Synopsis

Refereed articles

Information articles

Notes on contributors

Print friendly version

Fable 2 as Simulation, Game and Narrative: A Contest

Adam Ruch

The study of a videogame is a complicated undertaking.  One must address many features of the artefact, ranging from story and narrative, visual aesthetic, and musical score, to gameplay mechanics and interface.  Peter Molyneux's Fable 2 (2008) here will provide a case study for many of these features, as its mechanics design push its narrative to the edge of coherence.  The position this game occupies between algorithmic and expository makes it a fertile area for examining new ways of "story-telling", including the use of narrative architectures, database narrative, or interactive narrative.  Whatever term we apply, we must continue to apply stringent critical awareness to the material itself, as is the aim here.  As this article will assume prior knowledge of Fable 2, it is highly recommended that the reader at least be familiar with the material here1 before proceeding.  Better yet, play the game! 

Fable 2 is a classic example of the tendency of Peter Molyneux's games to err on the side of over-ambition.  Molyneux is well-known for his bombastic press releases and zealous language when describing the mechanics of his games. Words like "completely" riddle his speeches, as well as "revolutionary" or "totally new".  (Molyneux, Peter Molyneux Interview 2008) (Molyneux 2009) (Kikizo 2004) (Wilcox 2008) Molyneux's games tend to focus on "revolutionary"  mechanics to create gameplay experiences (Black & White, Fable and Fable 2 all support increasingly dynamic AI features, for example).  Fable 2 relies on its mechanics to create interesting gameplay, rather than on an original narrative.  However, the trouble that this leads to is that giving the player as much freedom as is allowed in Fable 2 means that the player is often out of control.  This does not lend itself well to the telling of story, and to combat this, often the mechanics are suspended entirely in order to press a certain experience upon the player.  The narrative steps in to simply change the circumstances the player/character finds him (or her) self in, and the player has no choice but to accept these changed circumstances, or abandon the narrative path altogether. 

This peculiar tension between an open world and story-based missions problematises Fable 2's ontology. This exploration of Fable 2 builds on the ludology/narratology debate, attempting to mobilise some of the approaches proposed throughout that discussion.  Is narrative still the opposite of interactivity, as Ernest Adams claimed in 1999?  Similarly, Greg Costikyan suggests that an interesting story will make for a boring game, and vice versa, in 2000.  Perhaps most famous was Markku Eskelinen's quote: "If I throw a ball at you, I don't expect you to drop it and wait until it starts telling stories." (Eskelinen 2001) When Juul wrote that narrative in computer games "tends to be isolated from or even work against the computer game-ness," Fable 2 was ten years away. (Juul 1998) The questions here are whether this ambitious game has actually achieved a new nexus of narrative and game.  Is Fable 2 a holodeck for story-telling envisioned by Janet Murray, or a cyber-stage Brenda Laurel might have imagined?  What is the nature of the experience one has in this kind of simulated environment, and does it satisfy the definitions of "narrative"? 

As a starting point, I suggest adding a third column of rules.  After the style of Gonzalo Frasca and Ian Bogost, that third pillar within a videogame structure is simulation.  Partly inspired by Espen Aarseth's recent assertion that videogames are not (only) games (Aarseth and Calleja 2009), I suggest that when looking at a videogame, one should analyse it in its entirety, rather than stripping out either the narrative or game to reach a core of what is most important.  The question is not which is most important, but how can the rules of each component be aligned within a particular videogame to create a coherent whole.  What follows will tease apart the three sets of rules, and identify areas where their rules are in conflict to the unfortunate detriment of this particular videogame, along with pragmatic suggestions for resolving the clash. 

Getting our Bearings in the Gameworld

Fable 2, like its predecessor, Fable, is an open-world adventure game with overtones of role-playing character development.  To use modern genre parlance, Fable 2 is a sandbox action-RPG.  The gameworld environment of Fable is a relatively contiguous world the player is at liberty to explore at any time (mostly).  There are breaks in between explorable areas, cordoned off by loading screens.  These breaks are between landmark regions and serve to make the land of Albion a large place, without actually allowing (or forcing, depending on one's perspective) the player to travel the vast distances between each region or landmark.  Gameplay is based around the player-character known as "Little Sparrow" , "Sparrow" and later by whatever title the player chooses to earn.  The environment of Albion is something between Tolkien-pastoral idyll, and a pre-industrial city setting.  There are vast stretches of meadow, forest, gloomy swamp, but also settlements from the gypsy camp to Bowerstone: a developing late feudal city with a castle at its centre. 

