This chapter elaborates on and synthesizes the results from the articles presented at the end of this study. The discussion here also draws on the previous theoretical chapters in a way that was not possible in the articles.

The three research questions answered in this study are presented in the introduction, but are repeated here for the sake of clarity. The three questions are:

  1. What are the preconditions for understanding how games create meaning? (Paper 1)
  2. How should games be defined and delimited? (Papers 2, 3 and 4)
  3. How do ludonarrative games create meaning? (Paper 5)

The following chapters present answers to these questions.

Preconditions for Games Creating Meaning

As was discussed earlier in this study, games differ from other media by being procedural, interactive systems that have some media-specific ways of meaning-making (see e.g., Bogost, 2007; Kücklich, 2002; Wardrip-Fruin, 2009). This chapter will look at three important aspects of digital games as preconditions for understanding how they create meaning. The three aspects discussed are procedurality, interactivity and temporality (see also Paper 1; cf. Murray, 1997, p. 71).

Previous chapters have highlighted the ways games are procedural. Games are systems that consist of processes. Some of these processes may be simple, but they may also be as complex as the current computing power allows.1 An average digital game will have a variety of systems handling the different aspects of the game from feedback systems and graphics rendering to world simulation and AI mechanisms.

As was shown in the previous chapters, these systems are not neutral, but necessarily provide a certain perspective to things (e.g., Bogost, 2007, p. 45; Wardrip-Fruin, 2009, pp. 4–5). This should not be construed as negative, since this perspective is a requirement for understanding something in the first place (e.g., Gadamer, 1960/2004, p. 277). Instead of hoping for some kind of impossible objectivity in the processes we create and engage with, we can hope to achieve “informed subjectivity” (Kücklich, 2002, p. 101) that is aware of the perspective from which we approach things.

This is necessary since games are not just procedural but also interactive. The interactivity of games is essential to this study because “in games we have to interpret in order to be able to configure”(Eskelinen, 2001). It is impossible to interact with games without first interpreting them. In this sense, interpretation is the cornerstone of all gameplay (cf. Eskelinen, 2012, pp. 277–279).

However, interactivity is not a simple concept. It helps to consider the different senses in which the concept of interaction is used (see Paper 1; cf. Ryan, 2002, pp. 595–603). Jensen (1998, pp. 188–190) identifies three traditions on how interaction has been understood, originating in sociology, communications and informatics, and their related fields.

The sense of interaction related to sociology refers to two or more people communicating, which is a very human-centric way of understanding interaction. In communications, interaction is further divided into two different approaches. In the cultural studies tradition, interaction has often been used as a way of referring to how people form interpretative relationships with texts. The concept of interaction has also been used in analyzing how people relate to different media, particularly as the consumers of mass media. In informatics, interaction refers to the relation between humans and machines.

While this study is most closely related to the tradition of communications, it benefits the most from the way informatics understands interaction. Jensen (1998, p. 200) also warns against confusing interaction with interactivity. While the former may refer to many kinds of different relations between humans, and humans and machines, interactivity is best reserved as a quality of media.

Interactivity implies that the player has some amount of agency in playing. The player interprets the game, makes choices based on those interpretations and then returns to interpreting the situation based on the consequences of the choices made. The consequences are not always the ones the player envisioned when choosing something. The player may fail to reach the desired goal either through having a flawed interpretation of the situation or by failing to perform the chosen action adequately. It is also entirely possible for the player to reach the goal by failing (Eskelinen, 2012, pp. 288–289).

For example, in a platformer game, the player may attempt an impossible jump and fail, but still end up on a platform that enables the game to continue. That situation could be described as a failure in both interpretation (which platform to aim for) and performance (how to get there) that nonetheless results in a successful resumption of the game.

The interpretive situation described above is very much like the one portrayed by the hermeneutic circle discussed earlier, except that the player has more power to determine the conditions of the interpretation. This makes the interpretation of games a combination of three different factors: the creative expression of the developers, the context of interpretation, and the player’s choices and preferences. This is not radically different from interpretation in other situations, but the configurative power gives the player more space to explore the meanings in a game.

