Ludonarrative Hermeneutics: A Way Out and the Narrative Paradox

Abstract

The practice of designing Interactive Digital Narratives [IDN] is often described as a challenge facing issues such as the “narrative paradox” and avoid-ing the unintentional creation of “ludonarrative dissonance”. These terms are expressions of a perspective that takes narrative and interactivity as dichotomic ends of a design trajectory, mirroring an enduring discussion in-game studies be-tween positions often cast as ludologists and narratologists. The dichotomy of ludo versus narrative is, in itself, problematic and is often the source of the very conflict it describes. In this paper, we investigate this issue through the example of the cooperative game A Way Out, in which two players team up to break out of prison. The game is designed with a narrative twist, involving the escalation and final resolution of the game’s competitive motif in the final scene. To understand the user experiences of this reveal, and the concomitant consequences, we engage in a discursive analysis of "Let’s Play" videos as a largely untapped re-source for research. By analyzing the interactions and performances in these videos, we can more clearly understand player responses to unsatisfying IDN design. As a result, we introduce the notion of a ‘hermeneutic strip’, extending Koenitz’ SPP model to locate and describe the involved processes of narrative cognition in IDN work.

Full text

Ludonarrative Hermeneutics: A Way Out and the
Narrative Paradox
Christian Roth1, Tom van Nuenen2, Hartmut Koenitz1
1HKU University of the Arts Utrecht, Nieuwekade 1, 3511 RV Utrecht, The Netherlands
christian.roth@hku.nl; hartmut.koenitz@hku.nl
2 King’s College London, WC2R 2LS London, United Kingdom
tomvannuenen@gmail.com
Abstract. The practice of designing Interactive Digital Narratives [IDN] is often
described as a challenge facing issues such as the “narrative paradox” and avoid-
ing the unintentional creation of “ludonarrative dissonance”. These terms are ex-
pressions of a perspective that takes narrative and interactivity as dichotomic
ends of a design trajectory, mirroring an enduring discussion in game studies be-
tween positions often cast as ludologists and narratologists. The dichotomy of
ludo versus narrative is, in itself, problematic and is often the source of the very
conflict it describes. In this paper, we investigate this issue through the example
of the cooperative game A Way Out, in which two players team up to break out
of prison. The game is designed with a narrative twist, involving the escalation
and final resolution of the game’s competitive motif in the final scene. To under-
stand the user experiences of this reveal, and the concomitant consequences, we
engage in a discursive analysis of "Let’s Play" videos as a largely untapped re-
source for research. By analyzing the interactions and performances in these vid-
eos, we can more clearly understand player responses to unsatisfying IDN design.
As a result we introduce the notion of a ‘hermeneutic strip’, extending Koenitz’
SPP model to locate and describe the involved processes of narrative cognition
in IDN work.
Keywords: Interactive Narrative Design, ludonarrative dissonance, narrative
paradox, hermeneutics, hermeneutic strip, SPP Model, role distancing.
1 Introduction
The practice of designing Interactive Digital Narratives [IDN] is often described as a
challenge that requires moderation between player freedom and the structured experi-
ence that interactive forms like video games internalize. Here, terms such as “ludonarra-
tive dissonance” and “narrative paradox” are often used expressions of a perspective
that takes narrative and interactivity as dichotomic ends of a design trajectory, reflect-
ing the enduring trope of the narratology vs. ludology debate. However, as we will point
out, the dichotomy of “ludo” versus “narrative” is in itself problematic, and is often the
source of the very conflict it describes as a design challenge.

2
In practice, the terms “ludonarrative dissonance” [1] and “narrative paradox” [2] are
used in relation to unsatisfactory user experiences in terms of agency and immersion
[3]. Little is known, however, about the exact nature of the underlying issues, of the
supposed tension between narrative and interaction. We investigate this issue through
the example of Hazelight Studios’ narrative action-adventure game A Way Out [4], in
which two players team up to break out of prison and escape the authorities. We focus
especially on player reactions to the showdown in the final scenes of the game where
agency is suddenly revoked and the players are forced into a deadly confrontation from
which there is no escape. In order to understand how this reveal and change of gameplay
is experienced, we turn to the material of "Let’s Play" videos on YouTube, which we
consider a largely untapped resource for user-experience research. By analyzing the
interactions and performances in these videos, we can more clearly understand player
responses to IDN design.
This paper argues that the concept of ludonarrative dissonance is insufficient in de-
scribing situations where games let their players down due to tensions between game-
play and narrative constituents. We go on to suggest that the concept should, at least in
this specific case, be extended and better defined by using the construct of interpreta-
tion of the ludonarrative to better accommodate the hybrid nature of the ludonarrative
construct taking both player and game designer into account than the entrenched dicho-
tomic ludology/narratology perspective.
2 Summarizing A Way Out
A Way Out is a game about the prison breakout of Leo and Vincent, two inmates who
meet in jail and become partners in planning a daring escape. Convicted of robbery,
assault and grand theft, Leo is introduced as a confident, pragmatic and headstrong
character with a tendency towards violence. Vincent, on the other hand, has been sen-
tenced for fraud, embezzlement and murder; he is portrayed as a smart, rational and
reserved character. An unlikely pair, the two are united by their wish to take revenge
on a common enemy – Leo’s former boss, Harvey, who has framed Leo and killed
Vincent’s brother. Determined to make their escape, the two have to work together.
Cooperation is required for most tasks, be it opening doors, subduing guards, hoisting
each other up, or creating distractions so the other character can accomplish a task.
Teamwork is at the core of A Way Out’s game design, realized as cooperative activity
for two players presented in a split screen. This configuration supports the narrative of
the two characters being dependent on each other, while placing the respective players
in the position of their avatars [5]. As Nitsche [6] argues, players understand the video
game space and their movement therein by ways of narrative comprehension: a form of
understanding of the events they trigger and encounter. To bolster the game’s syner-
getic disposition, both characters’ story unfolds simultaneously on the screen, allowing
each player to see what the other one is experiencing. In certain situations, the screens
merge, marking crucial points of interaction and shared experience. Once the two char-
acters have escaped, action scenes are intercut with slower narrative segments, reveal-
ing the gradual development of friendship. Throughout the game players encounter

