Macquarie University PURE Research
Management System
This is the Accepted Manuscript version of the following article:
Staines, D., Consalvo, M., Strangeby, A., Pedraça, S. (2019) State of play:
video games and moral engagement, Journal of Gaming & Virtual Worlds,
Vol. 11, No. 3, pp. 271-288.
which has been published in final form at:
https://doi.org/10.1386/jgvw.11.3.271_1
Copyright: Intellect Ltd. 2019.
State of Play: Videogames and Moral
Engagement
Abstract: In this article we examine three recent examples of ‘ethically notable games’ (Zagal,
2010) and highlight unusual or innovative design features for facilitating moral engagement.
Drawing on Sicart (2010, 2013) to frame our analysis, our goal is to highlight current trends in
ENG design and show how commercial games are moving beyond reductive “morality meters”
and treating moral choice with greater nuance, resulting – for the most part – in a more morally
engaging experience.
Keywords: Morality, videogames, virtue ethics, Firewatch, Deus Ex, Darkest Dungeon
Morality is a fundamental component of human experience, and art – from Greek drama through
to Saturday morning cartoons – is a powerful, pervasive means of disseminating and
interrogating moral principles and points-of-view (Carr, 2005, 2006). Videogames are no
exception to this rule and the history of ‘ethically notable games’ (Zagal, 2010) or ENGs is
almost as long as the history of gaming itself, with formative examples like Ultima IV: Quest of
the Avatar (Origin Systems, 1985) and Balance of Power (Crawford, 1985) appearing in the
early 1980s. Following the success of Star Wars: Knights of the Old Republic (Bioware, 2003), it
became something of a trend for videogames to feature ‘morality meter’ systems in which the
player’s moral decisions contribute to an overall ‘morality score’ for their character. Subject to
criticism from academics (Melenson, 2010), critics (Birch, 2014), and fans (Aristov, 2012),
morality meters have mostly fallen out of favour and there is now a ‘growing corpus’ of ENGs
that treat morality with greater care and nuance (Heron & Belford, 2014a). In this article we will
examine three such ENGs from 2016, focusing in each case on unusual or innovative design
features for facilitating moral engagement.
Literature Review and Theory
1
In their paper coining the phrase, Zagal (2010, pp. 1-2) identifies four key perspectives
for understanding ethically notable games. First, there is the ethical significance of the object:
should certain morally repugnant games (e.g. rape simulators) exist at all, and is it (un)ethical to
own and/or distribute these titles? Second, what does it mean to create games ethically – what
moral and professional standards should developers, publishers, and distributors abide by and
how should these be instituted across the industry? A third perspective concerns the ethics of
play and interrogates the meaning and application of concepts like rules, norms, and cheating.
Finally, there the ‘ethics of actions in games as defined by the games themselves’ (ibid.) – what
Zagal calls the game’s ethical or ideological framework. In this paper we are exclusively
concerned with this latter perspective, looking at how game design facilitates ethically
significant gameplay – defined here to mean in-game actions that provoke moral reflection. (For
more on what we mean by ‘moral reflection’ in this context, see the discussion of Sicart below.)
In recent years interest in ENGs has exploded and there are now multiple frameworks for
classifying, analysing, and designing ENGs. These include Flanagan et al.’s (2007) Values at
Play project, which explores how developers embed moral, social, and political values in their
games; Stevenson’s (2010) three-tiered scheme for categorizing ENGs; Schrier’s (2015) EPIC
framework for using videogames in ethics education; and Ryan et al.’s (2016) ‘four lenses’ for
ENG design. In addition to these, there is a growing corpus of empirical studies on ENGs. In one
such study, Weaver and Lewis (2012) examine ‘how players make moral choices in videogames
and what effects these choices have on emotional responses to games’ and find that players of
ENGs make in-game moral choices that are largely in keeping with their real-world moral
intuitions. This is consistent with Consalvo et al.’s (2016) qualitative study of player attitudes,
which finds that ‘rehearsing [an] ethos’ – playing a better version of their own moral selves,
basically – is essential to many gamers’ enjoyment of ENGs. This theme is reinforced by a
similar study (Lange, 2014) in which players overwhelmingly express a preference for playing
‘good’ characters in ethically notable role-playing games (RPGs).