Interactivity also requires considering the issue of temporality. As was discussed earlier, it is important to distinguish between two types of interpretation: the interpretation that is necessary to continue playing the game (real-time hermeneutics) and the interpretation of games as objects with meaning (game hermeneutics).

While the second relies on the first, they have different purposes. Making interpretations while playing a game aims mainly at the continuation of play, while interpreting games outside the moment of play may range from the aesthetic to the critical. The most common example of the second kind of interpretation is probably conducted when reviewing games, although most reviews are more focused on technical execution and entertainment value than aesthetics or critique.

Interpretations that happen during play have time-constraints and are limited by the ability of the player to perceive their surroundings while playing the game. This might mean that there is not enough time to perceive everything that happens in the game, although a well-designed game will provide enough information for a competent player to ‘go on’.

The idea that games happen in real-time can have misleading connotations (see also Paper 1). There are important exceptions: many games work with turns without set time limits. Asynchronous games are built on the idea that players can interact with the game when they have the time to do so. It is also possible that the game mixes real-time and turn-based temporalities. A game may proceed in turns, but while the game is processing the commands of the player, it may show animations of what is happening in the game, for example by showing the movements of units. The game is still broken into separate units of time, but those units show the passing of time.

Even games that happen in real-time may not be fast: glaciers melt and mountains erode in real-time. There are different speeds of real-time, so a more fine-grained understanding of temporality is needed to make sense of the issue. One basic distinction to be made is to separate different levels of temporality: fictional time, game-world time and real-world time (cf. Zagal & Mateas, 2010, pp. 846–851). Real-world time can further be divided into the player’s cognitive level and hardware level.

The fictional time of the game relates to the narrative or contextual elements within the game. For example, turns in the game might be called months. Taking 12 turns would then equal a year of fictional time. However, in game time that year might take only a few minutes, with turns flowing by as the player clicks on the button labeled “next turn”.

The real-world time of playing through a game sequence might or might not equal the game time. A player having trouble with a particular sequence may need a dozen tries before completing a difficult level. However, in game time only the successful sequence of play might count as game time. On the player’s cognitive level all the play-throughs would certainly make an impression, first by frustration and then with triumph.

Focusing on the hardware level is not really worthwhile for the purposes of this study. The hardware level is best measured in units, such as the gigahertz, that are too fast to affect the meaning-making in a concrete way. A notable exception would be a situation in which the hardware is too slow to run the game smoothly enough.

To sum up, games are procedural systems that often work in real-time but may not require fast reactions from players. The processes in games are not neutral, but have certain expectations and conditions built into them. Players make interpretations of the game and act on those interpretations, either succeeding or failing to reach their goals. Depending on the consequences of their actions, players then make new interpretations that take into account the earlier consequences. While interpreting in order to go on in the game, players also create supporting structures that enable them to make critical and evaluative interpretations of the game. This typically happens outside the moment of play when the player has the possibility to reflect on the game as a whole.

This chapter did not discuss narrative games or how narrative affects the meaning-making process of digital games. That will be the focus of a later chapter.

Defining Games

Up to this point this study has assumed that the reader knows what games are. They have been discussed from many different perspectives with the assumption that the reader has no trouble following the discussion, regardless of the fact that no clear definition of games has been provided.

This approach is in a sense very Wittgensteinian (2001). He thought that in everyday life we rarely have problems in understanding what words mean. It is only when a definition is demanded that we stumble and become unable to explain exactly what a word really means. The rest of this chapter shows how this affects our understanding of games (see also Paper 2).

There is no shortage of game definitions. Both developers and scholars have struggled to understand what games actually are, and this has resulted in a deluge of slightly different definitions. Some of these have tried to define only digital games (Esposito, 2005; Tavinor, 2008), while most try to define games in general (Abt, 1970, pp. 6–9; Avedon & Sutton-Smith, 1971, pp. 2–8; Costikyan, 2002, p. 24; Juul, 2003; Maroney, 2001; Myers, 2009; Salen & Zimmerman, 2004, p. 80, cf. 86–90; Suits, 1980; Waern, 2012; Whitton, 2009, pp. 20–28).

