Groundhog Day is not in my reading list at the moment. Maybe it should be, and I could swap something out for it (so much for Foucault or Genette). Recently I started thinking about the qualities of playing and restarting in games, and some of the sorts of unusual player behavior that occasionally results. Groundhog day has been described as a sort of cyclic and layered narrative, but I think that the appeal lies in the phenomenon of the experience of repetition. To understand this, we need to look beyond the story itself to the world indicated by the story.

Groundhog Day features the protagonist, Phil Connors, reliving the same day, February 2nd over and over again. Initially, his reaction is shock and confusion,  then he begins to indulge hedonistically, and eventually becomes depressed and attempts suicide in a variety of ways. Finally, in an effort to win the heart of Rita, the love interest, he sets out on a course of self improvement. Phil uses the repetition of time to become a better person, learning to play the piano, and helping the lives of others. Finally he reaches a point where he can improve the world around him and win Rita’s affection.

The film reverberates with conventions of games, the struggle for improvement, perfection and mastery achieved through practice, repetition in the face of failure, and the intermediate freedom that lies between. Groundhog Day represents a simulated world, whose mechanics become more visible when seen in repetition, and are reinforced when perturbed by variation. Like Phil, the audience gains a sense of the depth of the world by viewing it while it is repeated with different perturbations. Murray says that the film “is as much like a videogame as a linear film can be.” (HoH p. 36) This is fairly accurate, but thinking of it like a game yields some interesting conclusions. Murray describes the pleasure of the viewer as savoring the variety of reactions experienced by Phil, but ultimately this is frustrated by thoughts of how the viewer would do things differently.

Players in videogames games often react very similarly to the way Phil reacts in Groundhog Day. When the player first is met with failure and doesn’t know why, they will try different approaches to succeed, and if they continue to be frustrated, will often test the game’s boundaries. Players come into playing games with many different sets of expectations. Some players might be ready to be immersed right away, but other players are less invested in considering the game world as a participatory illusion, preferring to experiment and play with it. These are two major types of activities that players engage in while playing games. I think that most players do some mix of both, but the latter category of activity is often problematic. The activity of experimenting with the world often involves attempting to break it, to find and identify where the boundaries of the world are, and how much change and control may be exerted over the world by the player.

In a game centered on storytelling, an experimenting player will eagerly go to the NPC who is to give the player the key to get to the next area, and punch or kill them. The experimenting player might try to climb on top of the highest building in the game to jump off it, just to see if they can. Playing Facade, the experimenting player will try to flirt with Grace and Trip in the second or third acts (if they can get that far). They will take every effort to perturb and upset the narrative direction of the game, just to see what will happen. Developers have mixed reactions to these sorts of players. On one hand, it is pleasing to have players so interested in a game world that they will experiment with it, but it also makes it very difficult for the developers to present a coherent story or experience.

It is generally thought that the use of cut-scenes in games was due to technical constraints (I don’t have a source for this, but this perspective seems reasonably sound). It is easier to have a cut sequence where a static narrative bit is presented to the player after some sequence of gameplay than to have some complex system of having the player interact with the narrative segments or be able to participate during them. Another way of looking at this is as a way of preserving the narrative structure from the interference of the players. After all, if the player is given freedom within these narrative sequences, the player will try to mess them up, and then the game will need to accomodate for that interference. Ultimately, it is not useful from a game design perspective to percieve player action as interference, but if games are to be used for traditional storytelling*, then the story needs to be protected to the player.

* Whether games should or should not be intended as a storytelling medium is not my point. Games are used for storytelling, as nearly every mainstream game title has a story which unfolds during the course of play. There are of course, many that do not, and those are not the subject of my citique here.

Often games protect the story from interference but limiting the player’s actions while a narrative segment is in progress, or in the case of cut scenes, by preventing input completely (save for maybe a skip button if the player is lucky). However, if a game does not do this, then the player will experiment and behave erratically, much like Phil Connors in Groundhog Day. When he is given freedom without consequences, he begins to break implicit social rules, then explicit rules, becoming more and more erratic with each repetition. At his worst, Phil sees the rest of the characters as pawns or toys in a world which revolves around him, which is exactly the mindset of a player experimenting with a game, an exaggerated version of Phil’s already egotistical and jaded personality.

What is fascinating about Groundhog Day is the manner of its resolution. Much like a frustrated player, he eventually realizes that to progress and move forward, it is necessary to not only abide by the rules of the game, but also master them. At some point, he decides to treat the other characters as people and not pawns. Unlike the player of a game, Phil has no choice, and literally cannot continue until he figures out the rules of the world he is in.

What is at stake here is not a matter of figuring out how to stop or punish the player’s experimental behavior, the world must react to it of course, but it should not be seen as a problem or a thing to be insulated. Instead, it is necessary to figure out how to let the player earnestly want to progress and play by the rules of the game. Groundhog Day is interesting because Phil eventually decides to want to improve, to want believe in the world. It seems like it would be possible to put an experimenting player along a similar journey, to first see the world as a toy, and then as an expressive world which the player will want to participate in.

(Okay, I just put Groundhog Day on my reading list now. Now I need to find something to get rid of.)