Gameplay involves a great deal of travelling between locations, as is typical of the role-playing genre.  The character develops as the game progresses, along the lines the player wishes to pursue.  The triad of skills - hand-to-hand combat, ranged-weapons marksmanship and magic-using - is found here divided into Strength, Skill and Will branches of abilities.  One earns experience points much like any other RPG, though they are divided and awarded according to the abilities used to beat enemies.  So, if a player uses a sword in melee combat, he is awarded Strength experience which can only be used to upgrade Strength-combat related abilities.  Each fight will net some general experience as well, allowing for a degree of hybridisation, but in the long run, the character will specialise, as the amount of "XP" required to gain abilities jumps exponentially. 

So far, these features of Fable 2 are typical of the action-RPG.  What Fable 2 seeks to emphasise, however, is the dynamic nature of the gameworld.  Albion is reactive to the presence and actions of the player/character mostly through its NPCs.  Every NPC (other than the major characters such as Theresa, the other Heroes and the main antagonist) maintain an attitude toward the player/character.  This relationship is measured across different axes, from Love-Hate, degrees of Fear or Admiration, and will find the player/character (P/C) attractive or ugly.  The player can don different clothes, hairstyles, make-up and tattoos to influence the perception of the P/C.  Nominated actions, called Expressions will directly influence the relationship the player/character has with the NPC who sees that expression.  Drawing a weapon or casting a fireball will also affect the public opinion of the character.  Fable 2 (and Fable before it) are a mainstream example in the same vein as Façade by Michael Mateas and Andrew Stern, which was much more concerned with the "mechanics" of human relationships than with physical conflict.

Furthermore, certain choices will affect the environment of the gameworld in more fundamental ways.  Should the player make mischievous choices throughout the childhood experience, the Lower Bowerstone area will be run-down, full of criminals and altogether much less pleasant in the adult period of the game, than if the player chooses the more positive actions.  Similarly, the P/C can influence the prosperity of trading areas by frequenting the merchants, or not, of that area. 

Molyneux has worked steadily to produce games of emergence rather than progression, to borrow Juul's terms. (Juul 2005) Games of emergence rely on interactions of relatively few mechanics to, in combination, create huge numbers of possible experiences.  This makes the actual experiences of the gameplay session unpredictable, which destabilises authorial control over information.   For example, the author of a novel will be forced to make a choice: does Little Sparrow cause Lower Bowerstone to fall into corruption, or will his/her actions begin to turn the town's fortune positively?  The message may be largely the same either way: that Little Sparrow is an important, influential figure in the town, but the author of a narrative must pick one or the other.  What a game like Fable 2 can do well is to demonstrate Sparrow's importance, rather than simply say s/he is important. The reactive mechanics that support Fable 2 deliver this material through cause-and-effect models, rather than individual case examples.  This is simulation, rather than representation, but this will be addressed in more detail below. 

The world of Albion is beautiful, with a very particular aesthetic feel.  The NPCs that populate the world too bring their own particular personalities to the experience through dynamic interaction, rather than through scripted event at particular times.  Albion is a fantasy world not unlike Middle-Earth but instead of being delivered via the narrative that is The Lord of the Rings, where the reader absorbs each fact about the world one after the other, the player moves through the world dynamically.  Thus, the design process of world-building in Fable 2 does some of the work normally referred to as "story-teling" using an explorable environment, rather than a linear narrative.  One learns about Albion by "getting a feel" for the world, or by "seeing what happens," rather than by being told about it by an author's description.  How, though, does this interact with the narrative thread that is present in the game?
 
Videogame Ontology

Fable 2 is an enactment of Henry Jenkins' theory about "spatial storytelling" or a "narrative architecture" for videogames. (Jenkins 2004) While I agree strongly with the spirit of Jenkins' theory, a recurring word appears in his definitions that I feel weakens them:

Environmental storytelling creates the preconditions for an immersive narrative experience in at least one of four ways: spatial stories can evoke pre-existing narrative associations; they can provide a staging ground where narrative events are enacted; they may embed narrative information within their mise-en-scene; or they provide resources for emergent narratives. (Jenkins 2004: 123)

In this description, Jenkins uses the word "narrative" five times, and only appropriately once.  The first example, "an immersive narrative experience" is out of place because an experience is not analogous to a narrative.  The only kind of narrative experience one can have is reading/viewing a narrative, a decidedly different kind of experience to the active playing of a game.  The second mention is the appropriate one: by referencing other narratives, spaces can be designed to evoke memories the players already possess, and leverage them effectively.  "Narrative events" may be an appropriate designation, if these are events that are scripted and will always occur the same way in a given gameworld.  These "set pieces" are common in some videogames, such as Uncharted 2, Call of Duty 4: Modern Warfare, which consist almost entirely of set pieces.  Other games such as Grand Theft Auto 4 or Fable 2 contain far more unscripted than scripted events.  These are simply "events" in the context of a dynamic videogame world.  The next mention is similarly ambivalent.  Information may or may not be "narrative" and this has little to do with that information's usefulness in a particular videogame.  Information that can be accessed by the player at a time of the player's discretion can hardly be counted as part of a narrative, when the author has no control over when that information is accessed, or indeed if it is accessed at all.  Finally, the question of an emergent narrative is confounding in its acceptance of a narrative that is authored "just in time" as opposed to being pre-written by the genuine author of the videogame.  A real-time experience as equivalent to a narrative disregards the authorial process entirely, and inserts narrative where "simulation" ought to go. 

