Key Narrative Strategies in Computer Games. A comparative study

Table of Contents

Abstract. 2

1. Introduction

2. A Brief Overview of Classical Narratology

3. Key Narrative Strategies
3.1. Key narrative strategies by Genette
3.2. Story
3.3. Plot
3.4. Narrative Space and Setting
3.5. Character

4. Difference Between Literary Texts and Story-driven Games

5. Agatha Christie: The ABC MURDERS
5.1 Story and Plot in ABC.
5.2 Narrative Space and Setting in ABC
5.3 Characters in ABC

6. Orwell: Keeping an Eye on You
6.1 Story and Plot in Orwell
6.2 Narrative Space and Setting in Orwell
6.3 Characters in Orwell

7. To Be or Not To Be
7.1 Story of To Be
7.2 Plot of To Be
7.3 Narrative Space and Setting of To Be
7.4 Characters in To Be

8. Conclusion

Appendix

Abstract

A popular activity for many people are computer games, which can attract with compelling lore and stories. Taking Brian Richardson’s cause and effect as minimal standard for a narrative, this analysis aims to show narrative in computer games and motivate why they should be included in literary studies. In order to provide specific examples for narrative, the focus will be on key narrative strategies: characterization, setting (space), story and plot. Genette’s theory about order, voice, and mood functions as the base and is deepened, as well as broadened by further definitions and interpretations. The computer games chosen for this analysis are all based on literary works, in order to not only compare the games, but also have a reference to the original. Hence, establishing the place of games in the narrative field.

1. Introduction

“Narrative, and in particular storytelling, is an important part of the human experience.” (Riedel and Young 217)

Who has not sometimes thought that a literary character is doing something wrong, or should consider other possibilities? Generally in a book, the narrative, including the decisions a character makes, cannot be altered and have to be taken ‘as is’¹. In story-driven video games, the player can decide which decision they want to make, with some limitations.

‘Story-driven’ computer games² fulfill at least the minimal standard of a narrative, which is also in the foreground of the game. This minimal standard is based on Brian Richardson’s differentiation between a narrative and non-narrative cinema: cause and effect. However, the term ‘narrative’ can be understood in different ways and is not clear-cut. In order to provide specific examples for narrative in computer games and why they should be included in literary studies, this analysis is going to focus on key narrative strategies. This encompasses, an analysis of characterization, setting (space), story and plot. In order to also directly compare computer games and literary texts, specific games that are based on novels have been chosen for analysis. These games are Agatha Christie’s The ABC Murder (ABC), Ryan North’s To Be or Not To Be (To Be), and Orwell: Keeping an Eye on You (Orwell). The equivalent novels are Agatha Christie’s The ABC Murder (ABC Murder), Shakespeare’s Hamlet, and George Orwell’s 1984.

Narrative strategies themselves are “techniques writers employ to tell stories” (Arnold). In other words, they are strategic writing methods used by the author to gain certain effects. As an example, using a narrator that is not part of the story creates a more distant and objective view over the narrated action. In his work, Gérard Genette is sensitive to the different modes of narration and proves that narrative strategies can be applied to complex works (McKeon 44). Consequently, his theory functions as foundation for the later analysis. Due to the fact that Genette bases his categorization of narrative strategies on his theory of narrative, a brief summary will be given first.

Genette’s theory is the base model, but more refined theories were formulated later, which is why several different authors will expand upon his work. Marie-Laure Ryan’s definition of story will be used for a deeper, autonomous understanding of the term. Aristotle’s theory about poetics defined plot as ‘whole,’ which will be used in addition to Genette’s ‘order’ in section 3.3. Additionally and not covered by Genette, Sabine Buchholz and Manfred Jahn characterize narrative space within four complex parameters, broadened by Ruth Ronen’s addition of the concept of ‘frames.’ As a means to expand Genette’s ‘voice’ and ‘mood,’ Margolin’s definition of ‘character’ is added. Additionally, Aristotle’s differentiation of characters into mimetic or representational, as well as the further division of mimetic into the semantic, cognitive, and communicative model is used for further specificity. For the differentiation between literary texts and whether computer games could be considered narrative, Frasca Gonzalo’s description of ludology is used. Due to To Be being a detective story, Tzvetan Todorov’s definition, together with S. S. Van Dine’s categorization, will be utilized for the analysis, covering that a computer game narrative can fulfill the standards of the detective genre. Furthermore, Marie-Laure Ryan’s theory on narrative in digital media will be taken into account, however, due to the fact that it is outdated it will be used critically. Due to rapid change in technology and video games, most research in this field is outdated. Hence, most bigger studies are not applicable to today’s modern games.