One of the most daunting tasks in adapting fictional narratives into games is the treatment of narrative compression. This is the process by which certain events are extended or compressed as seen by the reader. There are two kinds of time in narrative, the time within the story and the time of the narration itself. Every narratologist under the sun talks about this at some point, but the primary scholar who explores on this in detail is Genette. The essence of narrative extension and compression are the means by which the story time, diegetic time, may be compressed by rendering it quickly in the narrative discourse. For example, the sentence, “five years passed,” indicates that a significant period has taken place in the diegetic time, but only was revealed quickly in the narrative itself. Similarly, narrative can dwell exceedingly on details and minutia which require considerable narrative time to read, but may describe a split-second in the story world.

Narrative compression is difficult to consider in adapting fiction to games because games treat time very differently than narrative media do. Film, theatre, and arguably all narrative media have conventions established for defining the relationship between the narrative time and the diegetic time. These relationships exist in games as well, but they are not as thoroughly developed. What conventions they do have are borrowed from other media (most notably film), as well as from other technologies (such as video playback). So for instance, a game might intersperse gameplay with cut scenes representing events happening across time, or the game might use a convention such as sepia tones to convey a flashback. Occasionally, a player might have the capacity to make choices within these periods of compression, but usually it is out of the player’s control. The other model that is frequently used, particularly in simulation games, is the ability to slow, pause, or speed up the passage of time in the world. This is most commonly used in simulation games where the player can configure something to happen and then watch as it does, where the processes in the game world are autonomous, and the passage of time is not tightly bound to the player. Very frequently, though, and this is especially the case in MMOGs, the time of the game world flows concurrently with the time in the world of the player. They might have different rates, but they flow consistently and the player has no power to control it.

Games certainly have systems of conventions for handling time, but something more is needed to be able to handle the sophisticated intertwined layers of player time versus world time required by the adaptation of fictional worlds. Pride and Prejudice has a great many convoluted sections where one paragraph will summarize what happens in the course of a morning, then the following several paragraphs will describe a conversation and some things that happen thereafter, and the next paragraph will indicate the passing of another day. Austen’s writing style has a particular quality of interjecting small observations, that a character might make in one moment, in a section indicating a longer passage of time, where the interjection is meant to be something that happens at some point within this period. Occasionally, some sections will be drawn out, where the characters might be waiting for some event to take place, or for some situation to become resolved.

The natural question to ask in all of this, is what is the relationship between the player and the passage of time? If the player controls a character, then that character is not logically in charge of the passage of time, the character is bound by circumstance, and is subject to the changes of time as described in the narrative. However, if the player is able to change the character’s actions, then the course of events might become very different. The player might want to say or do something more before the end of a scene, or the player might be done with a scene and simply be waiting for it to conclude. To prevent the player from being able to perform actions that the character could have arguably done is depriving of agency. While the character might need to sit through an awkward evening where nothing happens, it is not productive for the player to do so, either. So, it is necessary to extend, in some circumstances, control over time to the player. The logical questions are: what control is possible, and what circumstances should the player be able to exert that control?

One of the older conventions for time control in games is frequently found in RPGs, where the player is given freedom within a particular area of the game, but the diegetic time is effectively frozen until the player performs the next plot related event. This can be something simple such as leaving the house or hometown, or as dramatic as fighting the boss on the floating island that causes the world to end. This idea locks the player into a narrative moment until some special condition is met, which advances to the next moment, where the state of the world is different.

A way to extend this is to think of narrative moments as scenes or situations, where the scene itself has its particular temporal structure. For example, in a conversation, which is necessarily something that would take a reader a long time to read, and a player a long time to compose a response (using whatever conversation system is available), the time spent by the reader or player is considerably longer than the time that would be taking place in the story world, with respect to other events. Thus, time would be slowed down or paused within this situation. A broader situation, such as being bored and awkward at the end of a social event is something that would take a long time in the story world, but takes only a short time to explain or depict to the reader or player. At the end of a stage like this, the player must perform some sort of action to let the story advance, in order to have the time to do anything in case the player is not done yet. This moment of advancement is similar to both the trope in RPGs of needing to find some special action to trigger the next event, but it is also similar to the “next track” convention of a video or music player.

The player could thus have agency over time by having time slow down within more detailed situations, and being able to initiate those situations, and also by being able to skip past scenes when appropriate. Necessarily, it is not possible for skipping to occur in some places, for instance, if the player is being asked a question, or is expected to respond to something, or needs to do something important within the scene, then skipping forward is not allowable. A question emerges regarding how scenes are structured within the course of the game itself. If a scene is the general unit of a temporal block, then scenes must be composed together somehow. Does the player control what scenes he or she is going to participate in? Are scenes composed by the author of the game and the player must play out a fixed arrangement of them? It would be interesting and probably ideal if a player could fluidly tansition from one narrative moment to the next, but this is impossible within a structured system. Exactly what should take its place is not yet clear.

A final issue to consider, which may be the most difficult of them all, is the role of simulation within the play of temporal progression. If other characters are engaged in their own situations, then how does the course of the player’s activity affect their own passage of time? I do not know the answer to this either, but it is a useful constraint to consider.