is more contentious.There is, for example, an implicit assumption about quantitative measurement and random resolution underlying the arguments presented.This can even be seen even in the article title, which begins "On the Role of the Die" and in the early statement "I will try to expose the role played by dice in these games".Role-playing games exist which do not require random quantitative resolution.In fact, some do not require quantitative elements at all.This is not a reference to the Amber diceless system or similar rulesets which involve quantitative assessment of character skills and abilities but not random resolution.A form of role-playing game which does not fit any of Dorman's categories is that known in Australian hobby role-playing conventions as "systemless".That this form, described below, exists invalidates Dorman's four categories as a complete definition of role-playing games.More fundamentally, Dorman's categorisation has limited usefulness as a complete definition, as no attempt is made to analyse the forms to discover if there is any underlying commonality which could both group them together and separate them from other game types.If there are forms of role-playing game beyond is four categories no guide is given as to how to identify them.The definition therefore relies upon its own a priori completeness.Tychsen et al. (2006) provides a detailed analysis of multi-player tabletop and digital role-playing games, comparing the two forms.The analysis describes both the process-aspects of play, as well as providing an overview of the features shared between role-playing games in general.While fairly detailed, the discussion is explicitly stated as not being complete.The reader is referred there for the complete overview of the analysis, but in summary it states that role-playing games all share the following features: • At the heart of role-playing games, there is an element of "storytelling with rules", and each game form provides unique ways of expressing this feature.
• Rules, multiple (at least two) participants and is set in a fictional world, established via the game premise: A shared understanding among the participants of the game setting, rules and similar game framework issues.
• Most of the game participants normally control a character through which they interact with the fictional world • There is usually a game master (or digital system performing a similar function) responsible for management of those elements of the game and fictional world outside direct control of the players.
Tychsen et al. ( 2006) also discuss the functions of the game master and mention that the role of the game master may not be fixed, but move amongst the participants, and varies greatly in functionality across role-playing game forms.
The problem with the list of shared features in Tychsen et al. (2006) is that it appears to cover games which are not normally considered roleplaying games.If it is allowed that some digital games are role-playing games (which the authors of the current paper do) then consider the firstperson shooter.Once role-playing games are allowed to be digital than it must also be further allowed that software may take on the role of the game master.When looking at the list above, it would appear that e.g.first-person shooters would be considered role-playing games.They have a fictional world, multiple participants (at least one player and the game master/software) and a character through whom the player interacts with the game world.First-person shooters could also be argued to contain or create stories during play, notably games that specifically aim at creating an interesting storyline, e.g.Deus Ex, System Shock II and Bioshock.
It could be argued that the player does not roleplay a character in a classical FPS-style digital game, however, it is important to note that the player does have the potential role-play the character -but there would not be an in-game effect of this role-playing (except potentially in affecting some of the choices the player makes).Many contemporary FPS-games include features for solving conflicts in different ways -e.g.violent vs. non-violent solutions to problems in Bioshock or Crysis.It could be argued that this provides a lowlevel form of role-playing potential to these games.
Yet such games, with a few exceptions, are rarely considered or termed role-playing games.Some, such as Deus Ex, are said to have role-playing elements, but that is not the same as being a roleplaying game.Another game type that could be argued to feature the same series of elements are the three-dimensional platform games, such as Jak and Daxter and Ratchet and Clank, which again are not typically considered role-playing games.
Finally, Tychsen et al. (2006) point to the importance of the role-playing element of roleplaying games, but also note that contrary to the name, the act of role-playing is not a feature found in all the games popularly titled role-playing games.For example, digital role-playing games often feature a comparatively limited ability for the player to role-play their character.The authors do not however provide a definition of when a player can be said to be role-playing or not.
An essentially similar, if less detailed, definition is given by Morgan (2002), which in summary, states that players deal with an imaginary world, through the medium of a character, and that there is a game master who: "adjudicates rules disputes"; and: "guide[s] play much as a director would a movie".It can be seen to also be problematic in terms of identifying what and is not a role-playing game as it what it covers conflicts with the generally accepted usage of the term.
A slightly different approach is taken Mackay (2001, p. 4) who defines role-playing games as follows (italics as in original): "[A]n episodic and participatory story-creation system that includes a set of quantified rules that assist a group of players and a gamemaster in determining how their fictional characters' spontaneous interactions are resolved." It does not mention a fictional world and focuses on story-creation and interaction.It requires quantified rules, which were noted above to be unnecessary.Again, whole classes of digital games not recognised as role-playing games fit the definition.It could also be asked why the game must be "episodic".Many examples of short games, which can be completed in a single session, are known, particularly at non-digital role-playing conventions.The prominence given to "storycreation" is also debatable, given the arguments around the place of story and narrative within games.
Another, often referenced, definition of a roleplaying game is that of Stenros and Hakkarainen (2003), quoted above.In common with some of the other definitions discussed here it mentions players, game masters and interaction.However it eschews mention of a game world in favour of "diegetic framework", which includes the game world.²The concept of diegesis is an extremely useful in understanding what is happening within a role-playing setting and how players approach the act of role-playing.However, as it can apply generally to any game form it is less useful in separating out role-playing games.
Even the more detailed of the above definitions reduce to a game, set in an imaginary world, played by multiple participants, one or more of whom has a special role, commonly termed the game master, who controls aspects of the game world outside the control of the remainder of the participants, who typically control one or more characters.The presence of a privileged participant who controls aspects of the game world is hardly an identifying element of role-playing games.As well as the digital game forms mentioned above consider for example, referees in competition figure gaming, who may create the terrain upon which battles are fought (i.e, the game world), interpret rules issues and occasionally adjudicate outcomes.
This appears to meet the minimum requirements of a game master, yet such games are not considered role-playing games.Interestingly, though, it was from such games that the original table-top roleplaying game, Dungeons and Dragons, was derived.Which perhaps goes some way to proving the relationship between the role-playing game master and the figure gaming competition referee and that the mere presence of such a participant is not enough to make a game a role-playing game.
Most, and arguably all, of the definitions discussed in this section are successful in that the games commonly termed role-playing games meet their requirements.Unfortunately, despite their respective advantages, they are insufficiently precise for use in deciding which games are roleplaying ones and which are not as they also include within their scope games which are generally not considered to be role-playing games (or, in the case of Dormans (2006) do not cover games which are and give no guidelines for considering undecided cases).