The idea of a dichotomic relationship between narrative and gameplay is well-established in game theory—whether a particular scholar supports it or not.  I would argue, however, that these two things are not like terms, and thus cannot be reasonably compared to each other.  Narrative is a method of delivering subject material from one person to another (or to many) in a straight-forward fashion, limited by the linear nature of human perception of time and the host medium's ability to deliver that material.  After Genette, (and other Structuralist narrative theories such as Russian Formalism and Vladimir Propp or Seymour Chatman) one must distinguish between narrative and story.  The story-material (fable/sujet) is the raw material facts of the fictional (or real) world, the totality of events as they "truly" occurred.  Narrative (histoire/récit) is the actual text, the words on the page, in the order presented by the author for best dramatic effect. (Genette 1980, 1988)  Narrative is a way of presenting the story material that exists in excess, and crafting a narrative is as much a process of eliminating irrelevant details as it is inventing relevant ones.  Tolkien, for example, wrote huge tracts of historical background for the world of Middle-Earth that is not included in the narrative of the Lord of the Rings.  Those histories are presented as narratives themselves, and are relevant to that tale, but are not identical to the narrative that we know as the Lord of the Rings. 

Narrative media have existed for millennia, but are not the only way we explain concepts to each other: arguably visual media (painting, photography, sculpture) are not narrative, certainly not exclusively, since they do not move through time.  Frasca notes this, and highlights the computer's power to create simulation as opposed to representative narratives. (Frasca 2001) Similarly, Bogost's procedural rhetoric refers to the computer's ability to use simulations and active systems to describe other systems, and mount "procedural arguments" about the way the source systems operate. (Bogost 2007) Narrative relies on the author's ability to control the flow of information from the source to the receiver—which is the heart of the narratology/ludology debate.  Again, however, the delivery of a narrative is not exactly interchangeable with the delivery of story.  The distinction between narrative and story is imperative: one is the "raw data" of events-as-they-occurred (real or fictional) and the other is a piece of crafted media, handled with deliberate care by an author.  Narrative is not the only way to deliver the same story, though it is naive to believe that the reader/viewer/player's impression of the story will not be affected by remediation.  Humans are able to interpret a single image and gather information from it, instead of reading a linear piece of text detail by detail; this will have an effect on the information absorbed by that receiver. 
That being said, narrative media have enjoyed the most attention and development over the past ten centuries or so.  Since the advent of language, artists have been crafting narrative with increasing skill.  Critics have been pushing these artists to greater levels of consistency, precision and aesthetic unity.  Thus it is unsurprising that in this age we are rather pre-occupied with the notion of narrative: it is a common language most of us can comprehend.  Yet as noted above, the narrative is but one way to communicate.  There are others.  Lev Manovich discusses the new form of the database, which parallels very closely the argument I will make here. (Manovich 2001) Gameworlds are something of a visual presentation of Manovich's database story.  The database is not linear, not progressive, but experienced at random, or in any number of deliberate paths.  Alternatively, one could conceive of experiencing a database collection of information all at once, the way a computer does when a file is loaded into memory.  It is humans who must scan one line of data at a time, more because of our training than the ontology of data itself.  The way we actually experience a particular videogame like Fable 2 is influenced by characteristics of its "gameplay".  Gameplay can be understood as a particular manifestation of the "non-trivial effort" required of a user of Aarseth's cybertext which allows that person to "traverse the text". (Aarseth 1997) Gameplay consists of the rules that govern what control the player has over the game system, and are usually described by the fiction of the game.  In most cases, certainly with Fable 2, the player occupies an avatar who acts as a cursor within the gameworld/database.  Other games, mostly the real-time strategy genre, literally employ a cursor, combined with a much wider perspective over the gameworld than Fable 2 offers. 

The fictional world within which Fable 2 takes place is called Albion, and Albion is a database.  This is not a metaphor, in that the data that creates the visual representation of Albion exists in computer data on a disc.  An author, or auteur, composed this information in a creative way.  The colour of the sky, the arrangement of buildings, the look of NPCs, are all creative, artistic story-material, but they do not in and of themselves constitute narrative because they exist on the disc in an unordered and a format illegible to humans.  This data is collated, arranged and presented in legible form by the computer, for humans to experience.  The world of Albion is not "created" or even activated in sequence based on a set path through the story material, but is all available at any given time, for the user to access along whatever path he or she chooses.  This is not narrative, though it is a similar act of imparting fiction.  The process of moving through the database does impart information, description, from the author, through the medium, to the user.  That player does walk away from the experience knowing something more than he or she did prior to engaging with the videogame. 