3
dichotomic choices, usually juxtaposing Vincent’s careful approach with Leo’s brutish
one. The narrative comes to a climax as both players finally confront and kill Harvey
at his hideout in Mexico. Having fulfilled their mission, they return to the US, but in
typical Hollywood fashion, the game has a twist up its sleeve. Vincent turns out to be
an undercover cop whose sole reason for working with Leo is to get to Harvey. Feeling
betrayed, Leo takes Vincent hostage, leading to a series of actions that culminate in a
final showdown. Here, the game’s dominant mode changes from one of cooperation
(where players must help each other in order to proceed) to one of confrontation (where
players can and, in the end, must harm each other in order to propel the narrative for-
ward and achieve the game’s objectives), mirroring the narrative of the betrayal and the
resulting broken bond.
That is not to say that confrontation is absent beforehand: in fact, the game constantly
plays off of the dialectic between competition and collaboration to increase the bond
between players. From the start of the game, small, individual mini-games can be un-
dertaken, such as doing pull-ups in prison, the results of which can be compared to the
other player’s tally. However, in the game’s closing moments, narrative and gameplay
shift dramatically from a mode that is mainly co-operative to one that is solely compet-
itive. In the last interactive moment, players fight to reach a single weapon. Similar to
earlier friendly competitions, the player who presses X faster gets to the weapon first.
Once one of the characters reaches the gun, the game cuts to the perspective of that
character within a slow-motion scene. The scene produces a radical shift in what, fol-
lowing Genette [7], we might call ludic mood: the way the designer allows events to
unfold. We might say this is where the second betrayal occurs – this time, of the contract
between the game and its players [8]. The narrative only advances when the trigger is
pressed long enough so that the character in control raises his arm, aims for the other
character’s chest and then pulls the trigger. This is the only way to advance the narrative
of this scene; any choice of non-participation just causes the game to pause.
3 Theoretical Background
3.1 Harmony, Dissonance and Narrative Paradox
< don’t make me shoot him > ↑ I DON’T WANNA SHOOT HIM ↑
The above line is a Conversation Analysis excerpt from a Let’s Play video of A Way
Out [9]. The player, shouting at the game as he struggles to avoid killing his accomplice
and co-player, is typical for the response of players in the final section of the game,
having cooperated with the other player until this point. How can we understand these
strong, disapproving responses? We might be tempted to call this a matter of ‘lu-
donarrative dissonance’, which is typically understood as the result of a disconnect be-
tween game mechanics and narrative. The concept was first offered in a rather terse
blog post by Clint Hocking [1], who defines the concept loosely through the example
of Bioshock. In the game, the “ludic contract” between player and game, – an objectivist
morality that regards everything around them as a means to their own end – is in conflict