ROLE-PLAYING GAME FORMS
From the above discussion it can be seen that we do not currently have a definition of a role-playing game (as opposed to the role-playing activity) that both includes the set of games commonly described as role-playing games, while at the same time separating them out from other game forms.As this has not been the intent of the work cited in the above, this should not be taken as criticism.
However, attempting such a definition is useful as it offers a different perspective on role-playing than that offered by previous authors.
A definition which specifically permits the identification of a game as a role-playing game or not, could possibly be developed based on analysis of existing known examples, in an attempt to identify any similarities.
The analysis presented here will consider the following examples of role-playing games: • Pen-and-paper/table-top

• Systemless
• Live-action role-playing This is not intended as an exhaustive list of all forms of role-playing game, nor a claim that each is significantly different to all the others.For example, under some definitions Systemless could be considered a sub-type of pen-and-paper, under others a sub-type of live-action role-playing.For present purposes that some of the above may be closely related is however immaterial, what is important is that the examples in the list, as a whole, have been selected to ensure a coverage across the breath of role-playing games to provide a firm basis for developing a workable definition.
Some of the examples in the list are considerably more widely played than others, but the intent is to arrive at a definition that covers all role-playing games, not simply the more popular ones.
Descriptions of some of the above have been given elsewhere, for example by Dormans (2006), where four of the listed forms are described.For completeness, and ease of analysis, all seven are described below, although some of the following content differs minimally from the existing literature The concept of diegesis is extremely useful in understanding what is happening within a role-playing setting and how players approach the act of role-playing.However, as it can apply generally to any game form it is less useful in separating out role-playing games.

Pen-and-Paper/
including rules, play aids and the character descriptions.The last, termed a character sheet, usually describes the character in quantitative terms, with perhaps some qualitative description of the character's personality and history, with the latter varying greatly in occurrence and extent.The character sheet gives rise to the term "pen-andpaper", although the information is often written in pencil, not pen, to allow updating as the character evolves.Players may interact with the game world in any way that their characters, as inhabitants of that world, are capable of and play can potentially roam through any part of the game world.
The pen-and-paper form, being the one from which all others has originated, is well known and has been discussed in detail elsewhere, for example by Fine (1983), Mackay (2001) and Dormans (2006).In the interests of space, the form has therefore not been given as extensive an examination as some of the other forms discussed below.A closely related sub-variant, Systemless, is discussed in the next section.This formdisplays characteristics not highlighted in many previous descriptions of small-group role-playing games.