2. A Brief Overview of Classical Narratology

“A narrative text is a text in which a narrative agent tells a story.” (Bal 16)

Many theories about narrative and what it entails exist, making it harder to define what narratives really are. Hence, narratology deals with this topic and “is the theory of narratives, narrative texts, images, spectacles, events; cultural artifacts that ‘tell a story’” (Bal 3). It includes techniques to describe narrative systems (ibid. 3). Nevertheless, it should be kept in mind that reading is a subjective activity, meaning that every reader can have a different understanding of a text. Accordingly, there are divergent opinions about narratives.

Leading structuralists who dealt with narratology are Gérard Genette, Tzvetan Todorov, and Roland Barthes. As structuralists they did not try to interpret literature, but to disassemble it. These disassembled parts include structures and devices to analyze. Hence, their target was the system behind literary texts, which enabled the texts to have their respective conventions and meanings (Genette 8). Plot structure and the organization of details produce specific effects of, among others, suspense, characterization, settings, plot sequences, and story. Genette widened the restrictive study of narrative, to go against skeptics who claimed that “structural analysis of narrative was suited only to the simplest narratives, like folk tales” (Genette 9). Subsequently, Genette’s theory is used in this work as main base for analysis.

The term narrative itself is harder to define than might be expected, as some think that comic strips, for example, do not belong to the corpus of narrative texts. As can be seen, here the concept of a text can differ. Due to having visual images, comic strips are not solely language texts (ibid. 4). However, Bal’s definition of a narrative text is “a text in which an agent relates (‘tells’) a story in a particular medium, such as language, imagery, sound, buildings, or a combination thereof” (5). Thus, according to this definition a comic strip can be considered a narrative text.

From comics to an even more contrastive medium: movies. Brian Richardson did one simple differentiation between narrative and non-narrative cinema. He followed up on David Bordwell and Kristin Thompson’s example: “’A man tosses and turns, unable to sleep. A mirror breaks. A telephone rings’” (Richardson 93). These are just sequences of images, which do not constitute a narrative. However, “Richardson proposes the following narrativization: ‘The man can’t sleep because he’s had a fight with his boss, and in the morning is still so angry that he smashes the mirror while shaving; next, his telephone rings and he learns that his boss has called to apologize’” (Ryan 2004, 11). Summarized, a narrative is not simply made up of sequences of images, but of sequences with a cause and effect. This means that as long as a computer game has sequences with a cause and effect, it could be considered narrative. The story-driven games chosen for this analysis all have several characters which live in their world and are going through several events during gameplay. Accordingly, they fulfill the minimal requirements for a narrative. Nevertheless, there is a big discussion whether computer games can be considered literature. This difference between literary texts and computer games will be dealt with in chapter four.

It should be noted that although a lot has been written about narratology, Meir Sternberg criticized it for not being researched enough. In his opinion, narrative theory is still in its infancy, as disciplinary foundations still need to be laid (115). Gerald Prince summarizes Sternberg’s critique:

[T]here is no general agreement as to the differences, if any, between a narrative and an action description, a narrative and a story, or, most generally, a narrative and a nonnarrative […]. […] They [narrative theorists] have not sufficiently explored what is narrative in a narrative, what the constants of narrative are (as opposed to its variables), what its narrativity consists of as distinguished from its textuality and the extranarrative elements the latter comprises. (230)

The work of narratologists includes the following: In individual narratives, they seek out recurring structures that can be found in all narratives. The content of the stories moves to the background and structural functions are foregrounded. Herein, conventional criticism which focuses on character and motive is counteracted by foregrounding action and structure. Some theory and analyzes already in place for shorter narrative are often taken, refined, and expanded, so they can be applied to longer and more complex narratives. This also means that narratologists do not only focus on a small number of narratives, but on all narratives (Barry 233). Consequently, the study of computer games should be included.