4
with its “narrative contract”, in which the player only progresses when they help an
anonymous interlocutor over the radio. Bioshock thus contains a “dissonance between
what it is about as a game, and what it is about as a story” [1]. The concept has proven
fruitful outside academic circles and is frequently applied in games journalism and pub-
lic discourse.
In the case of A Way Out, if we take the game’s dominant ludic mode to be that of
collaboration, then, the narrative component seems incongruent, and – in the final scene
– violates the “narrative contract” through the removal of agency, eliciting strong neg-
ative reactions.
We argue, however, that “ludonarrative dissonance” does not adequately describe
this phenomenon, in part due to the unstable ontological premises of this term. While
the term ludonarrative dissonance seems to evoke diachronicity – a tension or clash
resulting from the combination of two disharmonious or unsuitable elements during a
musical piece – the term very often describes its problem in holistic terms. A game
either is ludonarratively dissonant, or it is not. The qualitative statement Hocking makes
about his case study (“Bioshock seems to suffer from a powerful dissonance between
what it is about as a game, and what it is about as a story”) demonstrates this pars-pro-
toto well. Of course, Hocking’s example is not entirely ludonarratively dissonant: he
notes that during the initial hours of his playthrough, what he experienced was more
like what Pynenburg [10] calls “ludonarrative harmony”, as he writes: “the game liter-
ally made me feel a cold detachment from the fate of the Little Sisters, who I assumed
could not be saved.” Harmony and dissonance seem, much like in music theory, co-
dependent terms. Second, when understood in a particular way, ludonarrative disso-
nance may constitute what Gilbert Ryle [11] designated as a category-mistake, in which
things belonging to a particular category are presented as belonging to a different one.
We argue that the “ludic contract” Hocking describes is, upon closer inspection, another
narrative contract: the world-making that players engage in through the game’s me-
chanics is a cognitive achievement, and one in which their identification with “Randian
objectivism” is a wholly narrative model, based on connected events. This is no attempt
to drag up the problematic dichotomy of narratology versus ludology; instead, we mean
this insofar as Jerome Bruner noted that “[w]e seem to have no other way of describing
‘lived time’ save in the form of a narrative.” [12] of course, the player does engage with
mechanics that are not (overtly) narrative in terms of their practical involvement. A
reflection on what those mechanics mean, however, undoubtedly needs to consider the
narrative aspect. This means we need to look more closely into the tension, between
narrative aspects in a game, instead of stopping short at the definition of “mechanics”.
We will do so in the analytical section of this paper.
As mentioned, ludonarrative dissonance also implies ludonarrative harmony. Lu-
donarrative harmony refers to the successful syncing of both ludic and narrative aspects
to build a consistent, immersive experience. When achieved, this harmony results in an
internally consistent world that “feels right” [10]. As such, ludonarrative harmony
seems to describe the status of the well-functioning videogame, and its particularities
are interesting to analyze. Interestingly, the lead designer of A Way Out, Josef Fares,
has implemented cooperative gameplay before in Brothers: A Tale of Two Sons [13].
In Brothers, the game mechanics require players to use one gamepad stick for each of

5
the two brothers they are controlling, reflecting the cognitively challenging task of co-
operation. The older brother helps the younger one with different tasks – notably, car-
rying him on his back while swimming. Later in the game the older brother dies, and
the younger brother has to finish the journey alone. Players are left with one of the
gamepad sticks to interact. Yet at the game’s end, players are confronted with another
stream to cross, and the younger brother refuses to go beyond the shore. The problem
is overcome when players realize that the older brother is still there ‘in spirit’: both
gamepad sticks can still be used. This narrative twist, told through the gameplay gambit
of first taking away agency and then giving it back, has brought much praise to Fares
and his design team. Similarly, in A Way Out, Fares’ design fosters the bond between
the characters (and the players) through cooperative gameplay right up to the climax of
the game, where agency is taken away. In this case, however, it is never given back.
Instead, the narrative design imposes either of two versions of an emotional ending,
with one character dead. On an analytical level, the question is whether the observed
problem is an instance of ludonarrative dissonance: the incongruence between what
players assume based on what they have thus far experienced, and what the game de-
signer has planned as the conclusion for the story.
Cliff Makedonski draws several conclusions from the notion of ludonarrative disso-
nance, considering designers responsible solely for the disruption. Overcoming the
problem would either mean “focusing its efforts in a mostly linear direction”, so that
the player cannot make decisions that would put consistency at risk, or creating a pro-
verbial map that spans the entire territory, and “molding a world where anything could
happen” [3]. Of course, creating an IDN system that allows for such high levels of
agency shifts the conversation to whether the resulting experience would still be mean-
ingful for both player and the designer.
This brings us to the concept of narrative paradox [2, 14] which describes the inherent
tension between authorship and participation, in which the player asserts agency, the
freedom to take actions, while the game designer refuses to relinquish control of the
narrative for the purpose of ensuring what they believe is a satisfying structure [15].
In IDN, interactors are encouraged to actively create belief by performing or, in effect,
inhabiting a role [16]. This perspective assumes a fundamental distinction between
agency and dramatic structure. A narrative is taken as a carefully woven product the
storyteller crafts in order to create maximum impact. Herein lies the problem.
Costikyan [17] writes: “To the degree that you make a game more like a story – a con-
trolled, pre-determined experience, with events occurring as the author wishes – you
make it a less effective game. To the degree that you make a story more like a game –
with alternative paths and outcomes – you make it a less effective story.” However, this
view has been challenged by a range of authors, including Jennings [18], Murray [16],
Mateas [19] and Koenitz [20]. Thus, in order to be able to identify more precisely the
tension caused by the ending of A Way Out, we need a better view of the ludonarrative
relationship between players and designers. To do so, we need to return to the problem
of what ludonarrative refers to in the first place.