Systemless
This game form, arising in the Australian roleplaying convention scene and elsewhere, is related to the pen-and-paper form and to psychodrama.³The number and functions of participants is typically the same as that for pen-and-paper, although the use of multiple game masters with substantial authorial control is more common.In these games characters are described in purely qualitative terms, by giving descriptions of their history and personality.There is no quantitative (or even pseudo-quantitative) definition of a character's attributes or skills.Character development is still possible, but is in terms of personality and emotion rather than the skills, attributes and levels typical of the pen-and-paper form.
In Systemless play emphasis is placed much more on enaction than description, players do not sit around a table, but move around the game space speaking as their character and portraying their characters' actions.The play of the game is the interaction between the players (including the game master) and the development of the characters and story.Actions are resolved based on the decisions of the game master, based purely on their assessment of the situation, and without reference to any quantitative character or world description or any form of random resolution mechanism.The possible range of player interaction with the game world and the range of play are the same as for the pen-and-paper form, though the means of resolving actions with the game world is markedly different, given the lack of quantitative and random elements which commonly feature in the pen-and-paper formsoften in conjunction with the same means of resolving actions in the game world as Systemless play.
This form appears to fall outside the categories of Dormans (2006) "Larp is a role-playing game, where the actual physical reality is used to construct diegeses, in addition to communication, both directly and arbitrarily." On the other hand some definitions of the border between live-action role-playing and tabletop, such as that of Lynch (2000), leave Systemless on the tabletop side of the divide.
Regardless of whether Systemless is an example of pen-and-paper, live-action or something else, its eschewing of quantitative elements while remaining a role-playing game is informative.

Live-Action Role-Playing
Live-action role-playing typically involves larger numbers of participants than the preceding forms, ranging from the dozens up to hundreds or even thousands.Emphasis in these games is placed on player enaction of the character's actions (similar to Systemless play, although live-action roleplaying can feature rules for player interaction), International Journal of Role-Playing -Issue 1 ³ "psychodrama; a method within group psychotherapy where the participants take roles in improvisational dramatizations of emotionally charged situations", Psychodrama (2007) costuming, props and setting.Real world locations, such as castles, parklands and warehouses, are used as the settings and are chosen to match the game world setting as closely as possible.As with the previous forms, participants are normally either players or game masters, with the players enacting a single character and the game masters, of whom there must be a considerable number due to the number of players, again controlling those parts of the game world beyond the players' characters.There may also be players who are assisting the game masters by carrying out preplanned actions, and so are not entirely acting at their own discretion.There are examples, such as some Scandinavian-based games, where attempts have been made to break the traditional game master-player boundary.In these games the relationship between players and game masters is fluid, changing over the course of the game through various game contexts.
Character descriptions can contain quantitative elements similar to the pen-and-paper format, but are usually based on qualitative information (e.g.personality, background) While player enaction is emphasised, formal rule systems are commonly used for determination of the outcome of many character actions, e.g. in the Minds Eye Theatre system, White Wolf (2005).The embodied nature of play, together with the emphasis on props and costume, allows players to have their characters interact with the game world in extremely varied and detailed ways.While the use of real world settings may appear to limit the areas of the imaginary game world which characters can inhabit, the game masters are free to extend the scope of play as they see fit.

Single Player Digital
The single player digital form of role-playing game, Hallford and Hallford (2001)

Massively Multi-Player Online
The most obvious difference between this and the previous category is the number of simultaneous There is potential for a much higher degree of attachment to these characters by the players than in the single player digital forms, which typically last 20 to 40 hours.

Freeform
The freeform style is reasonably well known in The United Kingdom, USA and Australia.In many ways it can be viewed as a specific form of the liveaction style, but is usually recognized among the player community as an independent category of role-playing game.Freeform is a form of live-action game with a clearer emphasis on character interaction in a more controlled environment than is possible in large scale live-action games.There is typically limited, if any, emphasis on combat.Normally the number of players involved is much larger than the table-top form, but less than is typical for live-action and also places less emphasis on setting, costume and prop.It tends to rely heavily on inter-player communication and negotiation and less on rules based action resolution.Again multiple game masters are required to handle the larger number of players and while most play is set in a single physical location, represented by the physical play space, the game masters are free to extend this into anywhere in the game world.