A leader in this field is games scholar Miguel Sicart (2010, 2011, 2013), who outlines a
model of ENG design based on the ‘Levels of Abstraction’ concept within information ethics.
The model posits that players interact with videogames at two levels of abstraction: as
procedural/mechanical systems to be mastered, and as semantic objects with cultural and ethical
meaning. For example: from a purely mechanical perspective, Call of Duty: Modern Warfare II’s
2
(Infinity Ward, 2009) infamous ‘No Russian’ level is not terribly different from any other level
in the game. The goal, as always, is to shoot people: the difference is that in this level the people
are unarmed civilians at an airport. This makes us uncomfortable: there’s dissonance between the
game’s procedural goals (shoot everyone) and their broader ethical and cultural implications,
resulting in what Sicart calls ethical cognitive friction – a ‘contradiction between what to do in
terms of gameplay, and the meaning and impact of those actions, both within the gameworld and
in a larger cultural setting’ (2010, pp. 6-7). The key to designing ENGs, Sicart argues, is to focus
on this dissonance, to provoke and exploit it and thereby compel the player to consider the moral
significance of the game’s procedural and semantic layers.
Crucially, Sicart recognises that ethical cognitive friction is contingent on a player
motivated to think about the moral significance of their in-game choices. Even the most morally
sophisticated ENG can be played instrumentally, as a series of ludic challenges devoid of ethical
resonance. Sicart calls this kind of play ‘reactive’ and contrasts it with the ‘reflective’ play of
someone who actively thinks about their choices and perceives dissonance when it appears
(Sicart, 2010, pp. 6–8). One of the defining goals of ENG design is to encourage players to adopt
a reflective stance – to promote what we refer to in this article as ‘moral reflection’ or ‘moral
engagement’. A morally engaged player is one who considers the moral significance of their ingame choices, who does not approach gameplay from a purely instrumental perspective but
attempts to understand the rules and assumptions that constitute a game’s ethical framework and
considers same in light of the real world ethical frameworks to which we are all, to some degree,
subject. To this end, Sicart recommends an approach that emphasises and exploits moral
uncertainty, restricting the player’s access to information and systems that disincentivise
reflection. Later we will see how design elements intended to support this approach – most
notably the concept of ‘aggregate of choices’ (2010, p. 10) – have surfaced in recent ENGs to
positive effect.
We highlight Sicart’s work particularly because it informs much of the analysis in this
article. The games we examine – Firewatch, Deus Ex: Mankind Divided, and Darkest Dungeon –
have been selected because they exhibit design features identified as either conducive or
detrimental to moral engagement: i.e. reflective, ethical play. All released in 2016, these games
provide an illuminating glimpse of current trends in ENG design while giving us an opportunity
to demonstrate the utility of Sicart’s approach for analysis and critique of moral content.
3
Additionally, because Sicart’s work is broadly compatible with the insights of other scholars
working in the field, we will draw on this work – most notably Nay and Zagal (2017) – to
supplement our analysis. This constitutes the secondary goal of our article, which is to highlight
concepts from the ENG literature that facilitate fruitful analysis and further discussion.
Of course, many old and new games could be productively analysed in this way and we
acknowledge that the titles featured here are not necessarily the best or most popular examples of
ethically notable game design. While it’s true that titles like Papers, Please, The Walking Dead,
and Spec Ops: The Line feature highly sophisticated ethical gameplay, these games have already
received a great deal of academic analysis (e.g. Formosa, Ryan, & Staines, 2016; Smelthurt &
Craps, 2015; Heron & Belford, 2014b) that we are not interested in recapitulating here. For this
paper we were particularly interested in spotlighting titles that were a) recently released, thereby
indicating emerging trends in ENG design, b) popular and widely-reviewed, and c) heretofore
absent in the ENG literature.
The analyses in this paper were conducted in three stages. First, three authors (Samia,
Adam, and Mia) played one game each while noting ethically significant scenarios and design
features warranting in-depth analysis. These notes were then handed to our project convenor
(Dan) who played all three games, paying particular attention to noteworthy commonalities and
discrepancies between the three in terms of their ethical design. In the second stage, the authors
convened and categorised the games in terms of applicable analytic frameworks, producing three
draft papers that were gradually refined and amalgamated by the project convenor. Finally, the
authors canvassed YouTube, Reddit, and other major internet forums, supplementing our own
experiences with insights from developers, reviewers, streamers, and ‘regular’ gamers. In so
doing our goal was to broaden the scope of our analyses, showing how the design features
discussed facilitate ethical engagement for players that aren’t academics writing a research
paper.