6
3.2 Ludonarrative hermeneutics
Adding to the confusion about ludonarrative dissonance is the fact that “ludonarrative”
itself “is variously understood as a structural quality of the video game artifact, an ex-
periential quality during the experience of a video game, or a high-level framework to
understand video games.” [21] In either case, as we mentioned in the last section, a
number of scholars emphasize the difference to linear narrative manifestations and
therefore work towards specific theories of video game narrative, seen as a variant of
interactive digital narrative (IDN).
As Montfort [22] and Koenitz [23] clarify, an IDN artefact is not itself a narrative,
it is an interactive computer program with the potential of instantiating narratives
through user interaction. Koenitz reflects this aspect in his SPP model [20, 23], by un-
derstanding IDN as comprised of System, Process, and Product. Through the process
of the player’s engagement with the interactive narrative system by choices and other
behavior - her performance (see also Knoller’s perspective on “userly performance”
[24]) - a concrete and personal narrative product is instantiated. Koenitz understands
the IDN system as a prototype for all potentially instantiated narratives and thus as a
protostory.
At any moment of reflecting on this narrative, it forms a story in the mind of the
player. This is the ludonarrative expression of the hermeneutic circle, in which a text
as a whole is established by reference to the individual parts and one's understanding
of each individual part by reference to the whole. Yet, in the case of interactive digital
narrative, the player also makes plans by evaluating the results of the interaction strat-
egy so far, by speculating about paths not taken and by considering the current potential
for interaction.
To illustrate how this system of ludonarrative hermeneutics works, we propose to
extend Koenitz’ SPP model [23] by adding a “hermeneutic strip”, incorporating two
interconnected loops of narrative interpretation processes (see Figure 1).
Figure 1: Extension of Koenitz’ SPP model [23] through implementation of a
double-hermeneutic circle (‘hermeneutic strip’)

7
The hermeneutic strip illustrates the player’s narrative meaning-making process that
happens a) when players interact with the system and b) when they interpret the instan-
tiated narrative at any point of the experience. The latter can be understood as the classic
hermeneutic circle, and together the two loops result in an interdependent double-her-
meneutic circle of the IDN experience (cf. [25]).
In the ‘upper hermeneutic circle’ players are reflecting what the system allows them
to do and what narrative they could extract, also in the sense of narrative game mechan-
ics [26]. Within the ‘bottom hermeneutic circle’, then, players interpret the currently
instantiated narrative. While players interact with an IDN system, they are continuously
extracting information to understand past and present events and to plan their actions.
As we mentioned above, this meaning-making process can be understood from a
cognitive perspective on narrative. Following cognitive narratologists Herman [27] and
Bordwell [28], Marie-Laure Ryan [29] narrative meaning is a cognitive construct, or
mental image, built by the interpreter in response to the narrative construct (text in
narratological terms). Intuitively, we might understand narrative as located with a nar-
rative product like a printed book or a movie, however, the cognitive perspective
stresses the point that narrative resides within the human mind as a mental construct.
“... it does not take a representation proposed as narrative to trigger the cognitive
construct that constitutes narrativity” [29]. Through interaction with a game such as A
Way Out, players create mental narratives in an effort to make sense of the simulated
environment and constantly check if their assumptions are consistent. New information
leads to an updated projection, and conflicting information can lead to perceived disso-
nance. Thus, we understand IDN to have different narrative agents: on the one hand
what is conveyed, implied, intended by the designer on a system level, and on the other
hand what is performed, interpreted, assumed, expected, speculated and planned by the
player. For the present discussion, we focus on player reactions to an IDN system.
Through an analysis of these responses, we will now attempt to shed light on the po-
tential dissonance players experience with the ending scenes of A Way Out.
4 Method: Analysis of Let’s Play Videos
For our analysis of player responses to A Way Out’s narrative twist, and the way in
which ludonarrative hermeneutics can be demonstrated, we turn to Let's Play (LPs)
videos on the popular video platform Youtube. Let's Play is a style of videos document-
ing the playthrough of a video game, usually including commentary by the gamer. Glas
et al. [30] identified LPs’ potential for game archiving and exhibition purposes within
cultural heritage institutions. Extending this perspective, we use LP as samples for the
evaluation of user experiences. The first selective step consisted of watching a number
of popular Let’s Play videos (N = 40), showing up under the search query “A Way Out
Let’s Play”. We narrowed the selection to 20 videos for the conversational analysis
when the responses to the telling moment seemed not to yield any new insights.
For the purpose of this analysis we focus on two ‘telling moments’ at the end of the
game, when the narrative twist takes place. In the first scene Vincent reveals that he is
an undercover cop and Leo becomes aware of his betrayal. The second scene is the