Structuralists tried to answer the question about what narrative is, in different ways. Summarized, Todorov in his book The Poetics of Prose introduces his theory of the six laws, namely ‘the law of verisimilitude,’ ‘the law of stylistic unity,’ ‘the law of the priority of the serious,’ ‘the law of noncontradiction,’ ‘the law of nonrepetition,’ and ‘the antidigressive law.’ For verisimilitude, Todorov notes that “[a]ll of a character’s words and actions must agree with a psychological verisimilitude–as if at all periods the same combination of qualities had been judged to possess the same verisimilitude” (54); according to the ‘law of stylistic unity,’ the “low and the sublime” (ibid.) are not allowed to mix; any comical version always follows the serious event due to the ‘priority of serious;’ ‘noncontradiction’ means that there can be only one version, for example if a character is called one name, but named differently later, the later one is directly incorrect and inauthentic; the ‘nonrepetition’ law states that there should never be a repetition in an authentic text, however, Todorov does concede that Odyssey has repetitions and even needs them; finally, any digression from the main action is seen, by Todorov, as added later by a different author (54-55). However, Todorov was mainly basing his findings on the Odyssey, which means that some laws are not always applicable: More than one person wrote as Homer, but Shakespeare wrote on his own. Furthermore, Todorov concludes that there is no ‘primitive narrative,’ as “all narratives are figurative” and “there is only the myth of the proper narrative” (55). Contrary to Todorov’s claim that the narrator never lies, a narrator can still withhold important information, or deceive (60).

In order to analyze narrative discourse, it should be considered that it “constantly implies a study of relationships” (Genette 27). In total, there are two relationships: The first relationship is between the narrative and the story. Connected by the same narrative, the second relationship is between this narrative and the “act that produces it” (ibid.), actually (author) or fictively (character talking about their adventure). Genette then attributes three terms to the different types of the word ‘narrative’: story, narrative, and narrating. The story is the signified, the narrative content of a work. Hence, everything that is happening. Next to it, narrative describes the signifier, statements, discourses, or the narrative work itself. Lastly, narrating stands for the author or fictive character (ibid.).

Genette’s terms will be important for the later analysis of narrative strategies in computer games. Accordingly, this is a brief summary:

Abbildung in dieser Leseprobe nicht enthalten

Applied to computer games, this would mean that the role of the author who is narrating is taken by the developer or, fictively, by a character in the game. The signified or the events happening are the story of the game. Finally, the signifier or narrative could be lore (background history), dialogs, statements, discourse, narrative texts in a game³. The narrative can be textually analyzed and the story is directly part in the analysis of characterization, setting, story, and plot. Genette points out his view clearly: “Story and narrating thus exist for me only by means of the intermediary of the narrative” (29). Likewise, the narrative is only a narrative if it is telling a story and being uttered by someone. Here, Genette defines it very precisely again: “As a narrative, it lives by its relationship to the story that it recounts; as discourse, it lives by its relationship to the narrating that utters it” (ibid.). Narrating can rather be seen as an influencing part. A further discussion on the differences between the narrative in games and literary texts will follow in chapter four. For now, in order to not overload this analysis, the player themselves, and who plays, is not relevant⁴.

Narrating can include methods and mechanisms for producing the narrative (ibid. 27). Accordingly, this means that for a reader only the narrative informs about the events and activities that occur in the story. Hence, the narrative has interpretable signs or traces, such as a narrator using ‘I,’ to show that the narrator is a character, or tense to indicate at what time the recounted event happened. Furthermore, the reader has to pay attention to the reliability (ibid. 28). Specifically in the game To Be. This game functions on a meta-basis, which merges the author with the narrator, covered in greater depth in chapter seven. The analysis of narrative discourse, therefore, is the “study of the relationships between narrative and story, between narrative and narrating and (to the extent that they are inscribed in the narrative discourse) between story and narrating” (Genette 29).

So far only the reader has been in focus, whose role differs from the role of the player. This is due to the fact that a game functions differently than a literary text: both media recount a story, however, in the case of games the narrative (the game itself) can become gradient. The player is able to choose their own directions, and thus becomes part of thenarrative. Nonetheless, there is, in most games⁵, a set number of different paths a player can take. Hence, the narrative is in some way restricted again. In the same way, narrating is not as clear-cut in games as it is in literary texts. As narrating is the way the story is told, but the player has some effect on how it is told, the line between the narrative and the narrating is blurred. A game is broadened by all the different ways the game can be played. This means that there are multiple ways to tell a story in a game, and the more options the game offers the more ways there are. For example, a pacifist player will tell the story of a game differently than a violent player, having used a different method for producing the narrative. The game Orwell is one example where different endings can be reached, depending on the way it is played. To Be also offers a wide variety of different paths and endings. More on this will be elaborated in the respective chapters. For now, the key narrative strategies which serve as a base for the analysis of the computer games will be defined more clearly.