Micromorality and Virtue in Firewatch
Firewatch (Campo Santo, 2016) is a first-person adventure game developed by Campo
Santo, published by Panic, and released in February 2016 for PC, Linux, and Mac OS, with PS4
and Xbox One versions following in September. Likened to so-called ‘walking simulators’ like
4
Gone Home (The Fullbright Company, 2013) and The Stanley Parable (Galactic Cafe, 2013) the
game was reviewed positively by critics and players, receiving particular praise for its wellstructured, engaging narrative.
Firewatch begins with a short interactive text adventure that guides players through the
shared history of protagonist Henry and his wife, Julia. Beyond choosing between superficial
alternatives, such as what to name the couple’s dog, the player’s agency during this section is
limited, and the same basic story – boy meets girl, boy marries girl, girl is stricken with early
onset dementia, boy has existential crisis – is told no matter what the player chooses. The
prologue serves another purpose, however, in that it immediately confronts players with the
almost aggressive ordinariness of its protagonist. Henry is a chubby, balding, everyman facing a
painful, uncertain future. Like all of us, he is flawed – a fact that informs much of Firewatch’s
subsequent narrative and moral content.
Firewatch features a branching narrative driven by choices the player makes during
dialogue. In keeping with its everyman protagonist and slice-of-life fiction, the moral scenarios
depicted in the game are parochial, even mundane, compared to the world-ending dilemmas
featured in other games. To its credit, Firewatch does not deal with big ‘life and death’ or ‘save
the world’ choices, focusing instead on the ‘micromoral’ issues (Rest, Bebeau, & Thoma, 1999)
that permeate everyday life. This approach has multiple benefits, two of which are especially
relevant here. First, micromoral scenarios are by definition more relatable and familiar than big
picture, macromoral scenarios, making it easier for the player to identify with the protagonist.
Second, micromoral scenarios facilitate the aggregation of choices: an approach to ENG design
where one-off moral scenarios with immediate, profound consequences are replaced with many
smaller scenarios that accumulate significance as the game progresses (2013). For Sicart,
‘[t]he aggregation of choices is a better fit for designing ethical gameplay because it
places the players in a narrative or world context in which many choices are offered all
the time, and the consequence of each is not readily traceable to a particular choice’ (p.
105).
One of the chief virtues of the aggregate approach is that it shifts the player’s focus from
outcomes to decisions, representing morality as more than big problems waiting for optimal
5
solutions, but as an expression of one’s identity – as something that one does, day-to-day, in a
multitude of tiny but important ways. In this sense, Firewatch embodies an Aristotelian, virtueoriented approach to ENG design. Virtue ethics ‘evaluates actors based on the habits of their
actions, and the temperament motivating them rather than the outcomes of those actions’ (Nay &
Zagal, 2017, p. 2). For virtue ethicists, choices matter chiefly because of how they shape our
character, and that is also true of Firewatch.
When Firewatch begins, Henry has taken a new job as a fire warden in a national park
while Julia, stricken with dementia, is relocated to a care facility in Australia. Soon after starting
his first day Henry finds a walkie-talkie and is contacted by Delilah, his supervisor, who asks
him to discover the source of the fireworks going off illegally above a nearby lake. Following a
trail of clues, the player eventually discovers that two skinny-dipping young women are the
culprits. Drunk and belligerent, these women, who can be seen only as distant silhouettes, berate
and abuse Henry for disturbing them. How Henry responds is left up to the player. Should he be
a professional – turn off the nearby stereo, pick up the beer cans, let the insults roll of his back?
Or is Henry the kind of guy who takes it personal, the kind of guy who might abuse his authority
a little and ‘accidentally’ drop some little snot’s stereo in the lake? Both choices are recognized
by the game as legitimate, but – significantly – neither impacts the way the rest of the narrative
plays out. Consequences aren’t the point; the choice itself speaks volumes.