8
showdown in which Leo and Vincent fight each other, leaving one of them dead. We
decided to exclude Let’s Play heavily edited videos as we were interested in a recording
situation that resembles an unedited think-aloud protocol. A smaller number of conver-
sations (6) were then transcribed using the Jefferson system, one of the founders of
conversation analysis (CA). [31, 32]. This method is interesting for several reasons.
Reeves et al. [33] note that investigations into video game play developed from ethno-
methodological and conversation analytic (EMCA) perspectives remain scarce. EMCA
approaches the need to engage both with play as social action and the ways in which it
is practically accomplished by players, “between players themselves, and between play-
ers and “the game” – as a moment-by-moment, sequentially organized activity” (ibid.)
They do, however, offer the opportunity to focus in detail on the embodied act of play
itself. Yet, EMCA typically attempts to work with what we could refer to as “natural-
istic data”: audio and video recording of activities that are, as much as possible, occur-
ring in their original settings.
While the kinds of performative behavior that Let’s Players demonstrate are likely
not the same kinds of behavior found in everyday – and unrecorded – home scenarios,
we contend that the performativity here allows us more insight, not less, into the ways
in which players respond to the tensions arising out of the particular combination of
gameplay and narrative. Youtubers “act out” their responses, often using what Ytreberg
[34], following Goffman [35], refers to as a performative aspect of mass media, namely
hyper-ritualizations and dramatic scriptings. The former involves a concentration and
intensification of certain traits of unmediated behavior and interaction. The latter,
Goffman notes, refer to “all strips of depicted personal experience made available for
vicarious participation to an audience or readership” [35]. It is this dual nature of the
Youtube performance that makes it indicative of the way players “ought” to respond
to, in this case, a game. Goffman writes that “[the] deepest significance [of dramatic
scriptings] is that they provide a mock-up of everyday life, a put-together script of un-
scripted social doings, and thus are a source of broad hints concerning the structure of
this domain.” We view the videos we analyze, then, as “over-performed” interactive
narrative user experiences, telling of the ways in which games produce affective re-
sponses in their players.
To transcribe the findings of the LP sessions we turn to Conversation Analysis (CA):
cf. [31, 32], which focuses on “the shape and form of the ways in which contributions
to interaction form a connected series of actions”, while attempting to “discover and
describe the architecture of this structure: the properties of the ways in which interac-
tion proceeds through activities produced through successive turns” [31]. In the process
of annotation, we made use of the "Jefferson system" after its developer, the late Gail
Jefferson. One of the additions we decided to make to this system is the inclusion of
physical behavior, which due to the video format includes important interactional cues.
We further made a distinction between physical player behavior (throwing the control-
ler away) and in-game character behavior (pointing the gun towards something), as the
latter is obviously controlled by the player but adds further contextual information.

9
5 Analysis
As the final confrontation between the characters commences, all of the players re-
mained highly motivated to continue, as this event at first seems in line with the gradual
escalation of the competitive elements that it has included from the beginning. Then,
after a long fight scene between the characters Leo and Vincent (both badly wounded)
stumble to a gun at the edge of the building. Still enthralled by the game’s ludonarrative
harmony, players shout and curse at each other as their movements bring them closer
to the gun. Then, as one player reaches the gun, they are confronted with the results:
the camera pans behind the winning character, and all they can do is to press the trigger.
In all of the analyzed LPs in which Vincent was the victor, players did not want to
shoot the other, trying out a myriad of ways to avoid the outcome. Some temporarily
put down the controller in refusal [9], others started playing meta-games with the mech-
anism of pointing the gun at the player [36]; yet others simply waited for a long time
[37]. A closer look at a particular response of the players in The Let’s Play channel T&J
Nexus [38] demonstrates the disbelief of players as the game forces their hand and ne-
gates the ludonarrative harmony of comradery that has characterized the majority of the
playthrough.
P Vincent: wha- aim, i don’t wanna shoot. (.5)
P Leo: oh::: my god (.4)
P Vincent: i don’t gonna i’m not gonna shoot (.) are you kidding me? {switches be-
tween pointing and pointing away at Leo} (.3) do I have to? (.3) no, what
if i don’t. (.6) i don’t wanna d(h)o this (.h) ↑ it’s gonna f(h)orce me ↑
P Leo: (h) (.h) you don’t have an option. (.3)
P Vincent: (h) i don’t wanna do this (.3) I really don’t wanna do this (.2)
P Vincent [holy shit]
P Leo: [ahh:: this is so] (.hh)
P Vincent: {shoots} ohh:: ↓ my god. ↓ why? (..) ohh:: come o::n
P Leo: (.hhh) (hhh)
P Vincent: fuck this ga::me. No:: i didn’t w(h)ant it to e(h)nd like this
What we see here is a familiar “mapping” of players onto their characters, despite
the fact they have played a round character for the entirety of the game – putting further
tension on Troisi’s distinction between “objective” and “subjective” narrative positions
[39]. Players refer to their characters in first person yet simultaneously reflect on their
own position as players. Additionally, they use the subject pronoun “it” to refer to the
agency of the game itself, clearly no longer a “game” in the sense of “free movement
within a more rigid structure” as Salen and Zimmerman [40] put it, but a system of
predetermined choices. We also see players rationalizing the choices made by their
characters and, instead of resisting, “playing along” and roleplaying as the idiosyncratic
character they are playing. One player who gets to shoot while playing the cop, Vincent,
rationalizes by saying “Well, it’s my job” and “there’s no honor among thieves” to his
co-player. At the same time, players who control the character who shoots the other
discursively distance themselves from the act they engage in. One player, while having