Fable 2, as with many of Molyneux's projects, is directed at simulating life, or at least a particular variety of imaginary life.  Other games take a different perspective on the same task, Sid Meir's Civilisation series takes a long view on life of a culture.  The Sims series, by a very like-minded Will Wright, attempts a similar project at the scale of a group of families in a community.  One can return again to Frasca, who takes SimCity as exemplary of a "simulation" rather than "representation" of a city.  (Frasca 2001) The world we as real humans live in is not a narrative either.  It bears more resemblance to the database simulation described above.  The Sydney Harbour Bridge is stretching over the harbour right now, whether I can see it or not.  The conflict in the Middle East ebbs and flows irrespective of my attention.  The fact that our consciousness is bound up in our physical singularity leads to the idea of narrative: the world seems to occur in a single line, stretching out before each one of us in time, uniquely.  Narrative is storytelling from that perspective.  Simulation is (or would be, in the ideal Holodeck) storytelling from all perspectives.  Whether a human individual can perceive the whole simulation simultaneously or not is irrelevant: the simulation exists, like the real world, all at once.  We may (and often do) take a particular position within that simulation, and experience it sequentially—but do we have any other option?  Bound up as we are to our perception of time, one event following the other, even simulated events have to occur in some order for us to perceive them as we have learned to do with real world events.  In short, narrative is the mediation of a (or many) perspective(s), while simulation is the mediation of a world system.  Finally, the "player-narrative" does not come into being until the player relates the chain of events to someone else, as in any other act of narration.  Working from Genette again, sequential experiences or events are not in and of themselves narrative until they are given form and shape by a narrator, who expresses that narration into a medium to an audience (actual or potential, a novel contains an at least latent narrative even if someone is not reading it at the time). 

Videogames straddle a liminal position between these two media types.  On the one hand, all videogames must be simulations to some degree, as they are computer models of systems.  While the kind of modelling varies wildly, and will not satisfy many scientific definitions of the word, these are perhaps artistic simulations, taking poetic licence to achieve a different end than a scientific discovery.  This term might allay some of the simulation fever that Bogost describes by openly acknowledging the subjectivity of the simulation.  Bogost notes that designating something as a simulation often leads to a distrust or denial of the results (or experience) because the users are aware of the subjectivity of the model rules the simulation is based on. (Bogost 2001: 107-108)  But in a videogame, the subjectivity and simulation bias are precisely the point of the exercise.  The "bias" is also the "art".  The new literacy that Turkle describes is precisely the process of learning a new game. (Sherry Turkle, in Bogost 2001: 108) In order to discriminate between simulations, participants would have to be able to apprehend their rules based on the experience of play.  This is how new gamers learn the finer details of a videogame, and how they go on to pick and choose from those on the market.  This literacy is essentially what Bogost goes on to describe in his later work, Persuasive Games

Competing Rule Sets in Narrative-based Videogames

It is easy to conflate the rules of the simulation with the rules of the game.  This is not always justified, however, and can cause a great deal of confusion in discourse about a particular videogame.  Aarseth has recently noted that videogames are not "games" and I agree with him to the extent that the object known as videogames often contains content which is not game, but film. (Aarseth and Calleja 2009) Perhaps a videogame contains many games in the case of Fable 2, a simple co-ordination game in the form of the Blacksmithing activity is quite distinct from the rules of a fight against skeleton warriors.  In the case of Grand Theft Auto 4, there are shooting challenges, driving and racing challenges, as well as "Easter egg" hunt games involving hidden pigeons.  Each of these activities has a different set of game rules, particularly winning conditions, states and consequences; but very little changes (if anything) in the simulation rules. 

Prior to Aarseth's recent commentary, Ian Bogost performed some valuable work by identifying and explaining the ontology of "game engines" as part of the material of a videogame, but not synonymous with the videogame, nor with its narrative.  I further argue that the engine that drives the simulation is also distinct from the game quality of the videogame.  Game engines vary greatly in size and complexity, and can be utterly unlike each other, but generally they are software development tools meant to abstract away some of the more tedious work of game development.  Physics engines will store consistent models of object behaviour such as gravity and collision detection.  Game engines will often assume perspective, such as a first or third person view, and indeed assume a player/character avatar, meaning that the game developer will not have to construct the computer algorithms for these features.  The game engine will have an impact on whether the videogame features an open-world environment, interactive conversation between characters, gunplay or driving—but the designer's goals and vision will influence the selection of a particular engine over another. 