Accordingly, computer games can be considered to have narratives, as well as Genette’s story, narrative, and narrating can be adapted to them. However, a computer game has generally more options to tell a story, meaning the difference between narrative and narrating is not as distinct. In order to examine computer games more intensely, the next chapter focuses on key narrative strategies.

3. Key Narrative Strategies

“Narrative Strategies are the techniques writers employ to tell stories.” (Arnold)

In order to study literature, more precisely literary fiction, specific terms are needed for analyzing. Culler argues that Genette did not specify if he used Proust’s novel À la recherche du temps perdu simply to illustrate his development of a narrative theory, or if the theory was developed more as a byproduct for a better understanding of Proust’s great novel (Genette 7). In fact, Genette refused to decide⁶. His dilemma to choose between sacrificing Proust’s novel for the sake of literary theory or prioritizing the novel before the theory makes clear that although it is good to categorize, it should be kept in mind that all works are also originals. There are special features in works that might not fit perfectly into some categorizations, still, this should not exclude them. Michael McKeon argues that “the abstract theory of the narrative mode is largely drawn from the concrete practice of the novel genre” (41). Accordingly, narrative theory “is for the most part really the genre theory of the novel” (ibid.). Computer games were still rather new at the time and not as highly developed as today. Consequently, some of Genette’s arguments need to be modified to include computer games or do not fit. Nevertheless, the most important factor is that Genette wrote his theoretical study, enabling everyone interested “to experience the strangeness of the text” (Genette 10).

3.1. Key narrative strategies by Genette

Main elements are herein the differentiation of point of view into mood and voice. Furthermore, he split up narrative and vocalization, as well as clarifying story and plot (Genette 10). By using computer games, and therein from specific examples, this analysis includes a different media genre with more contemporary examples in order to broaden literary theory.

To study these fields more closely, Genette categorized into order, duration, frequency, mood, and voice. To analyze narrative discourse, he sub-divides into “three basic classes of determinations” (31). These are tense, mood, and voice. The first one deals with “temporal relations between narrative and story,” the second is about “modalities (forms and degrees) of narrative ‘representation,’” and the third relates to “the way in which the narrating itself is implicated in the narrative” (Genette 31). Hereby is meant the narrative situation or its instance, and includes the narrator and their audience, may they be real or implied (ibid. 31). It examines the relations with the subject, or the instance, of the enunciating. In this sense, tense and mood both approach connections between story and narratives, whereas voice deals with the “connections between both narrating and narrative and narrating and story” (ibid. 32).

The narrative voice is related to characters. It shows the story’s verbal representation; to be precise, the information that is given in the narrative. A story can be represented differently, depending on which point of view it focuses on, which perspective and distance the narrator has. Herein also focalization is included. Focalization serves as the point of view from which the story is told. There are many possible ways, but the three main ones are ‘external focalization,’ in which the story is told from a perspective outside of a character, more or less like a camera view; ‘internal focalization,’ which shows the story from the viewpoint of a character inside the story; and ‘zero focalization,’ which could also be called ‘omniscient narrator,’ meaning that the mind of more than one character could be freely entered (Barry 225). If the story is told by a character, Genette then differentiates between ‘heterodiegetic’ and ‘homodiegetic’ narrator. The first refers to a character who is not part of the story and narrates from an outside perspective, while the ‘homodiegetic’ narrator is part of the story they tell. Furthermore, in ‘order,’ Genette refers back to German theoreticians for the distinction between ‘erzählte Zeit’ (narrative time) and ‘Erzählzeit’ (discourse time), in which the former refers to the time passing inside the story and the latter indicates how long a reader needs to read (33). He also calls flashbacks ‘analeptic’ and a flash forward in the story ‘proleptic’ (Genette 48-67, 67-79). This leads to the presenting of the story, the plot. Herein he exemplifies via Scheherazade’s storytelling that there can be ‘embedded narratives’ within stories. The primary story is then the first, most general one, in which all others are embedded. This means Scheherazade is the first story and further stories she tells the sultan, for example Ali Baba, are embedded in this first story. When Ali Baba tells a subsequent story, it is embedded in his own story and further embedded in Scheherazade’s. These embedded stories can, however, be interrupted by ‘intrusive’ frames, which refer back to the first story (ibid. 228-229). In the chapter ‘Narrative of Words,’ Genette then discusses ‘mood’ in which the perspective and distance of a narrator are described (169-185). There are several different methods, for example, direct speech, reported speech, and indirect speech. This is also linked to who tells the story, the ‘voice.’