Pervasive
Pervasive and ubiquitous games are typically digital games which extend the game play beyond the computer screen.For example, where player movement in the real world equates to avatar movement in the game world, as in Botfighters.
Pervasive role-playing is slightly different, in that it does not necessarily include a digital component.Instead it is essentially an extension of the liveaction form.In the latter there are usually boundaries (of various strength) defining which parts of the real world are being used to represent the game world.In pervasive role-playing these boundaries are much weaker or even essentially non-existent, to the extent that anything in the real world, even people not playing the game, can take on a significance for the play of the game.As any part of the real world, or anything in it, can potentially be part of the game, it is obvious that the geographical range of, and the possible ways of interacting with, the game world are extensive.In most other ways this form resembles live-action role-playing.For more detail see Montola (2007) and Jonsson (2007).

FEATURES OF ROLE-PLAYING GAMES
As can be seen from the above there is significant variation amongst role-playing games, including the mechanisms supporting game play and the play styles that typify them.While this may make it appear unlikely that a useful overarching definition can be found there are also considerable areas of similarity.

Character
All the examples discussed share a use of playercontrolled characters.These characters are the primary (in most cases the sole) means by which the players can interact with the game world.The methods by which the characters are defined vary, in some cases being purely quantitative, in others extensively qualitative and in others a mixture of the two, but in all cases the characters are regarded as individuals, with their own unique place in the game world (some experimental Scandinavianproduced role-playing game modules have experimented with replacing the typical character with e.g.abstract concepts, such as a group of emotions).This contrasts, for example, with the use of characters in simulation based education and training exercises, where characters are more often described by their roles (teacher, medic, etc.) than by reference to their individuality.The players are able to effect, and influence, the development of the game world through actions expressed via their characters.
More than being merely character-based, characters in role-playing games are in the vast majority of cases capable of development, as noted in the definition quoted above.Again this development might be in quantitative, skill and ability, terms or in qualitative personality terms.
While the form of the development might vary widely between games it is always subject to at least some player control.There might be skill points which the player chooses to allocate, specialisations to select or decisions made about emotional changes.This separates role-playing games from other games with character development, but where that development is fixed and pre-decided by the game designers, with perhaps some limited choices by the player within a defined framework.Pre-defined development is seen, for example, in games where the character obtains a new ability when a particular point in play is reached, such as is often the case in 3D platformers, or some adventure games, or which demonstrate emotional change in the central character, but where this is again under the developer's, not the player's, control.It might be contended that, in any game that is characterbased, the player may imbue their character with a personality and develop that personality over the play of the game.While this is true, a defining feature of role-playing games is that they are capable of reacting to changes in the character(s).The game reacts to skill and ability changes.If the game focuses on the personality of a character then when that personality is changed, the game can react, in the shape of the reactions of the other players (and the rest of the game world, as expressed by the game master).Even if the player of, for example, a racing game imbues their character with a personality and then changes that personality the game will not be able to react as it is not designed with this capacity (in fact, it would not be able to react to the initial personality state either).
The above is not to argue that character development must occur for a game to be a roleplaying game, only that it be possible within the design of the game, offer some control to the player and that the game will respond, in some manner, to the changes.Character development is not a requirement on every player or character, but is a potential play feature existing within the structure of the game.It is perfectly possible, for example, to "play" World of Warcraft, by creating a character, and then merely touring the world without ever acquiring additional equipment or experience points.Similarly, a player of Monopoly could simply move their piece around and around the board without ever buying a property.But the intent of development is there, even if ignored in some particular play examples.

Game Master
While most participants in the games discussed are players controlling a single character, all of the forms also have other participants who control the game world beyond the players' characters.These participants are typically referred to as game masters.The exact duties of the game master vary, with the power relationship between players and game master varying between game to game and even at different points within the same game.Game master functions also vary, making defining them difficult, although some attempts have been made such as Stenros and Hakkarainen (2003) and Tychsen et al. (2005).Whatever their exact nature, the viewpoint of a game master is very different to that of the players.While players are primarily concerned with their particular character, game masters are primarily responsible for presenting the world to the players, elaborating story elements and adjudicating results.This is also the case where the game master is represented by a game engine in a digital role-playing game.Where there is extensive use of props the game masters are responsible for the selection and positioning of these.Even in pervasive games; game masters will often place game objects within the real world, structuring and controlling it according to the needs of the game.
The presence of a game master helps differentiate role-playing games from other forms of character based games, such as board games where a player controls a single character, for example Zombies!!, and from children's games, such as cops and robbers.The game master may be called upon to adjudicate outcome of events in he game world, and will rely upon a rule system to do so.However, that rules system does not necessarily include any quantitative representation of characters or game world or include random resolution of any kind.