Videogames "project fictional worlds" as Juul says, "possible worlds."   (Juul 2005:122)  These fictional worlds are built like simulations of the real world, in practical terms, using consistent cause-and-effect models.  Marie-Leure Ryan describes the text as being a virtual reality, conveniently allowing us to reverse the analogy as well.  So videogames are simultaneously computer simulations, games and texts.  The newness of simulation as a medium is that instead of focusing on the perception of the how the world works from a singular and specific, usually human, example, we instead create the rules of how the world works in all cases, and can create millions of test examples to demonstrate the system.  The Havok physics engine is nothing more than this, a simulation of how the physical world works, that can be applied and re-applied to infinite situations.  Other systems, those of human relationships, are much harder to model.   Yet that is the ground which Fable 2 (with The Sims) is breaking.  For Molyneux, rendering believable reactions of physical objects is not enough.  Similarly to the creators of Facade, he seeks to model the social system of human life—at least some version of it.  Molyneux seeks to capture, highlight and express the feeling of being human, albeit in a fantasy world.  "What we need is for the player to experience emotions… including excitement, fear, and achievement… and love." (Molyneux, in Boy of Tomorrow 2007) And within this rule-bound fictional world of physical and social interaction, there is a narrative thread to follow, which introduces its own set of rules. 

Using the above discussion as a platform, the remainder of this paper will analyse the techniques Fable 2 utilizes towards Molyneux's goal, and interpret some results and implications of their application. 

Molyneux commented in an interview about Fable 2 that the videogame was intended to describe what it felt like to attain power.  From this simple statement we can extrapolate a great deal about the success or failure of this game.  As discussed before, the balance of Fable 2 is towards the generalised theme, the system coded into simulation, rather than the narrated example.  Yet the game still contains a narrative-driven set of quests that lead the player/character towards a final goal.  The over-arching plot is the task Theresa assigns the hero: to find the other three heroes in order to defeat Lucien.  Each of the three must first be located, then somehow convinced to join the team.  The first, for example, requires the player character to work up a significant amount of reputation with the local village before the leader will even speak to the player/character. 

Fable 2 is essentially an open-world game, though certain areas are not accessible by the player/character until after certain quests are completed.  Like many of the now-typical genre, the questing can be taken up and left behind at the player's discretion.  One can wander the countryside for hours, work in the blacksmith, buy and sell property or cultivate relationships with NPCs instead of doing quests.  That said, those narrative quests are always there, quite obviously inviting the player to follow a particular, prescribed path.  The relationship these linked quests have with the openness of the world is troubling.  The primary question is one of motivation: why would I choose to take up these dangerous and difficult quests in this open world?  Motivating the player, and the character, to take up these tasks seems under-developed in Fable 2.  This may be partially because of the underwhelming narrative that is told through the "interactive" cut scenes, and the illogical choices that the player's character is forced to make at certain points.  (See links below.)

This videogame has instantiated three sets of "rules".  The first are the fundamental rules that allow the gameworld to exist in a way fungible to the human player: physics, rendering and interface rules.  The second set are the rules of the game that allow the player to take certain actions, gaining points in terms of money or relationships with other characters.  These also include the gaining of experience points, skills, armour and weapons.  The third rule is the motivating imperative of narrative. 

The incentive to complete a videogame lies partially outside the game.  After purchasing a game, one is ostensibly interested enough in that videogame to want to play it.  So there is an external pressure to play all of the game, rather than just the first level and then quit.  In games of progression, the only (normative) way to play the game is to progress forward.  So, rather like finishing a book, the "beating" of the final boss has been a given desire of the player, and is rightly relied upon by the designer in games of progression.  Because of this relative predictability and the avoidance of significant decision-making, strict games of progression can generally be more closely linked to narrative, so the incentive to finish is increased by the desire to see a dramatic narrative played out to its resolution.  Games of emergence are much more complicated. 

A game of emergence relies more on the systematic rules to create unpredictable gameplay, allowing, or forcing, the player to contemplate the rules more than the narrative.  In Fable 2, there is less impetus to beat the final boss, simply because there is so much else to do.  My character is not prevented from living quite a good life in Bowerstone by the evils of Lucien who is far out of sight, out of mind.  The choice is open to the player to take up the adventure, or not.  Like Grand Theft Auto, inFAMOUS, Prototype, Crackdown, Just Cause, Assassin's Creed and other "sandbox" games, there are many other activities available apart from the progressive story missions.  If the player chooses not to progress beyond Bowerstone, there is little pressure applied after the initial urging from Theresa, and the much earlier scene of the player/character's sister's murder. 
This is the crux of Fable 2's problems: the simulated world is very interesting, and fun to play in.  it is much more like occupying an avatar in The Sims where one can increase in blacksmithing, bar tending and lumberjack skills, while acquiring money to buy real estate.  The player can maintain multiple families in multiple towns, having children and affairs with other women (or men in both hetero- and homosexual relationships).  The player/character can become fat or thin, acquire dozens of different pieces of clothing to dress him or herself differently.  There are also more formal side quests involving "demon doors" and gargoyle statues to spend time on.  Albion extends far beyond what is required of the narrative involving Sparrow, Sparrow's sister, Theresa and Lucien, so much so that the world seems somehow divorced from that narrative, because there is so little effect of one on the other.  This is most simply demonstrated by the seemingly innocuous fact noted earlier: the only NPCs that the player/character does not maintain a love/hate, attractive/ugly relationship with are those that are important to the narrative.  The player is unable to interact with those major players in the same way as is possible with the other hundreds of characters throughout the gameworld.