Summarized, Genette categorized narrative into tense, mood, and v oice, which can be further sub-divided into key narrative strategies. These techniques used by writers to gain certain effects include point of view, audience, characterization, plot, structure, genre, and diction (Arnold). The main focus for the following analysis is on the techniques of story, plot, narrative space, including setting and frames, as well as character.

3.2. Story

It is common practice among narratologists to divide narrative into smaller parts, referring to it as a combination of story and discourse. H. Porter Abbott defined these two elements as follows: “story is an event or sequence of events (the action), and narrative discourse is those events as represented” (16). Accordingly, narrative becomes the “textual actualization of story, while story is narrative in a virtual form” (Ryan 2008, 347). This goes along with Genette’s view of story being narrative content and events happening, as well as the fact that story can be autonomous, but not discourse. Seymoure Chatman divides a narrative text into story and discourse, then story into events and existents, while events are actions and happenings, and existents are characters and settings (19). Due to them being a mental image, or cognitive construct, stories can also be formed as a response to life (Ryan 2008, 347). Marie-Laure Ryan provides a tentative definition of story:

1. The mental representation of story involves the construction of the mental image of a world populated with individuated agents (characters) and objects. (Spatial dimensions.)
2. This world must undergo not fully predictable changes of state that are caused by non- habitual physical events: either accidents (‘happenings’) or deliberate actions by intelligent agents. (Temporal dimensions)
3. In addition to being linked to physical states by causal relations, the physical events must be associated with mental states and events (goals, plans, *emotions ⁷ ). This network of connections gives events coherence, motivation, *closure, and intelligibility and turns them into a plot. […] (347)

As can be seen from these steps, story itself is also closely linked with plot. Already Aristotle called story “the essence of the plot” (65). Furthermore, E. M. Forster divided story and plot with his simple sentences “The king died and then the queen died” and “The king died and then the queen died of grief” (87). Due to having a cause and effect, the second sentence is a plot. Accordingly, the next key narrative strategy that will be examined is plot.

3.3. Plot

For Aristotle, plot is the imitation of the action, as in the arrangement of the incidents (25). According to him, a “proper structure of the plot” is ‘whole.’ ‘Whole’ means containing a beginning, middle, and end, which is why he concludes that a “well constructed plot, therefore, must neither begin nor end at haphazard (sic!), but conform to these principles”(31).

The simplicity of this definition is, however, deceptive. In fact, plot is one of the “most elusive terms in narrative theory” (Dannenberg 435). Some definitions treat plot synonymously with story, in relation with story, or point out its temporal sequentiality. Plot has been used as the translation for the Russian Formalist concept sjuzhet as well⁸. Additionally, the poststructuralist period broadened the concept of plot by adding gender constructions, and the exploration of plot “as a sense-making operation or mental configuration, as a force which affects the reader as a narrative unfolds, and as the interplay of virtual and actual narrative worlds” (ibid.). Furthermore, Bremond is notable in his contribution, because he included “alternative courses of events as part of plot” (Dannenberg 436). New theories regarding the exploration of plot branch out into different theoretical directions: “*cognitive, *feminist, philosophical, psychoanalytical, ethical, and ontological, as well as different genre-based categories of plot and Freudian arrays of plot types (ibid. 436-439), however this extends beyond the scope of this analysis.

In summary, most theories agree in differentiating between story and plot. Along with Aristotle, they see plot as a complex construction out of the basic chronology of events. This distinction and the construction of plot becomes obvious, when considering that it is possible to ‘tell a story,’ but not a plot (ibid.). Hence, the story is told and in this telling, the understanding of the narrative and the temporal events in its chronological order is called plot. On this topic, many refer back to the Russian Formalists who distinguished between the ‘fable’ (the story) and the ‘subject’ (the plot). The story is how events would have unfolded in real life and the plot is the special arrangement of these events given by the author (Todorov 26). Where the story and plot unfolds is then examined in the narrative space and setting.

3.4. Narrative Space and Setting

Although the study about narrative space started slowly, it now has a place in narrative theory, especially due to genres such as cyberpunk fiction, fantasy, science fiction, and dystopian fiction (Buchholz and Jahn 551-552). Previously it was more commonly seen as a general background setting. Notably, a story can be fully understood without a setting, for example when a theater stage is left bare, or the above-mentioned sentence from Forster ‘The king died, and then the queen died of grief.’ However, the notion nowadays is that setting “is as crucial as the locus amoenus in medieval and romantic literature or indicative of the social status of characters as in the novels of Jane Austen” (ibid. 554).