10
used first and second person pronouns throughout the scene, slips into third person di-
rectly after his avatar shot the other.
P Vincent: ↑ why did he ↑ shoot im? (.) i mean i guess he-
P Leo: leo definitely was gonna kill him
P Vincent: ↑ yeah i mean ↑ he (.) had (..) all the right to
The player demonstrates what Goffman calls ‘role distancing’ [35] which pertains
to the act of presenting one’s 'self' as being removed or at a distance from the role one
is being required to play. Role distancing is one strategy which allows the individual to
play the role but to resist it – for example, by keeping your eyes open when asked to
pray or say grace, you communicate to the group by role distancing, that you are making
no commitment to the role. As such, people get to deny "the virtual self that is implied
in the role for the allocating performers" [41]. While role distancing is a well-known
trait from everyday life, it has as of yet not been connected to the complex performances
of streamed video game play. Let’s Players have to negotiate the space between them-
selves as performers vis-a-vis both the game (which fixes them into certain roles) and
their potential audience (which may expect certain roles to be played). In this particular
case, their linguistic behavior signals a dissonance not between their play and the nar-
rative, but between their prior investment in the totality of their prior role and the new
one they are forced to take. Taking stock of such conversations, where Glas [42] argues
that “the aim with LP videos is not to create stories”, we would strongly disagree. More
than “annotating their game-play in real time” [43], Let’s Players engage in all kinds of
narrative life writing acts [44]. Moreover, it is the negotiation of that self that occurs in
the moment that players are confronted with a forced action they do not subscribe to.
In sum, the use of role distancing in the broader narrative construction of “demon-
strative play” that the Let’s Play genre consists of, shows the subtle ways in which
players are claiming authorship over the uncontrollable experience that the game offers.
It also demonstrates that the dissonance, here, is not primarily one between mechanics
and narrative, even if we would agree upon such a binary: there is not necessarily a
tension between the narrative of the betrayed partner and the mechanical consequences
that flow from that. Instead, the dissonance here arises between what we could describe
as the hermeneutics of the designer and that of the player.
6 Discussion
The term 'ludonarrative' and related vocabulary ('video game narrative', 'narrative-fo-
cused game' 'interactive story') has been used to describe the phenomenon of narrative
in interactive media. However, clear and generally accepted definitions of what this
term actually entails are elusive. Accordingly, Koenitz states that no generally accepted
definition of video game narrative exists as of yet [21]. He identifies a range of positions
(e.g. the experience of a video game, its content and or an analytical perspective), which
clearly demonstrate the vagueness of existing terminology. In this paper we have seen
how current definitions of interactive narrative are not sufficient to describe the

11
problems players have with the narrative of a particular game, and its narratively inter-
preted mechanics.
Again: the fact that we cannot conceptualize gameplay without a resort to narrative
is not an attempt to police the borders of narratology and ludology. What it allows us
to see is that there is a friction between the player’s unfolding ideas about the characters
they inhabit, and the designer’s intended experience for that player. In this respect, lu-
donarrative dissonance describes the consequences of deeper-seated systemic tensions.
In order to conceptualize this, we have proposed a different categorization by incorpo-
rating a hermeneutic approach to Koenitz’ SPP model [23]. This conflict we identify
lies beyond the moniker of ludonarrative dissonance: it is a disturbance in the interac-
tive narrative experience caused by the unbridgeable distance between players’ expec-
tations (the assumptions they develop as they experience the narrative through game-
play, as shown in the interaction and interpretation circle in figure 1) and what is im-
posed by the designer, with the intention to create an engaging, immersive, and mean-
ingful experience.
This view is supported by our analysis of Let’s Play sessions as we observed players
to be confused and in some cases even “rebelling” against the removal of agency, es-
pecially when the narrative that ensued subverted their conception of its integral unity.
The hermeneutic strip we have introduced seemed to break, as the particular instantia-
tion of the story turned out to differ (due to player agency) from the protostory players
had imagined. This, we argue, demonstrates how players develop a sense for the level
of granted agency, for instance when choosing to follow Leo’s or Vincent’s approach
to the issues they encounter throughout the game.
As discussed above, IDN is a shared meaning-making activity between designer and
audience. In the last interactive scene of A Way Out, shared meaning-making turns into
an imposed narrative by the author. Here, the creator changes from the role of an inter-
active narrative designer, offering agency to the players, to the role of the traditional
author who wants to convey a specific narrative meaning. Players revolt, when encoun-
tering the withdrawal of their agency as the co-creative process turns into an authored
one, subverting their own co-created hermeneutical ideas about the ‘protostory’ (the set
of all possible narratives ) and the current concrete instantiation (the product created so
far). This issue is related to the concept of design conventions as the conventional un-
derstanding of design methods by the audience [45]. A Way Out, in this context, applies
the IDN design convention of “meaningful collaborative choice” throughout most of
the game before finally frustrating audience’s expectations when a novel, limiting in-
teraction design is imposed.
This confusion leads to perceived dissonance, a rupture in the experience, leading to
a break in the ludonarrative process. We argue that players, as a result of this interrup-
tion, instead of being invested in further narrative progression, started focusing on the
change of their interaction experience – a move towards a meta-communicational level.
As we know, in literature the “whole”, as part of the hermeneutic movement, refers to
the entirety of the story. When it comes to IDN, that whole is more than that: as the
protostory it also refers to anything the player “could have done” as well. This is, in
fact, what Murray [16] considers part of the transformative aspect of games: knowing