Bogost's "procedural rhetoric" is a useful concept to apply to this rift.  By procedural, he nominates the active, algorithmic nature of a simulation and rhetoric is the art of persuasive expression. (Bogost 2007: 1-64) His work mainly focuses on persuasion of players about subjects relevant to the real world, but persuasion can be somewhat more innocuous; simply a way of suggesting that the fictional world works according to some system.  There are, as described, multiple systems or procedures in Fable 2, and one will contradict the other, creating a disharmony that weakens both.
 
Proposals for Procedural Harmony

One can imagine a game that would encourage the player to leave Bowerstone to fight Lucien by making Bowerstone (or wherever the player decides to while away his or her time) a less and less hospitable place to live.  Like Frodo Baggins or Luke Skywalker, the Hero will leave if home is threatened in tangible ways.  The town could slowly become inhabited by aggressive agents of Lucien, as a kind of protectorate.  Important NPC merchants could be abducted.  The player/character's family (which Molyneux stresses as of the utmost importance to the his ideal player) could be killed or abducted by those agents.  Instead, however, the player/character can gain power, wealth and fame without actually achieving victory over his/her nemesis.  Instead of a yellow question mark over the head of the next quest-giver in the chain, the world of Albion should push the player/character towards his next goal.  Arguably, these hypothetical events that lead the player to take action are "narrative" and though they might share some features, by applying Genette's definition strictly, we require a new word to describe these contingent, unpredictable and conditional events.  Jenkins might be tempted to call these events part of the "emergent narrative" though I remain dissatisfied with using the word narrative in this way.  The actions of the player afterward are again not strictly narrative until they are packaged as such by a conscious narrator.  They are instead, gameplay experiences, no less meaningful in their own right, no less capable of teaching the player something, but not indistinct from narrative composition that might seek to do the same thing. 

As I have described before, the videogame medium is ill-suited to the dual purposes to which it is often simultaneously put.  The first is generally the more obvious one of telling a story by putting the player in the shoes of a character who is restricted to a certain set of actions along a well-described and limited adventure.  This is the model of videogames from Super Mario Bros. right through to Halo 3.  Yet the medium is capable of the far freer simulation-based game we see beginning to flower in the open-world or sandbox genre.  Still, these open games cling to the idea of telling a particular story, by using the decidedly "incoherent" (Juul 2005) elements such as the question mark/exclamation point above a quest-giver's head.  These two modes of gaming are so far not well-balanced, as demonstrated by Fable 2.  The first tells the story the author has in mind, reveals something of the mind of that author perhaps, but can certainly be read for a message.  The other is more unfamiliar territory; the free mode of play tells the player about the designer's vision of the world, but also about himself, reflected in his own choices throughout the game. 
The first, the game of progression, posits a win condition, and the player must simply agree with this as a valuable outcome.  This is reinforced by the dramatic narrative pull towards resolution.  Once this agreement is established, the player moves through obstacles hindering his progress towards that goal.  Each obstacle can be overcome, and should the player fail to beat a challenge, he will try again until the challenge is defeated.  Each available choice, whether it is between left and right hallways, between handgun or rifle, or between saving one party member or another, can be measured against the benefit or liability the results of each choice represents.  This requires that each choice be weighted mechanically to provide the player/character with either a benefit or a liability.  The game designer must decide which one of these choices will provide which result.  In more complicated games, several choices can be offered with varying degrees of benefit or disadvantage, but the end result is the same.  The player does not make a meaningful choice, the designer does.  The player only elects to follow the path of least resistance or not, with the incorrect choice stalling progress. 

In an open game of emergence, in a simulation-based game, "progress" is a much weaker term.  In this kind of videogame there are either no prescribed goals, or there are so many that none are the ultimate winning condition.  With this kind of game, the designer should incorporate consequences into the choices available, reactions to the player/character's actions, but these should not be readily identifiable as good or bad, and only good or bad.  In an open-world game, the player is responsible for judging his or her own actions in an interpretive sense, rather than in an empirical one.  Choices should be ambiguous, as they are in the real world.  Real human beings do not have the benefit of a manual that describes the point value of each choice made throughout life.  Further, life does not stop and let a person try a challenge again.  We act, are reacted to, and endure the consequences of our decisions. 