Treatment of Space
Role-playing games consistently make use of a fictional game world, and this element is found in many of the definitions discussed in section 2. Yet it can hardly be said that role-playing games are unique in this.Many, if not most, games are set apart from the real world by the placement of their action in a fictional world.However, role-playing games make use of the fictional world in a manner that is consistent and distinctive, although not one which is unique to them.This can be seen both in the parts of the game world encompassed within the game and the means by which characters can interact with that game world.
Most game forms are limited, by their structure, in the amount of the game world that the players can experience.This is true of both digital and nondigital games.A military board game covers a fixed amount of territory.Asteroids is set in a small part of an asteroid field.Cluedo, in both digital and nondigital forms, is limited to a single house.Play can never proceed beyond these limits.Non-digital role-playing games are under no such fixed restrictions.They offer the promise (if rarely fulfilled) of the ability to go anywhere and do anything within the game world.The players and game master are free to investigate the entirety of their imaginary universe as they please.Even if the play of the game is currently geographically very limited within the game world (perhaps even to a single building or even room) this is a conscious choice of some or all of the participants, not inherent in the structure of the game and could potentially change at any moment.This concept is touched upon by Young (2005) in his discussion of game mastering styles, but not investigated there in great depth.
Digital role-playing games are nowhere near as free, being almost always limited to the pre-created game content.Indeed this has lead Schut (2003, p. 10) to suggest: "maybe we should use [Janet] Murray's term and call digital game narratives participatory stories".Even then digital roleplaying games tend to encompass a high proportion of the imaginary world, higher than first person shooters, perhaps equalled by adventure games and some 3D platform games, such as Ratchet and Clank, Jax and Daxter and Beyond Good and Evil.Such games closely resemble roleplaying games but lack the character development aspect.The need for pre-play preparation of the graphical representation of the accessible areas of the game world should not be considered a hard and fast limit on digital role-playing games.As technology improves the ability to present interesting, non-pregenerated, space will improve, bringing to the digital the possibilities currently only available in the non-digital.Movement in this direction can be seen in the recently released game Hellgate: London.
Not only do role-playing games allow access to relatively large sections of the game world (and in some forms potentially all of it), they also allow extensive choice in how players may explore that space.Players are generally free to choose their path through the world (at least to an extent noticeably greater than many other game forms) and even revisit areas.Again, this is not true of character based games that divide the play area into levels (such as most first-person shooters), where the player is restricted both in their path through the environment and from revisiting completed levels.Adventure games likewise tend to move players through the world a section at a time, limiting the ability to revisit areas.Roleplaying games, especially the non-digital forms, then, can be viewed as treating space in a (pseudo-) realistic manner.Characters have choice as where they visit, what order they visit areas and whether they wish to revisit areas, just as the players of such games do in their real lives.
Role-playing games are obviously an example of Murray's (2000) concept of the "tangled rhizome" mode of spatial navigation in games.However, while they allow the choice of direction she posits, they are not alone in this.It is the scope of the choice offered to the player(s) that sets such games apart.
Obviously global or galactic strategy games such as Civilization or Space Empires offer significant spatial scope (in raw quantity at least) within the game world and flight simulators typically present both expansive areas and free player navigation.However, these games are typically not character based, and the options they present for interacting with the world tend to be more limited than is found in a role-playing game, as discussed in the next section.