Often these definitions highlight the similar aspects of games: they are systems, have rules, are played and have goals. Sometimes some of these aspects are left out and some other features – like players, competition or detachment from everyday life – are brought to the fore. These are all valid definitions and this study will not try to refute any of them. On the contrary, this study considers all of them as potentially useful and fitting. This should not be read as an endorsement of an “anything goes” mentality. Instead, the focus is on what the definitions are doing and how well they are doing it.

The approach taken here is based on the philosophical research of Ludwig Wittgenstein (1953/2001). While he famously discusses games, his remarks are not to be taken as a theory of games. In short, Wittgenstein is not doing game studies. Instead, his interest in games is simply to provide an example on how language works. While his ideas about games might be of some interest, his ideas about language are much more useful, even for a games scholar. Wittgenstein’s ideas have been applied in game studies before (Bojin, 2008; Treanor & Mateas, 2011, p. 8; Whitton, 2009, p. 20).

Wittgenstein’s view of language is based on the basic premise that in most cases a word’s meaning is its use (Biletzki & Matar, 2014). Words derive their meaning from how they are used by people, which means that the meanings may be contradictory, complex or vague. There is no single core of meaning for any word, but instead their meaning is composed of groupings of related meanings.

This argument is already present in Paper 2, so it will not be repeated here. However, a brief summary of the main results is in place to provide enough context for the discussion that follows.

Games are an example of the type of language-games presented earlier. There is no single core of what it is to be a game, but different types of games are related to each other through some shared features (cf. Bogost, 2006, p. 5). Wittgenstein (1953/2001, para. 67) called these relations family resemblances. Family resemblances are a network of relations where none of the objects share all their features, but all share some features with their nearest relations.

These family relations are understandable because they are embedded in what Wittgenstein (1953/2001, para. 241) called forms of life. Forms of life are all the social surroundings people live and interact in. These forms of life make the family resemblances meaningful by being embedded in different kinds of social relations and contexts.

Because the contexts in which games are played and discussed are different, it is only natural that there is a lot of variation among the types of games people think of first when thinking of the concept of ‘game’. This leads to different views on the things that are considered central to being a game, on different ways of including and excluding phenomena from the notion of games (cf. Juul, 2003, p. 39; Whitton, 2009, pp. 22–23). Interestingly, the act of excluding some forms of games necessarily creates borderline cases of things that are almost, but not quite, games. These are often revealing of the thought processes behind framing games in a particular way.

Role-playing games are an interesting example of game definitions in two senses (see Paper 3 for details). First, they are often seen as a borderline example of games, or simply as non-games (e.g., Juul, 2003, p. 39; Salen & Zimmerman, 2004, pp. 81–82). Second, they are also interesting in that both players and scholars have a hard time agreeing on what constitutes a role-playing game. For example, in their excellent attempt to define role-playing games, Hitchens and Drachen (2008, p. 13) view the gamemaster as a central aspect of role-playing games. Unfortunately, this leaves out examples that many participants of the hobby would be happy to call role-playing games.2

While role-playing games have traditionally had a gamemaster, it does not seem like a necessary part of role-playing games today. Role-playing game developers have built all kinds of games, questioning many different traditional aspects of role-playing games, including the role of the gamemaster.

This seems like a fortunate example of the process of definition discussed in the theory section of this study. Role-playing games are a cultural and historical phenomenon, and as such their definition will need to change as time passes by and new kinds of examples of role-playing games surface. This process is exemplified by the variety of different forms of role-playing games found today: when role-playing games grew out of war-gaming in the 1970s, neither live-action nor digital role-playing games were part of the phenomenon, but today it would be impossible to deny their existence as part of the practice of role-playing. Role-playing games are a prime example of why the definition of games needs to be a hermeneutic one.

Ludonarrative Games and Meaning

While a big part of this study has discussed things relating to games in a broad sense, this chapter focuses on the issues that are particular to ludonarrative games and meaning. To do so, matters relating to narrativity and narration are discussed here. As discussed earlier, ludonarrative games are a particular case of expressive games. While all games may be expressive in some sense, ludonarrative games are so in a particular way.