Fable 2 juxtaposes these two modes, to its detriment.  There is a narrative adventure of progression within this open world simulation.  That open world damages the narrative, though, by weakening its power over the game experience.  The quest adventure relies on the player believing that defeating Lucien is a valourised win condition, that is, desirable under any circumstances, and by any means necessary.  The decisions involving these quests become mechanical, economic and empirical.  "What choice will lead me to victory in the shortest possible time, along the most efficient route?"  By embedding these kinds of decisions in a much larger world where many decisions are much less straight-forward, both kinds of decisions are rendered less meaningful.  The player is torn between experiencing Albion, and "playing" the story.   We can interpret this as a disconnect between the rules of Albion as a simulation or open-world sandbox game of emergence, and the relatively aligned rules of a mission-based game of progression and the narrative.  Fable 2 is not the only game responsible for this.  Though generally better, Grand Theft Auto 4 suffers from this somewhat when the narrative asks us to feel touched by the death of certain characters, when the wider world allows us to cause the deaths of any number of civilians we choose.  The significance of death, at least, is completely different within the narrative and the simulated world. 

One can propose that Fable 2 is very close to success with this model, however.  A videogame could very easily examine the mindset that "victory at any cost" in a simulation world model.  The designer could then establish a set of techniques that would lead the player/character to victory over the adversary in a short period of time, but at great moral cost.  There would also have to be a longer, more arduous path, one of personal sacrifice for the player (in terms of time) and the character (in some narrative way) that would also lead to victory over that adversary.  However, the game could not end after the defeat of that enemy.  The player would have to observe the results of his decisions along the way.  Fable 2 comes very close to this, but fails to capitalise on the framework it erects.  Possibly the single greatest flaw is that no matter what path the player chooses, the purest good or vilest evil, the three choices at the end of the game remain the same and they are not actually fungible in subsequent play.  Where is the sister we can bring back to life?  Who were the millions of people Lucien was responsible for killing?  I did not notice the vast fields of dead in my adventures through Albion.  And wealth beyond imagination would be something closer to infinite than one million gold, which can be spent in a single transaction, by buying Castle Fairfax.  Describing this hypothetical game in terms of "what if" and parallel logics is suggesting a new set of procedural rules, which work in concert to persuade the player that victory can be achieved in more than one way, but if one choice is made, then some consequences will be suffered, else there will be different consequences.  This system is a more coherent procedural rhetoric than those Fable 2 employs.  

What Fable 2 can teach us is that it is technically possible to create a reactive, complex world full of human relationships between player/character and NPCs.  In Fable 2 these relationships may be somewhat primitive, and based on a limited number of metrics, but they are interesting to play with.  This model could be further developed and lifted out of its host game the way that physics engines have become independent entities, and applied to other games.  Fable 2 presents the player with a great many activities to pursue, and the freedom to choose which to take up, and this can be (in the broadest terms) fun.  The consequences of the actions are what make those activities interesting, though, and Fable 2 fails to deliver convincing consequences in many cases.  As explored earlier, what is the consequence of not pursuing Lucien, and simply becoming a land baron?  Fable 2 also delivers precise and non-negotiable consequences to other actions that are less than logical.  So, on the one hand, Fable 2 leaves the player free to make up his or her own mind about the vast majority of the activities available in-game, but force a certain interpretation of a smaller number of ostensibly more important (narrative) actions.  Further, while pursuing the narrative quests, the game suspends much of the freedom of action the player/character enjoys during the rest of the game. 

While in Bowerstone, for example, a player/character can suddenly go berserk and start slaughtering civilians.  This results in terrified onlookers running away from the player/character, racking up huge evil and fear points, and summoning a legion of guards to defend the town.  The consequences of this are that the player will probably be run out of the town, the NPCs will be dead, of course, and the player will have difficulty buying items from a dead vendor.  There is nothing to stop the player from taking up such a course of action, but he will have to suffer a set of results.  Yet while in the Spire tower Lucien rebuilds, the player is never able to simply attack Lucien outright.  We must instead endure ten years of torment in order to rescue Garth, then flee the castle—again, without ever actually striking the nemesis who stands right before the player/character. 

Timothy Walsh notes in his article "Everyday Play: Cruising for Leisure in San Andreas" that Grand Theft Auto: San Andreas does not change the rules of the game during mission-based episodes, which is also true of the subsequent GTA4. (Walsh 2006: 129) That is, if a player can drive a car off a ramp and cause another truck to explode during "free-play" outside missions, he can use that same strategy during a mission to launch an attack on mission based enemies.  This is a much more coherent approach to game design, contrasted to the sudden emasculation of the player/character in Fable 2.  Indeed, as noted above, though part of Molyneux's project was to highlight the feeling of power contrasted to powerlessness, at that point in the game missions, it doesn't make sense.  At that point in the narrative, the player/character has the power of Strength, Skill or Will, and should simply throw a fireball at Lucien and be done with it.  If any other NPC opposed the player in this way, the rules of the wider gameworld would allow that kind of reaction, but while pursuing the narrative, we again see a different set of rules applied. 