Interaction with the Game World
The previous section dealt with the treatment of the game world on a macro scale.Role-playing games also have a consistent approach to microlevel interactions with the game world.
While all games involve a configurative element role-playing games differ in the potential scope of the configuration available to the player from other character based forms.Players of non-digital games can have their character interact with the game world in any way that is possible within the limits of that world.Even digital games, limited by current technology, tend to offer a wider range of possibilities, usually including combat, dialogue, object interaction, etc, than is found in most games.Adventure games lack the combat option, firstperson shooters generally offer less rich dialogue and object interaction, etc.⁴While role-playing games provide more interaction opportunities, they tend to be generalist rather than specialist in how they allow players to exercise their configurative options.They may allow many different ways of interaction with the world (such as driving, shooting and talking) but do not go into any in as much detail as games dedicated to such activities.Consider a racing game, such as Formula One Championship Edition.
Interaction with the imaginary world is limited to partaking in races, although this may be covered in exceptional detail.The player cannot stop racing and start flying a plane.Non-digital (and some digital) role-playing games will allow players the opportunity for both (assuming they exist in the game world) and much more, but will typically not cover any activity in as much detail as a game dedicated to that pursuit.While some role-playing games may have extremely detailed coverage of some of these (typically combat and occasionally vehicles) this treatment does not extend to all possible world interactions -any attempt to do so would lead to a game too rules-heavy to be easily playable.In a non-digital game a player may decide to cook, paint or any other possible activity, but the resolution of these actions will be typically handled in a cursory manner.In general roleplaying games offer a comparatively wide choice of configurative options, but present many of them in a relatively abstract manner.A particular game or particular group of players, may emphasise one or another (such as vehicle combat) but for every one so detailed, many are handled abstractly.
Another difference between role-playing and other games is that typically role-playing games (including the digital forms) place a lesser requirement on users to provide continuous input.
In digital racing or flying games a player that provides no input for an extended period will likely crash.Most first person shooters require players to be constantly on their guard (at least during times when the game allows player input).Conversely a role-playing game typically allows its player large sections of time when they can choose how much input they will give.Players have at least partial control over the balance between configurative and interpretive in their approach to the game.

Narrative Backing
Role-playing games typically demonstrate strong narrative influences.While the exact relationship between stories and games is still debated, roleplaying games demonstrate more story-like elements than many other game forms.The history of the game world and the narrative support for ingame tasks is more apparent in role-playing games.
Making sense of the game play in these games requires an understanding of the wider game world.For example, in Asteroids a player is not told, nor do they need to know to play the game, what the social structure of the game world is, how interstellar travel world or what race the pilot of their ship is.In a science fiction role-playing game all these elements are likely to be at least known to the player, and possibly important to the actual game-play, as discussed by Tychsen et al. (2006).
Role-playing games introduce this element as a consequence of their individualisation of the characters and their presentation of events in the game world.In fact it could be argued that the narrative elements in role-playing game are a result of other, defining, elements and that it is a corollary, not a necessary element in itself.In essence, role-playing games cause narratives to emerge on a running basis, they do not contain narratives as such.
Players of role-playing games experience a sequence of (typically) related events.These can be said to form a narrative, of some sort, in much the same way that narratives are formed from real life experience.That the traditional definition of story from narrative theory, for example Bal (1997), may not apply to role-playing games⁵ is beside the point, the "story-like" element is commonly strong in role-playing games.This can be seen both in the International Journal of Role-Playing -Issue 1 While the exact relationship between stories and games is still debated, role-playing games demonstrate more story-like elements than many other game forms.
presentation of events, their reception by players, and in the common provision of supporting material, detailing the game world and events in it.

DEFINITION
The above discussion allows for a definition of a role-playing game based on the analysis of existing forms.

Game World:
A role-playing game is a game set in an imaginary world.Players are free to choose how to explore the game world, in terms of the path through the world they take, and may revisit areas previously explored.The amount of the game world potentially available for exploration is typically large.

Participants:
The participants in the games are divided between players, who control individual characters, and game masters (who may be represented in software for digital examples) who control the remainder of the game world beyond the player characters.Players affect the evolution of the game world through the actions of their characters.

Characters:
The characters controlled by players may be defined in quantitative and/or qualitative terms and are defined individuals in the game world, not identified only as roles or functions.These characters can potentially develop, for example in terms skills, abilities or personality, the form of this development is at least partially under player control and the game is capable of reacting to the changes.

Game Master:
At least one, but not all, of the participants has control over the game world beyond a single character.A term commonly used for this function is "game master", although many others exist.The balance of power between players and game masters, and the assignment of these roles, can vary, even within the playing of a single game session.Part of the game master function is typically to adjudicate on the rules of the game, although these rules need not be quantitative in any way or rely on any form of random resolution.