Here it can be seen that the distinction between narrative space and setting is not always clear, or not dealt with properly by every theorist. The most basic definition of narrative space is “the environment in which story-internal *characters move about and live” (ibid. 552). These include “landscapes as well as friendly or inimical conditions (including climatic and atmospheric ones)” (ibid.) and tend to be specific, even though they are indeterminable. As an example, even vague fairytales are somehow specific ‘once upon a time, in a land far away.’ Sabine Buchholz and Manfred Jahn characterize narrative space within four complex parameters:

(1) by the boundaries that separate it from coordinate, superordinate, and subordinate spaces,
(2) by the objects which it contains,
(3) by the living conditions which it provides, and
(4) by the temporal dimension to which it is bound. (552)

This entails that on a coordinate level, the sense of place for the narrator and the character is the same. Superordinate space can contain a geographical position, or the time in which the story takes place, whereas the subordinate space is the immediate surrounding of a character. Here, the distinction between ‘story space’ and ‘discourse space’ should be made clear as well. These are parallel to the German ‘Erzählzeit’ and ‘erzählte Zeit,’ which are mentioned by Genette in his category of ‘order.’ Story space is the immediate spatial environment of the characters, where the action is happening. Discourse space refers to the environment of the narrator and is linked to focalization (Lothe 2).

Ruth Ronen made another division: She defines space as “the domain of settings and surroundings of events, characters and objects in literary narrative, along with other domains (story, character, time and ideology), constitutes a fictional universe” (421). Then she distinguishes between frames and setting. According to her, a frame is a “fictional place, the actual or potential surrounding of fictional characters, objects and places” (ibid.), whereas the setting is defined by being “the zero point where the actual story-events and story-states are localized” (ibid. 423). From her definition, a setting is more stable, like a theatrical stage, while frames can change a lot (ibid. 423-425). For Umberto Eco, a frame “looks like something half-way between a very comprehensive encyclopedic sememic representation expressed in terms of a case grammar […] and an instance of overcoding” (Eco 20). Thus, it is a data structure for representing a stereotyped situation, (cognitive) representations of knowledge about the world as chunks or units of concepts denoting certain courses of events or actions involving several objects, persons, properties, relations, or facts, making them a condensed story (ibid. 20-21).

Furthermore, for Ronen, frames can be either open or closed. Open frame means a space is fully described, whereas a closed frame entails that a space lacks certain descriptions and leaves the reader room for their own imagination. As an example for closed frames, she provides Miss Havisham’s house from Great Expectations, in which, apart from two rooms, the house is never fully described. Because of these closed frames and the non-description, the imagination of the reader is required, which in turn creates a sense of mystery (Ronen 431). Contrarily, open frames are fully described (Ronen 431).

In order to conclude where the story and plot happens, narrative space is the broadest and most general part, then setting defines the direct surrounding of a character and frames can be seen as mini stories that are happening in a setting and can be either open or closed. This leaves as last point to be examined, the characters living in the world.

3.5. Character

In a narratological view, character “refers to a *storyworld participant, i.e., any individual or unified group” (Margolin 52). However, in a stricter definition, it is “restricted to participants in the narrated domain,” which excludes the narrator and narratee (ibid.). In everyday usage, the term includes a person’s personality. Based on Plato, but with a different notion, Aristotle presented the concept that “art is essentially mimetic” (Hagberg 366). However, Paul Woodruff argues “Mimēsis and its Greek cognates defy translation. Besides ‘imitation,’ we find in English such renderings as ‘image-making,’ ‘imitation,’ ‘representation,’ ‘reproduction,’ ‘expression,’ ‘fiction,’ ‘emulation,’ ‘make-believe,’ and so forth” (73). The first view treats the character as a human or human-like being, whereas the latter reduces the character “to a text-grammatical, lexical, thematic, or compositional unit” (Margolin 52). The non-mimetic version can be seen, for example, by the structuralist Barthes (ibid.). The mimetic field of study is mainly divided into “semantic (*possible-worlds theory), *cognitive (readers’ mental models), and communicative (the process of narrative mediation […]” (ibid. 53).