Another illuminating example of Firewatch’s virtue-oriented micromoral scenarios
occurs about half-way through the game when Henry decides, without the player’s input, to take
off his wedding ring and leave it on his desk, a deliberate and significant action in the context of
his evolving relationship with Delilah. From talking to her over the radio, it’s clear Delilah’s a
friendly, expansive, and funny woman, and it’s therefore natural that Henry (and the player) may
begin to develop feelings for her. At times Delilah, whose Biblically symbolic name is surely not
accidental, seems to flirt with Henry, hinting strongly at the possibility that something more
could develop between them. So when Henry takes off his wedding band, a physical
manifestation of his moral obligation to his dying wife, the significance is obvious. However,
what’s particularly compelling about this situation is that the player may, without prompting,
have Henry pick up the ring and put it back on his finger.
6
Figure 1. Henry’s wedding ring. Source: Firewatch (2016, Campo Santo)
In addition to setting up a subtle and profound micromoral choice, Henry removing his
wedding ring is a compelling example of what Sicart refers to as a ‘slow technology’ approach to
ethical game design (Sicart, 2013, p. 73). This approach advocates limiting the player’s agency
in ways that promote a ‘slower’ more contemplative mode of engagement. In this mode, the
player’s attention can be drawn to questions that arise from frictions within and between the
game’s semantic and procedural layers (ibid., p.76). In the present example, the player’s agency
is temporarily restricted when Henry independently removes his wedding band, prompting the
player to consider the significance of this act – and what it says about Henry’s character – in the
context of the evolving story. Having drawn the player’s attention to the ‘question’ in this way,
the game then gives the player agency to ‘answer’ it by either putting the ring back on or
ignoring it. Notably, the game does not explicitly indicate that Henry can put the ring back on,
and it’s entirely possible that less attentive players may miss the opportunity altogether. When a
developer refrains from drawing attention to a choice in this fashion, it often indicates the choice
is insignificant or unnecessary. But in this case, the opposite is true: by not drawing attention to
the possibility of replacing Henry’s wedding band, Campo Santo emphasises that this is not a
7
ludic or narrative choice, but a personal one that reflects the nature of Henry’s character. It is
significant, in other words, partly because it can be overlooked and because the game does not
change as a result of making it. Indeed, the only thing that changes is Henry – and the player’s
view of him.
Delilah, meanwhile, is noteworthy in part because she is resistant to change, and because
she exercises an unusual (for videogame NPCs) degree of independent moral agency. When it
emerges later in the game that the young women from the lake have vanished, Delilah asks
Henry if she ought to tell the police about the fireworks. Irrespective of how the player responds,
Delilah decides to keep the information to herself, later telling Henry that she wanted to ‘save
[them] the trouble’ of dealing with the cops. This is consistent with what we know about her
character, a woman
who has worked the job for nearly a decade and been seemingly unchanged by her
experience. She's not growing, and the fact that Henry and the player are still not enough
to change her mind in many situations is refreshing in a medium where you can often
achieve impossible persuasion simply by having high enough stats. (White, 2016)
Compare to a game like Mass Effect (Bioware, 2007), where feats of ‘impossible persuasion’ are
commonplace and achieved by investing points into the appropriate stats – in this case, Charm
and Intimidate. As Commander Shepard, the trilogy’s protagonist, the player becomes a kind of
galactic moral arbiter whose proclamations are as good as law for the NPCs they encounter. In
the first game this is taken to an absurd extreme when Shepard overhears a bereaved wife and
brother arguing passionately about whether to perform invasive surgery on the unborn child of
their deceased husband/sibling. At the player’s discretion, Shepard may intervene in the
discussion and offer trite advice that is taken at face value and presumably acted upon by the
volitionless duo. The net effect of this interaction is that these NPCs don’t feel like ‘Cs’ at all,
but automatons waiting in stasis for the player to arrive and resolve whatever deeply personal
issue is troubling them.
In drawing this comparison, we recognize that Firewatch is a very different sort of game
to Mass Effect and that feats of impossible persuasion, like feats of impossible physical prowess,
are part of what make Bioware’s grand space opera fun to play. What we wish to highlight is that
8
meaningful moral scenarios are not contingent on player-determined outcomes, and that letting
players exercise absolute moral authority may in fact negatively impact moral engagement,
robbing NPCs of their agency and virtual personhood. Conversely, Delilah’s stubbornness
regarding the police is potentially frustrating, but it’s in that moment of frustration that the player
is permitted to slow down and consider what Delilah’s decision says about her as a person.