12
that things could have been different is constitutive for the experience, as it makes one’s
decisions personally meaningful.
7 Conclusion
When creating interactive digital narrative systems, designers need to find a balance
between player agency and a dramatic structure to offer a consistent, meaningful nar-
rative. The resulting set of ludonarrative experiences is an essential part of narrative-
rich games that cannot be ignored. Ludonarrative dissonance can be best understood as
a temporal dissonance between the player’s interpretation of the ludonarrative process
and the interpretation of the overarching narrative (the instantiated narrative product
created so far), and a function of the player’s iterative hermeneutic oscillation.
“Ludonarrative dissonance” as an umbrella term does not explain the actual disso-
nances that occur and which can easily be misinterpreted as creating a false dichotomy
between ludic and narrative elements. Our extended SPP model is a first step towards
gaining a better understanding of such dissonant experiences. The challenge that lies
ahead involves identifying and explaining the actual dissonances that can occur as the
construction of narrative is tightly connected to our meaning-making. To understand
the complex relationship between interaction and narrative it is crucial to include a
cognitive perspective in the analysis. This destabilizes the dichotomy between “ludo”
and “narrative”, as narrative interpretation can be situated on several dimensions of the
ludonarrative experience.
Future work might focus on expanding the IDN model to include the different nar-
rative dimensions in between which dissonance can occur. Researchers might also in-
vestigate into temporal aspects and, for instance, the dissonance between different play-
through sessions (e.g. for systems that offer cross-session memory like Save The Date).
In terms of methodology, a further exploration of Let’s Play Videos as qualitative sam-
ples for the evaluation of the interactive narrative user experience is warranted – the
fact that the captions of these videos can be computationally scraped is noteworthy.
A Way Out creates a collaborative experience, while highlighting the friction be-
tween collaboration and competition by which many friendships are characterized.
Each player is capable of relating this “social model” to their own life. Interactive nar-
ratives, in other words, are significantly more flexible in achieving what Barthes called
the writerly text, in which the reader is located as a site of the production of meaning,
and for which the goal is to make the reader no longer a mere consumer, but a producer
of the text [46]. Interactive Digital Narratives are not played for escapist motives alone
but also for the transformative aspect: “The right stories can open our hearts and change
who we are. [...] Enacted events have a transformative power that exceeds both narrated
and conventionally dramatized events because we assimilate them as personal experi-
ences.” [47]

13
References
1. Hocking, C.: Ludonarrative Dissonance in Bioshock: The problem of what game is
about, http://clicknothing.typepad.com/click_nothing/2007/10/ludonarrative-d.html.
(2007)
2. Aylett, R.: Emergent narrative, social immersion and “storification.” Presented at the
1st International Workshop on Narrative Interaction for Learning Environments, Edin-
burgh. (2000).
3. Makedonski, B.: Ludonarrative Dissonance: The roadblock to realism,
https://www.destructoid.com/ludonarrative-dissonance-the-roadblock-to-realism-
235197.phtml. (2012)
4. Hazelight Studios: A Way Out, (2013).
5. Klevjer, R.: What is the avatar? Fiction and embodiment in avatar-based singleplayer
computer games. (2006).
6. Nitsche, M.: Video Game Spaces. MIT Press, Cambridge, MA (2008).
7. Genette, G.: Narrative Discourse, an Essay in Method. Cornell UP, Ithaca: NY (1980).
8. Roth, C., Vermeulen, I.: Breaching Interactive Storytelling’s Implicit Agreement: A
Content Analysis of Fac̨ade User Behaviors. In: Koenitz, H., Ferri, G., Haar, M.,
Sezen, D., Sezen, T.I., and C̨atak, G. (eds.) Interactive storytelling : 6th International
Conference, ICIDS 2013, Istanbul, Turkey, November 6-9, 2013, Proceedings. pp.
168–173. Springer International Publishing, Cham (2013).
9. Lets Play Pals: A Way Out - It Doesn’t Have To Be Like This (End),
https://www.youtube.com/watch?v=4Nvi3bguqoE.
10. Pynenburg, T.: Games Worth a Thousand Words: Critical Approaches and Ludonarra-
tive Harmony in Interactive Narratives, http://scholars.unh.edu/cgi/viewcontent.cgi?ar-
ticle=1069&context=honors.
11. Ryle, G.: The Concept of Mind. University of Chicago Press, Chicago. Il (2003).
12. Bruner, J.: Life as narrative. Social research. (1987).
13. Starbreeze Studios: Brothers: A Tale of Two Sons, (2013).
14. Louchart, S., Aylett, R.: Solving the Narrative Paradox in VEs – Lessons from RPGs.
In: Technologies for Interactive Digital Storytelling and Entertainment. pp. 244–248.
Springer Berlin Heidelberg, Berlin, Heidelberg (2003).
15. Louchart, S., Aylett, R.: Narrative theory and emergent interactive narrative. Interna-
tional Journal of Continuing Engineering Education and Life Long Learning. 14,
(2004).
16. Murray, J.H.: Hamlet on the Holodeck: the Future of Narrative in Cyberspace. Free
Press, New York (1997).
17. Costikyan, G.: Where stories end and games begin, http://www.cos-
tik.com/gamnstry.html. (2001)
18. Jennings, P.: Narrative Structures for New Media. Leonardo. 29, 345–350 (1996).
19. Mateas, M.: A Preliminary Poetics for Interactive Drama and Games. Digital Creativ-
ity. 12, 140–152 (2001).
20. Koenitz, H.: Towards a Theoretical Framework for Interactive Digital Narrative. In:
Aylett, R., Lim, M.Y., and Louchart, S. (eds.) Interactive Storytelling: Third Joint
Conference on Interactive Digital Storytelling. pp. 176–185. Springer (2010).
21. Koenitz, H.: Narrative in Video Games. In: Encyclopedia of Computer Graphics and
Games. pp. 1–9. Springer International Publishing, Cham (2018).
22. Montfort, N.: Twisty Little Passages. MIT Press (2005).
23. Koenitz, H.: Towards a Specific Theory of Interactive Digital Narrative. In: Koenitz,
H., Ferri, G., Haahr, M., Sezen, D., and Sezen, T.I. (eds.) Interactive Digital Narrative.
pp. 91–105. Routledge, New York (2015).