Interaction:
Players have a wide range of configurative options for interacting with the game world through their characters, usually including at least combat, dialogue and object interaction.While the range of options is wide, many are handled in a very abstract fashion.The mode of engagement between player and game can shift relatively freely between configurative and interperative.
6. Narrative: Role-playing games portray some sequence of events within the game world, which gives the game a narrative element.However, given the configurative nature of the players' involvement, these elements cannot be termed narrative according to traditional narrative theory It should be noted that this definition does not provide clear boundaries.Exactly how much of the game world is presented, how wide the choice of interaction possibilities and how much story element is contained vary between the forms of role-playing game and are not amenable to precise quantification.This leads to a blurring of the boundaries between what is and is not a roleplaying game.However, the definition provides very clear support for categorising games, as discussed in the next section.The definition specifically focuses on the structure of the games, not on the playing styles employed within them.This can vary greatly, from player to player and moment to moment, ranging from convincing acting to the purely instrumental and beyond.
The definition also, as a consequence, demonstrates that digital role-playing games do not represent the full spectrum of role-playing games.For example, some role-playing games blur or even remove the boundary between player and games master.Digital role-playing games are more restrictive, with the software having a non-negotiable role and rely on quantitative character representation and event resolution, while not allowing purely qualitatively description or arbitrary resolution.
They also limit, in advance, what portions of the game world the characters can engage.Where a human game master can, on the fly, detail and present any aspect of the game world, this cannot be done in the digital realm, if only through the need to prepare the graphical assets.

DISCUSSION
If the definition proposed above is to have use it should at least be able to distinguish role-playing games from similar forms.In this section a number of game forms are examined, highlighting how the proposed definition distinguished them from roleplaying games while previous definitions do not.
First person shooters fulfil the requirements of many existing definitions of role-playing games.
They have participants, a game world and a controlling power outside the players.The stereotypical form of a Dungeons and Dragons game finds the players involved in a "dungeon-bash".
Here they move through a maze of corridors, killing and looting as they go, This is little different to the play of many first person shooters.It also has player controlled character development and extensive means of interaction with the environment.Where it fails in meeting the definition is that the environmental interaction is not quite what would be found in a role-playing game.In particular, the player has no choice in the interaction with non-player characters -the player is spoken to, but never speaks back.Interacting via dialogue is an important aspect of role-playing games, as noted in the definition.
Adventure games, such as Monkey Island or Syberia, on the other hand, make extensive use of dialogue interaction including, most importantly, giving the player some choice of dialogue options.These games lack character development as it is found in role-playing games, often limit the player's navigation of space and usually limit interaction with the world to dialogue and certain object interactions.
Other examples can be found of games which are similar to, but not quite, role-playing.Such games fit within various of the existing definitions but are excluded by the one presented here.3D platformers have been discussed above.Cops and robbers, and other similar children's games, lack a directing influence which could be labelled a game master.Board games where players take a single character role, such as Zombies and Talisman, lack a game master, impose strict limits on the areas of the game world that can be visited during play and have limited options for interacting with the game world.The existence of a game master (or equivalent) by itself is not enough to make a game a role-playing game.Consider certain double blind board war games.A double blind game is where players have a copy of the game board o which they manoeuvre their pieces.The have only limited knowledge of the movements of the other player.These games may involve a referee, who adjudicates the action and informs players of events outside their control.In fact in one example, Flat Top, the referee is called a game master.Such games have participants, a game world and a game master, yet are clearly not role-playing games.It should also be noted that some of these games, such as variants of Squad Leader, place players in control of pieces representing individual people (although admittedly more than one).These games also differ from role-playing games in the type of interaction with the world allowed, being purely combat-focussed.