The semantic theory has possible worlds, in which the “character is modelled (sic!) as an individual who is a member of some non-actual state of affairs” (ibid.). In this, a character has, prototypically, human-like properties, but is also located in space and time. Human-like properties include “physical or external, actantial (including communicative), social, andn mental or internal (cognitive, emotive, volitional and perceptual)” attributes (Margolin 53). Furthermore, the character becomes more human-like through “enduring personality traits and dispositions, knowledge and belief sets, intentions, wishes, attitudes, *desires and *emotions, and, of course, internal states and actions” (ibid.). At the very least, the character must contain an agential capacity and should, next to bare existence, have at least one property assigned for each state they exist in. Unlike people, characters are based on the information presented in the text. This means that this information can be radically incomplete. Depending on the narration mode, information about a character can be more, or less restricted. As an example, an omniscient narrator⁹ knows everything about a character and can reveal as little or much information as they please (Prince 2008 443). Contrarily, if everything is described through a camera perspective, only the outside is visible (Margolin 53). It is even possible for a character to solely exist in another’s imagination, belief, wish, or intention. These different presentations of a character make them unique. Each character should indeed be unique and distinguishable from others and have their own identity. If any of these conditions are not fulfilled, a character appears empty and it is even reduced to pure verbal expression (ibid.). However, especially when proper names are involved, characters end up with the same name, even when they are written by the same or different authors. Therein, regarding the question if they are the same individual, opinions vary, but the most convincing view sees the characters and their namesake as a counterparthood, differentiating them and not making them the same. However, this relation is a matter of degree and has no agreed upon minimal condition (ibid. 54).

The cognitive theory approaches the character from the mental model a reader forms in their head while reading. From this perspective, many classification types exist: For example, characters can be flat or round, where a flat character is said to not be able to surprise (Margolin 55). This is also possible for games, if characters are not visible, for example first person characters, or when they are just being talked about and never appear visibly in-game. The characters can undergo mental changes, like declines or developments, especially in different phases. When a character has undergone changes in between phases, another second-order category is needed to integrate these two (ibid. 54-55). The mental model is created based on the identification of textual data, “around which a minimal situational frame is constructed, consisting of this individual, time, place, and state or *event” (ibid. 54). Hence, the data collection from the text is bottom-up, or data-driven processing, as well as knowledge based. The latter includes scripts, schemas, stereotypes, and can include “psychological, social, communicative, and physical dimensions” (Margolin 54). Accordingly, the ‘character’ is intuitively generated by each reader or player and can thus change, based on pre-knowledge and textual understanding. When a character is modeled after a real-world person, model readers and professional literati might prefer the literary one over the actual-world model, while ordinary readers approach it from the opposite perspective (ibid. 55). The same applies to games, which can have characters based on novels or a real-world person. When enough textual data has been collected and a category established, categorization can start, which is then a top down method. However, due to the diversity of characters, the reader might have to adjust their mental model, which keeps them in a state of heightened awareness and on the bottom-up processing (ibid. 55).

In the communicative theory, the question is more about the process of narrative mediation, meaning the division of a narrative transmission into a narrative agent, a focalizer, a narrator, or a narratee.¹⁰ Next to these four divisions are three main questions that need to be answered: “[W]here does information about the individual occupying any of these positions come from? What are its nature and scope? What is its truth-functional status or reliability?” (ibid. 55). Information sources are explicit or implicit characterizations. A narrative agent can characterize themselves, as well as be characterized by co-agents, or the narrator. Focalizers are unable to characterize themselves, but, implicitly, information about how they focalize and how they process this information can tell a lot about them (ibid. 56). Global narrators can only self-characterize, whereas a narratee is always characterized by either the narrator or having the narrator quote the self-characterization. Especially first- person narrators need to be treated cautiously on their reliability: General factors like intelligence, knowledge, honesty, and contextual factors like who characterizes whom, for whom, in what situation, and with what intention should be considered. Moreover, when a narrator characterizes others, they implicitly also characterize themselves. As a convention, characterizations made by omniscient impersonal narrating voices can be trusted and taken as true, apart from narratorial irony. Furthermore, there are three major sources of information to extract narrative agent’s properties, especially mental ones: dynamic elements, static elements, and formal compositional patterns of character-grouping. The dynamic elements deal with the “character’s physical and verbal actions or behaviour, their content, manner, and context” (Margolin 56). The static elements deal with the “character’s appearance, natural setting and man-made milieu, assuming that contiguity implies similarity between physical and mental, the physical serving as signifier for the mental” (ibid.). Lastly, the formal compositional patterns of character-grouping are done by “way of similarity contrast, assuming that forms of organisation reflect forms of content” (ibid.).