In Paper 4 it was discussed how the relation of games and stories has been theorized in game studies. The analysis was started out with the controversial statement that “stories are just uninteresting ornaments or gift-wrappings to games, and laying any emphasis on studying these kinds of marketing tools is just a waste of time and energy” (Eskelinen, 2001). The conclusion reached was that there has been a variety of ways of reading this statement, framing it in an assortment of ways. However, a productive way of reading this statement was discovered and different examples, like Super Mario Bros., World of Warcraft, chess, text adventure games and digital role-playing games were observed through this lens.

While story-elements are less important for games like chess, they become increasingly important as one comes closer to the ludonarrative end of the spectrum. Text games like Varicella, role-playing games like The Witcher or detective games like L.A. Noire would make little sense without their stories. They would most likely be unplayable, as all the narrative cues for understanding how to play them would be gone. For example, what would interrogating a suspect mean in a game like L.A. Noire if the narrative framework for investigating a crime would be removed? Actions like interrogation, accusation and investigation need a framework in order to make sense.

Some readings would claim that ludonarrative games are less ‘game-like’ than other examples, as they require the narrative aspects in order to be played – the core ludic structures are not enough to play them effectively. Instead, this study simply identifies them as games of one type, ludonarrative games, which are not fundamentally more or less valuable as games. It is not inherently problematic to recognize both chess and digital detective stories as examples of games. It is only problematic if we insist on there being a core for games.

Ludonarrative games have particular tools for expressing meanings. For example, they may employ the mechanic discussed above in relation to temporality. Games may span over years of fictional time, without the player having to grow old waiting for the game to finish. Jumps, pauses, summaries and ellipses provide games with tools to play with time in order to provide certain experiences for the player (see Paper 1). Textual cues like “Earlier…” in a Spec Ops: The Line cut-scene can show the player that the game is not played in sequence, hinting at what is going to be experienced later. Or games may intentionally mix the chronology of the narrative, like in the paranormal stealth game Second Sight in which the revelation of the actual chronology of the events is a significant plot device.

Games can also use narrative tools familiar from other media, like focalization, mode of narration and granularity (see Paper 5 for details). Because games are a multi-modal media, they can use all the forms of expression available to literature and cinema, but they also have expressive tools not available for the typical examples of those forms of expression.3

Focalization is one of the possible tools. Focalization is the perspective from which things are seen. This might not be the literal point of view, especially in games in which the point of view can be changed.4 Instead, focalization refers to the narrative perspective: whose perspective is the game using. Despite the possibility of changing the perspective, games are still often focalized through a central character. Games use focalization in a variety of ways, and some have found ways of using changes in focalization to convey certain experiences. A typical example is the change of perspective that follows the character’s death in many of the games portrayed from the first-person point of view: at the moment of death, the perspective floats outside the character’s body to signify that the player has lost control of the character.

Games can use narration in a number of manners by combining different modalities in multiple ways. For example, the horror mystery Alan Wake combines voice-over and dialogue with texts found in the environment. All of these form parts of the narrative. But games also have at their disposal the possibility of using systems for storytelling. For example, Dragon Age 2 begins with an unreliable narrator recounting to an interrogator of past events that the player simultaneously plays through. Because of the narrator’s tendency to exaggerate, the player can easily mow through the hordes of attacking enemies in a cinematic fashion. The game mechanic corroborates the narration by having the player character cause large amounts of damage with their attacks and by providing access to special attacks. When the interrogator forces the narrator to stay closer to the truth, the player character’s attacks become far less impressive.

Games portray things in different granularities. Granularity refers to how fine-grained or detailed some description, graphic, sound or simulation is. Games can vary all of them, choosing an appropriate – or intentionally inappropriate – granularity for each. These are often determined by the conventions inherent to the genre or the media. Some research in game studies uses the concept of fidelity in the same or similar sense than I use granularity (e.g., Breuer, 2010, p. 7; Möring, 2012, p. 2). The difference is slight, but relates to two things: First, fidelity seems to imply a simulation of a “real world” (Breuer, 2010, p. 7), while granularity does not rely on such relation. Second, fidelity is more related to visual studies and cinema, while granularity has its roots in narratology.