Still, the framework of Fable 2 seems like a promising platform that other games could develop from.  Almost like a physics or game engine, the work Lionhead Studios have done to create interactive NPCs should probably be exported from Fable and into other similar open-world games to add a layer of interactivity and realism to games that would benefit from it.  The game oscillates between admirable subtlety and deplorable bluntness.  Albion has a wonderful flavour to it, distinctly its own atmosphere provided by a quirky sense of humour, competent voice acting, and believable architecture.   These facets are difficult to quantify, which is partly why they work so well; they are not obvious or heavy-handed.  The narrative quests, however, are exactly the reverse.  They are delivered from on high as imperatives.  With more practice, these subtle techniques of flavour will be harnessed to push the player to act in relatively predictable ways, and will deliver a more coherent experience, more in line with what the auteur would wish to express. 

References

Aarseth, Espen. (1997) Cybertext: Perspectives on Ergodic Literature. Baltimore, Johns Hopkins University Press.

Aarseth, Espen, and Calleja, Gordon. (2009) "Conference Lecture." The Third Place: Computer Games and Our Conception of the Real. University of Oslo.

Bogost, Ian. (2001) Unit Operations: An Approach to Videogame Criticism. Cambridge: The MIT Press.
 - (2007) Persuasive Games: The Expressive Power of Videogames. Cambridge: The MIT Press.

Adams, Ernest. (1999) "Three Problems for Interactive Storytellers." Gamasutra, December 29, 1999.
Boy of Tomorrow. "Fable 2 is All About Love." gaygamer.net. 2007 8-March. http://gaygamer.net/2007/03/molyneaux_fable_2_is_all_about.html (accessed 2009 10-September).

Costikyan, Greg. (2000). "Where Stories End and Games Begin," Game Developer, September 2000.
Eskelinen, Markku. (2001) "The Gaming Situation" Game Studies, vol 1, 1 available at http://gamestudies.org/0101/eskelinen/ (accessed January 2010).

Frasca, Gonzalo. "Simulation 101: Simulation vs Representation." Ludology.org. 2001. http://www.ludology.org/articles/sim1/simulation101.html (accessed June 2009).

Genette, Gerard. (1980) Narrative Discourse. Translated by Jane E. Lewin. Oxford: Basil Blackwell.
—. Narrative Discourse Revisited. (1988). Translated by Jane E. Lewin. Ithaca: Cornell University Press.

Jenkins, Henry. (2004) "Game Design as Narrative Architecture." In First Person: New Media as Story, Performance and Game, edited by Noah Wardrip-Fruin and Pat Harrigan. Cambridge: The MIT Press.

Juul, Jesper. (1998) A Clash Between Games and Narrative." Master's Thesis, available at www.jesperjuul.net/thesis/ accessed January 2010.

Juul, Jesper. (2005) Half-Real: Video Games between Real Rules and Fictional Worlds. Cambridge: The MIT Press.
Kikizo. "Fable Kikizo Documentary." GameTrailers.com. 2004 30-November. http://www.gametrailers.com/video/kikizo-doc-fable/1248 (accessed 2009 йил September).

Manovich, Lev. (2001) The Language of New media. Cambridge: The MIT Press.

Molyneux, Peter. "Peter Molyneux Extended Cut." GameTrailers.com. 2009 13-April. http://www.gametrailers.com/video/peter-molyneux-lionhead-studios/47899 (accessed 2009 September).

—. "Peter Molyneux Interview." GameTrailers.com. 2008 15-October. http://www.gametrailers.com/video/peter-molyneux-fable-ii/41557 (accessed 2009 September).

Walsh, Timothy J. (2006) "Everyday Play: Cruising for Leisure in San Andreas." In The Meaning and Culture of Grand Theft Auto, edited by Nate Garrelts, 127-142. Jefferson: McFarland & Company, Inc.

Wilcox, Jon. "Molyneux: Black & White Team 'Revolutionising' AI." totalvideogames.com. 2008 йил 19-February. http://www.totalvideogames.com/Black--White-Team-Title-unannounced/news/Molyneux-Black-amp-White-Team-039Revolutionising-AI039-12270.html (accessed 2009 September).

Young, Shamus. "Fable 2: Plot Failures Parts 1-4." TwentySided. 2009 4-February. http://www.shamusyoung.com/twentysidedtale/?p=2105 (accessed 2009 September).

1 Videogames Reference Wiki, maintained by the author.  Fable 2 Page URL: http://flickeringcolours.net/vgwiki/index.php?title=Fable_2