That said, it’s important to acknowledge that designers who stymie player agency in this
fashion walk a fine line and that a frustrated player can easily feel cheated or misled, which may
negatively impact their capacity or willingness to reflect on the ethical dimensions of a given
scenario. Looking at negative user reviews for the game on Steam, one of the more common
refrains is that the game’s overall value is diminished by ‘railroading’ and a lack of
‘replayability’ – a direct consequence of the approach to agency outlined above. ‘Crap game. No
replayability. Storyline railroads you through worse than a "choose your own adventure" book,’
gripes review author Krellian. ‘The ending lacks a twist and it's the same ending no matter what
you do’ (2017). This is a problem admitting no easy solution and, as far as we can see, will
always be a danger associated with implementing the ‘slow technology’ approach advocated by
Sicart. All a developer can realistically do is tamper player frustration by contextualising their
lack of agency in the broader narrative and themes of the game. In the case of Firewatch, this is
achieved by positioning Delilah as stubborn, capable, and fiercely independent – the kind of
person who wouldn’t hesitate to say ‘no’ if she felt it was in her best interest to do so.
From a production perspective, one of the major advantages of leaving the player to
determine the moral significance of their choices is that it frees designers and writers from the
responsibility of authoring outcomes for those choices. As an indie studio with fewer than five
full-time employees, this was surely an important consideration for Campo Santo during the
development of Firewatch. Practical considerations of this sort are an important, but oft
overlooked, element of ENG design.
‘Social Battles’ in Deus Ex: Mankind Divided
There are currently four major titles in the Deus Ex franchise of first-person stealth/action
RPGs, with the latest released in 2016 for PC and console. The series is set in a 21st century
dystopia where multiple factions compete for control of world-changing technology, particularly
human augmentations. Originally developed by Ion Storm and published by Eidos, the series has
9
since passed into the hands of Eidos Montreal, who – with new publisher Square Enix – has
released two prequels: Deus Ex: Human Revolution (2011) and Deus Ex: Mankind Divided
(Eidos Montreal, 2016). For the purposes of this analysis, we will focus on Mankind Divided
with some reflection on Human Revolution. These two games play very similarly and have a
great number of mechanical and narrative commonalities that are notable from an ethical
perspective.
In keeping with the ‘immersive reality simulator’ design philosophy pioneered by its
predecessors, Mankind Divided can be played in a variety of styles, from pacifistic stealth all the
way through to explosive lethality. As an ‘aug’ or bio-mechanically augmented human,
protagonist Adam Jensen is designed to encourage the player to take a stance on the social and
moral issues built into the game’s ‘new bad’ dystopian fiction. A walking confluence of flesh
and tech, Jensen is distinguished from other augs by the fact that he does not suffer from
‘rejection syndrome’ (the biological body rejecting mechanical augments) and is therefore not
dependent on Neuropozyne, the drug used to suppress it. Jensen is special, in other words, and
this specialness is reflected in his almost absolute moral agency. Unlike Firewatch’s Henry,
Jensen is not limited to merely sharing his opinion, but has the power to effectively impose it on
recalcitrant NPCs via the game’s ‘social battle’ dialogue mechanic.
In these battles, the player must convince an NPC to cooperate by selecting dialogue
responses at each juncture in a branching conversation. Different choices provoke different
reactions from NPCs, positive and negative, with each juncture featuring at least one ‘optimal’
response. Failing a social battle does not lead to a game over or otherwise preclude further
progress, but succeeding is always associated with a reward of some sort – usually experience
points, but sometimes money, items, and information. Unlike Firewatch, the ethical significance
of social battles is intrinsically connected to their immediate narrative consequences, with
success typically facilitating non-violent resolutions to the conflicts embedded in the game’s
many quests and side-quests. Successfully ‘winning’ the social battle against the unhinged cultist
Allison Staněk, for example, facilitates her surrender and arrest; losing the battle, on the other
hand, results not only in her death, but the deaths of her fellow cult members as well. In this
sense, social battles are the ‘moral option’ for the diplomatically-minded concerned with
minimising violence and needless of life.