This concludes the mimetic, human or human-like treating of characters. The other theory is about non-mimetic characterization, which does not treat characters in a human or human-like manner, even “refuse[ing] to go beyond the textual, intensional, or semiotic profile of the narrative discourse” (ibid.). In this theory, the character is restricted to the text at hand and the information given in it for lexical features, agents, patients, achieving an effect, like laughter or horror, an architectonic pattern, or a functional piece in the plot (ibid.). For the non-mimetic theory, the character has more of an ideological position, a “point of intersection of *motifs or themes, and as an exemplification of an issue, problem, attitude, value, or idea” (ibid.).

According to Todorov, there is “no character except in action, no action independent of character,” where character and characterization are more important, as every narrative “is ‘an illustration of character’” (66), and every “new character signifies a new plot” (70). Due to the introduction of a new character, and a new plot, there is an embedding in the story, through which characters can pass (70-72). This means that stories can have multiple layers with multiple plots, but characters are not limited to one layer and one plot, but can move between them. Only if there is no narrative, there is death (74). Characters cannot exist if there is no narrative about them (76)¹¹.

In summary, characters are storyworld participants, which can be categorized into mimetic, human-like treating of characters, and non-mimetic characterization. Furthermore, characters can move through embedded stories, but without a narrative, they cannot exist. Accordingly, story, plot, narrative space, setting, frames, and characters can together built a well rounded narrative, not only in literary texts, but also in computer games. Nevertheless, there are some differences in the two medias.

[...]

¹ This excludes choose-your-own-adventure books, for example The Cave Of Time.

² The concept of ‘video games’ includes software for all potential gaming platforms. Due to the fact that the term is so broad, this analysis will focus primarily on computer games. This, however, does not mean that arguments brought up in the analysis are solely applicable to computer games. In fact, a lot of console games (Horizon Zero Dawn, Uncharted, and The Last of Us) have intricate narrative stories as well. From here on ‘video game’ will stand broadly for any game, while ‘computer game’ will only refer to games that are made to be played on a computer.

³ For simplicity, it is assumed that a player progresses in a game and does not try to break the gameplay or its mechanics.

⁴ There are several studies, books, and articles dealing with gender and computer games, including sexism. As examples for further information about this topic are: the study about female dislikes by Tilo Hartmann and Christoph Klimmt, Justine Cassell and Henry Jenkins’ From Barbie to Mortal Kombat: Gender and Computer Games, as well as the BBC article from Kim Gittleson “Why does sexism persist in the video games industry?”

⁵ Modern games with an open world style, for example, Far Cry or Skyrim have a lot of different paths a player can choose, but at one point the game ends. Some games leave the player the option to keep playing even after the scripted narrative has ended (Lego Marvel Superheroes 2 or Grand Theft Auto), but the player cannot progress in the storyline any more. Examples would be The Witcher or Diablo.

⁶ Refer to Genette’s own statement from his preface “I confess my reluctance – or my inability – to choose between these two apparently incompatible systems of defense (22).

⁷ The asterisks are part of the Routledge Encyclopedia of Narrative Theory, marking these words as being possible to look up.

⁸ For this see the entry by Ann Banfield in the Routledge Encyclopedia of Narrative Theory: “A term used by Russian Formalists to denote the way in which a narrative text presents (or cues readers to reconstruct) a chronological sequence of situations and *events. In the account developed in Seymore Chatman’s influential book Story and Discourse (1978), the sjuzhet is the ‘discourse’ level of narrative, i.e., the narrative ‘how’ encoding the narrative ‘what,’ which in formalist terminology is called the fabula” (535).

⁹ See for differentiations of point of view Gerald Prince’s entry Point of View (Literary) in the Encyclopedia of Narrative Theory (442-443).

¹⁰ See for this also Aristotle: “the poet may imitate by narration – in which case he can either take another personality as Homer does, or speak in his own person, unchanged – or he may present all his characters as living and moving before us. These, then, as we said at the beginning, are the three differences which distinguish artistic imitation, – the medium, the objects, and the manner. So that from one point of view, Sophocles is an imitator of the same kind as Horner– for both imitate higher types of character; from another point of view, of the same kind as Aristophanes-for both imitate persons acting and doing.” (13).

¹¹ Todorov goes into more detail and describes further the grammar of narrative, the quest of narrative, the secret of narrative, and narrative transformation, which will not be described in detail, as it is out of scope for this analysis.