A good example of simulative granularity is how a significant portion of games handle character health. Health is often reduced to hit points, one quantity that is easily tracked, simulated, and visualized for the player (Jørgensen, 2013, p. 9, 41). Different attacks on the body are then quantified as loss of hit points.

First-person shooters are a good example, with different games quantifying the hit point loss from gunshots differently. Some games handle all damage from bullets in the same way: each hit is worth a certain number of points. A typical additional detail is to add damage to those shots that hit a character in the head – headshots, as they all called in the gaming lingo. Some games take this logic further, simulating different amounts of damage to different parts of the body. For example, Counter-Strike: Source awards different amounts of damage from hits to the chest, arms, stomach, legs and head.

As was discussed above, choices about what and how to simulate are also rhetorical choices, since they determine what kind of things the game expresses through its processes. For example, the turn-based strategy game Civilization IV simulates pollution and its adverse effects on living in cities. Civilization V removes this feature, removing environmental considerations from the things the rulers of civilizations need to consider. Intentional or not, this has an effect on how the environment is viewed in the game.5 A simulation can be more or less fine-grained and focus on different aspects of the system it is simulating. In the Civilization example, the simulation is less fine-grained in Civilization V, but this has other effects than simply making the game less complex than its predecessor.

Despite the variety of tools available, games tend to utilize only a small set of established forms of narration, perspective and choices in granularity. This is understandable, as the narrative language of games has developed over time and has become bound to certain genre- and style-related expectations. However, the few notable exceptions to these established forms reveal that it is possible to play with conventions and create new kinds of gaming experiences.

Results Summary

Games comprise a variety of different forms, which are clustered in groups that share qualities. For example, role-playing games share more qualities with larp than Tetris. This sharing of qualities can be understood through the concept of family resemblance. Our current concept of games should be understood as culturally conditioned and historical, and suspect to change as the culture around games changes. An example of this kind of change is the way role-playing games grew out of war gaming and then spread to digital platforms as technology enabled this change.

Games have elements that make them distinguishable from other media, but these elements may overlap with other playful or procedural forms of media, for example experimental literature. The distinction between games and other media is not clear-cut and it is conditioned by the surrounding culture that views some media or activity as games and others through a different lens. That distinction may soon become obsolete as the lines may be redrawn when interpreted from some future perspective.

The elements used to identify games include procedurality, interactivity and some special types of temporality. The systems perspective is typical to game studies and it highlights how games are complex systems that consist of processes. These processes codify a certain perspective into the world and may represent ideas to the player who enters an interpretive cycle while interacting with the game. This interpretation works in real-time as the player tries to understand the game through their earlier experiences and prejudices in order to continue playing. This real-time interpretation can then work as a platform on which the player builds an interpretation of the game as an object with cultural meaning.

Ludonarrative games are a particular form of games. They combine generally large amounts of narrative elements to ludic elements. Despite some earlier definitions, they should not be understood as a marginal or an exceptional form of games. Games contain many things and different forms of games can contain different things. In ludonarrative games, the narrative aspects are more prominent than in many other forms of games. This should not affect the evaluations of their value as games any more than the digitality of digital games does.

Ludonarrative games also have certain types of tools for meaning-making that they share with other forms of media, like literature and cinema. Perspective, granularity and narration can be used to convey certain meaning effects to the player. Using these as narrative tools gives designers ways of designing meaning into their games.

  1. Even that might not be the limit. Crysis was known for being too demanding to run on the gaming hardware available at the time of its publication, with technology slowly catching up to meet the game’s requirements. 

  2. For a discussion of examples, see Paper 3. 

  3. I write “typical examples” because experimental cinema and literature play with interactivity, and can also use the techniques mentioned here. They are just more typical to games. 

  4. The exact meaning of focalization is slightly different in cinema studies and narratology. Importing the contested concept into game studies is still in progress, even though some headway has already been made (Ciccoricco, 2012; Nitsche, 2005). This study follows the usage found in narratology more closely. 

  5. The Civilization series of games is not a typical example of ludonarrative games, but the mechanic discussed is a good example of how the focus of simulation creates certain types of rhetoric effects.