10
Among the many augmentations – super powers, essentially – available to Jensen, there
is one specifically designed to assist in social battles: a so-called ‘social enhancer’ named the
Computer Assisted Social Intelligence Enhancer or C.A.S.I.E. for short. C.A.S.I.E.
surreptitiously analyses the mental state of targeted individuals and includes a ‘dialectic
enhancer’ which allows Jensen to dominate conversation through chemical-pheromone
manipulation. In effect, Jensen releases a cocktail of pheromones into the air around his target to
render them psychologically malleable and cooperative.1
When the implant is unlocked, it provides a HUD overlay displaying key information
during social battles. Alignment analysis identifies NPCs as alpha-, beta- and omega- personality
types, their major personality traits are listed, and a pithy psychological profile is provided for
the player to review. When the NPC speaks, alphabetic symbols in the HUD light up to indicate
the optimal dialogue response. Animated graphics also pop up where C.A.S.I.E. makes special
notes about NPC mood indicators, such as increased heart-rate or dilated pupils. Finally, the
HUD displays a response level meter which will fluctuate during the conversation, based on
choices made and in such a way that optimal choices will see the meter fill up to its maximum,
visually confirming for the player the success of their strategy.
1
The ethics of using neurochemical coercion to get your own way remains sadly underexplored in both Mankind
Divided and its prequel.
11
Figure 2. Talos Rucker social battle. Source: Deus Ex: Mankind Divided (Eidos
Montreal, 2016)
Mankind Divided is replete with social battles but we will look at just one: a pivotal
encounter that takes place about half way through the game involving one of its major NPCs,
Augmented Rights Coalition (ARC) boss Talos Rucker. The goal of the battle is to convince the
charismatic but unstable militant to surrender without a fuss, avoiding further bloodshed, and for
players without C.A.S.I.E. this presents a serious challenge. With three dialogue choices
available at each of the four conversational junctures, there are dozens of unique paths through
the battle, many ending disastrously. Identifying optimal responses implies paying close
attention to Rucker’s rhetoric and pointing out its inconsistencies by drawing on the game’s
broader fiction, including Rucker’s own history. Small visual cues – the way he fiddles with his
glass, a shift in posture – provide insight to Rucker’s emotional state, but these are easy to miss
and even easier to misinterpret.
With C.A.S.I.E. activated, optimally resolving the Rucker battle is trivial. Reading
between the lines of Rucker’s rhetoric before making a considered response is no longer
required: the HUD, with its blinking glyphs and readouts, tells players everything they need to
12
know. Some simple text and graphics – the see-sawing of a meter, keywords to reductively
characterize Rucker, some flashing lights over basic symbols – capture the player’s attention as
the three-dimensional vividness of face-to-face conversation is reduced to a simple game of
Simon Says. As Reddit user Some_Guy_87 remarks, ‘Instead of really listening to the
conversations, I just watch a spot on the screen and count pings’ (2016).
In Sicart’s (2010) terms, C.A.S.I.E. explicitly draws the player’s attention to the
procedural layer of the social battle, to the nuts and bolts of its underlying systems, and away
from its semiotic layer. The result is a system that encourages the player to treat social battles
instrumentally, as trivial ludic challenges devoid of moral significance, undermining the
possibility of ethical cognitive dissonance while simultaneously reducing NPCs like Rucker to
treasure chests requiring the right sequence of commands to yield their goodies. Here again we
see how affording the player godlike moral agency undermines moral engagement: Jensen is not
talking to these people, he is commanding them, rendering their point-of-view, their objections,
their status as moral agents completely moot. C.A.S.I.E. affords the player the opportunity for an
immaculate playthrough, leaving them with nothing to reflect on ethically (Sicart a, 171). The
player need not interpret consequences and readjust in moments of limited agency because
C.A.S.I.E. always provides the optimal solution.
Mankind Divided’s prequel, Human Revolution, also features the C.A.S.I.E.
augmentation, but implements it in a slightly different way. In this game, C.A.S.I.E. displays the
personality type of the targeted NPC, but does not label responses with corresponding glyphs,
motivating the player to – as Reddit user revanchisto puts it in the previously quoted thread –
‘actually think about [their] response based on the psychological profile [of the NPC]’ (2016). In
this regard, Human Revolution’s implementation of C.A.S.I.E. is – from the perspective of moral
engagement – arguably superior to the system featured in Mankind Divided, giving players just
enough information to make an informed choice without reducing the decision-making process
to pure pattern matching. This maintains a rough balance between the game’s procedural and
semiotic layers, placing players in a position where they can contrast the system’s ludic
imperatives – select this to win the conversation – with the semantic content of what Jensen and
the NPC are saying to each other, creating the possibility of ethical cognitive friction. However,
because the optimal responses in Human Revolution tend to also be the most ethical and in
keeping with Jensen’s character, this possibility is sadly underexploited.