CONCLUSION
Role-playing games, although they exist in a variety of forms, which include great differences in player number and cross the divide between the digital and non-digital, possess a range of common features that allow them to be distinguished from other game types.Existing definitions have typically not captured this distinction, typically being concerned with the aspect of role-playing or specific types of role-playing games.An analysis of various examples of role-playing games in this paper has enabled the identification of a range of characteristics.On that basis a definition for them could be proposed which is much more successful at separating role-playing games from other, similar, game forms.
Much of the outline of the proposed definition is not new, sharing game world, characters and game masters with many existing definitions.However the analysis has shown that not only are these important components of a role-playing game but that such a game includes each in a very specific manner, which together provides an a means of identifying this game form.Particularly important to the proposed definition is the treatment of character, space and interaction within a roleplaying game.Characters can develop, under player control, within the game world and through them the player is able to interact with that world in a great variety of ways and throughout the imaginary geography.A cornerstone of roleplaying is the range of imagination it encourages.This recognises one aspect of role-playing games, seen in previous definitions, in that they allow players, through their character to do whatever would be possible in the imaginary world of the game.Of course this misses the role of the game master, and other important points, such as character development, but it does highlight the range and depth of interaction possible within the game world.This use of space and the possibilities for players to explore the game world, in both geographic and configurative terms, have been important omissions from previous definitions.
The role-playing format continues to evolve and mutate.The analysis presented here does not attempt to exhaustively cover all the existing forms (for example, mobile-phone based massively multiplayer role-playing games were not considered).The definition given above is a presented as an advance, not the final word.While our contention is that it covers existing forms it would need to be revisited and possibly revised as new ones emerge.
Having proposed a definition for this idiosyncratic game form it is worth giving some thought to how this relates to definitions of games in general.Roleplaying games are identified as a limit case in by Salen and Zimmerman (2004, p.81) in their discussion of the definition of a game (although it should be noted Lindley (2005, figure 2) places them in the middle of the spectrum of games).If role-playing games are accepted as games (as Salen and Zimmerman further state (2004, p.81), to not so accept them would be: "a ridiculous conclusion"), it is necessary to ask what such acceptance means for the definition of a game.While the purpose of this paper is not to enter into a discussion of a broad definition of games, one particular issue arising from the current examination of roleplaying games deserves further examination -that of outcomes.
Many definitions of a game include the need for some defined goal or outcome, including those of Parlett (1999), Abt (1970), Suits (1990), Costikyan (1994) and Salen and Zimmerman (2004).The definition of a game proposed by Salen  Exactly how such a model is formulated is not of immediate concern here.But what this does demonstrate is the utility of role-playing games in testing more general games theory.Whether they are regard as typical games or not is less important, although limit cases, as they are termed by Salen and Zimmerman (2004) are always a good test of a theory.Their peculiar nature, similar but not the same as other game forms, existing in both the digital and non-digital worlds, the broad scope they give for interacting with the game world, both lends them a fascination for their players and makes them a fertile field for research.In their diversity they display many faces but, to wrench a quote from Campbell perhaps beyond its limit, we may here have started to approach the one face behind the many.
, for, asCopier (2005, p.3) says: "Table-top or pen and paper role-play does not involve any form of physical acting."Whileonecould argue that a definition of pen and paper roleplay could be given which includes such games, it then becomes a definition simply based on the number of participants, which tells us little, if anything, about the nature of the activity so categorised.Similarly, certain definitions of liveaction role-playing appear to include this form of gaming, for example those ofGade (2003, p.67): "I define a larp as: An interactive medium where one or more participants take on roles.The roles interact with each other, and with the surroundings and the world of the larp."and Montola (2003, p.86):

International Journal of Role-Playing -Issue 1 In a non-digital form the players can interact with the game world in any way the game master allows -with the game master improvising resolution mechanisms if necessary -whereas digital forms are limited to the interaction forms implemented prior to play by the game designer.
Yet the dungeon-bash is regarded as a role-playing game and first person shooters are not.Most of the latter lack the character development aspect, which is crucial to role-playing games.They also typically feature a very narrow range of options for interacting with the game world, e.g. the option of communication with dungeon inhabitants which, however rarely exercised, does exist in the tabletop form.Even those digital games that do include character development, and are said to have a roleplaying aspect, lack some other element covered in the proposed definition.For example, Deus Ex, follows the traditional first person shooter treatment of space, dividing it into levels and not allowing players free return to already explored areas.One first person shooter which does allow free exploration and revisitation is System Shock 2.
Costikyan's (1994)4, p.80), for example, includes a "quantifiable outcome".As we have discussed above quantitative elements are not a requirement for a game to be a role playing game in any sense, outcome included.Even in role-playing games with quantitative aspects, the outcome is generally not subject to exact quantification.Role-playing games are able to proceed indefinitely.Costikyan's (1994)argument that they have continuous goals is tenuous at best -all human activity can said to have a goal, even something as simple as passing the time.Including this in a definition tells us nothing, as it does not separate games from other activities.It could be argued, asJuul (2003, p.40)does, that "Pen and paper role-playing games are not normal games because with a human game master, their rules are not fixed beyond discussion."Followingthis one could further argue that role-playing games are not a useful test for general game definitions such as Juul's (2003, p.35) classic game model:"A game is a rule-based formal system with a variable and quantifiable outcome, where different outcomes are assigned different values, the player exerts effort in order to influence the outcome, the player feels attached to the outcome, and the consequences of the activity are optional and negotiable."However, that would almost leave such definitions circular-they are defining the games which meet their definition, and consigning other games to a "half-real" status.That Juul qualified the title of his model with the term "classic" implies that a more general game model may exist.Perhaps that more general model should not have such an emphasis on outcome.