14
24. Knoller, N.: The expressive space of IDS-as-Art. In: Oyarzun, D., Peinado, F., Young,
R.M., Elizalde, A., and Méndez, G. (eds.) Interactive Storytelling: 5th International
Conference, ICIDS 2012, San Sebastián, Spain, November 12-15, 2012. Proceedings.
Springer Berlin Heidelberg, Berlin, Heidelberg (2012).
25. Karhulahti, V.-M.: Double fine adventure and the double hermeneutic videogame. Pre-
sented at the Foundations of Digital Games, New York, New York, USA (2012).
26. Dubbelman, T.: Narrative Game Mechanics. In: Nack, F. and Gordon, A.S. (eds.) In-
teractive Storytelling. pp. 39–50. Springer International Publishing (2016).
27. Herman, D.: Story Logic. U of Nebraska Press, Lincoln, NE (2002).
28. Bordwell, D.: Poetics of cinema. New York : Routledge (2007).
29. Ryan, M.-L.: Narrative across Media: The Languages of Storytelling. University of
Nebraska Press, Lincoln (2004).
30. Glas, R., de Vos, J., van Vught, J., Zijlstra, H.: Playing the Archive. In: Mol, A.A.A.,
Ariese-Vandemeulebroucke, C.E., Boom, K.H.J., and Politopoulos, A. (eds.) The In-
teractive Past: Archaeology, Heritage & Video Games. Sidestone Press, Leiden
(2017).
31. Wooffitt, R.: Conversation Analysis and Discourse Analysis. SAGE (2005).
32. Liddicoat, A.J.: An Introduction to Conversation Analysis. Bloomsbury Publishing
(2011).
33. Reeves, S., Greiffenhagen, C., Laurier, E.: Video Gaming as Practical Accomplish-
ment: Ethnomethodology, Conversation Analysis, and Play. Top Cogn Sci. 9, 308–342
(2016).
34. Ytreberg, E.: Erving Goffman as a theorist of the mass media. Critical Studies in Me-
dia Communication. 19, 481–497 (2010).
35. Goffman, E.: Frame analysis: An essay on the organization of experience. Northeast-
ern University Press, Boston, MA (1974).
36. GameGrumps: A Way Out: Finale - PART 20,
https://www.youtube.com/watch?v=FWFMPkU0hKc.
37. #TheEscapeBros: Ending - a Way Out [Co-Op] - Part 11,
https://www.youtube.com/watch?v=WUfru2Gk92k.
38. T&J Nexus: A Way Out, Vincent Ending (Spoilers Duh) Reaction ft Bes,
https://www.youtube.com/watch?v=yXYLd0J2Lh0.
39. Troisi, A.: Interactive Narrative Design in Mass Effect 2. Vancouver, Canada (2010).
40. Salen, K., Zimmerman, E.: Rules of Play. MIT Press, Cambridge, MAS (2004).
41. Goffman, E.: Encounters: Two Studies in the Sociology of Interaction. Allen Lane
(1972).
42. Glas, R.: Vicarious play: Engaging the viewer in Let’s Play videos. Empedocles: Euro-
pean Journal for the Philosophy of Communication. 5, 81–86 (2015).
43. Newman, J.: Videogames. Routledge (2013).
44. Eakin, P.J.: How Our Lives Become Stories: Making Selves. Cornell University Press
(1999).
45. Roth, C., Koenitz, H.: Towards Creating a Body of Evidence-based Interactive Digital
Narrative Design Knowledge: Approaches and Challenges. Presented at the Alt-
MM'17, Mountain View, CA August 21 (2017).
46. Barthes, R.: S/Z. Hill and Wang, New York, NY (1975).
47. Murray, J.H.: Hamlet on the Holodeck. The Free Press, New York (2016).