13
Systemic Cruelty in Darkest Dungeon
Darkest Dungeon is a cross between a dungeon crawler and RPG created by BritishColumbia based Red Hook Studios, released for the PC and Mac on January 19, 2016. Prior to its
release the game appeared on Steam Early Access, first for Kickstarter backers and then for a
general audience. The game has also been released for Sony’s PlayStation 4 and Vita platforms,
as well as for Linux and iOS tablets. The game has received largely positive critical attention,
and has sold more than one million copies (Sigman, 2016).
Darkest Dungeon builds on a familiar premise – the player must assemble parties of four
heroes to descend into dungeons, in order to slay monsters and gain loot. Dungeons are
procedurally generated, and players can level up their heroes over time to take and deal more
damage. The four dungeon areas feature gradually more difficult quests and bosses to defeat,
culminating in a final boss battle. Complicating an already difficult game with tough enemies
and little margin for error is permadeath. If a hero (or several heroes) dies on an adventure, they
are gone for good. It’s possible – and actually entirely likely – to experience full party wipes of
veteran-level heroes, which (ironically) the game will award you an achievement for. To manage
the flow of new and old heroes required for the game’s larger campaign, the player controls a
town where she can recruit novice heroes of various classes. There are multiple types of heroes,
but most offer some combination of direct combat, ranged attacks, area-of-effect spells, and
healing abilities. The town also features facilities where players can upgrade heroes’ weapons,
armour and abilities, as well a Tavern, an Abbey, and a Sanitarium for heroes to recover in after
tough adventures.
Many similar fantasy-themed role-playing games offer players various ethical or moral
dilemmas, usually built into the game through mechanics such as morality meters or various
types of karma systems. In contrast, Darkest Dungeon does away with alignment and karma
systems, instead opting to focus on the mental health of the heroes in the players’ care. As the
game’s creators explain ‘what really matters is how the heroes feel, not how the player feels’
(Sigman & Bourassa, 2015). Perhaps because of that shift in focus, there are no obviously signposted moral or ethical decisions for the player to make in the game’s storyline. One is not asked
to save one character instead of another or whether or not to activate a nuclear device and wipe a
14
small city off the map like in Fallout 3 (Bethesda Game Studios, 2007; Staines, 2010).2 The
story remains the same no matter how players approach the tasks set to them. The system is
again reminiscent of Nay and Zagal’s discussion of virtue ethics in games, where ‘even when
choices are inconsequential, they can be morally meaningful to their players when they are used
to gain insight (or develop insight) into the moral fibre of those characters players control’ (Nay
& Zagal, 2017). Similarly to Firewatch, choices and their consequences play out through the
small decisions that accrete over time concerning how players want to treat all, some or none of
their heroes.
Managing and monitoring mental health is one of the core mechanics built into Darkest
Dungeon. In addition to each hero’s hit points and various gear and weapon related stats, the
game displays a stress meter that slowly – or rapidly – climbs as heroes make their way through
a dungeon. Stress can be accumulated from participating in combat, and especially from the
attacks of certain classes of enemies who specialize in dealing stress rather than physical
damage. One such creature is the Madman, whose two main abilities are ‘Doomsday’ and
‘Accusation,’ which cause only 0 and 1 hit point of damage respectively, but instead add varying
levels of stress to their targeted victim, who is usually the hero with the highest stress level in the
party.
However, a hero’s stress also accumulates when torches lighting the way in a dungeon
burn lower, from lack of food, from seeing other heroes taking damage or being killed, as well as
through backtracking during exploration and in response to certain quirks a hero may have. A
hero’s stress meter slowly fills from such events, and eventually triggers the game’s Affliction
System. At the moment of triggering, the game will inform the player that the hero’s ‘resolve’ is
being tested, and things can go one of two ways – most commonly the hero develops an
affliction such as Fearful, Selfish or Paranoid. Occasionally the hero will instead gain a Virtue
such as Stalwart, Courageous or Powerful. Either way, Afflictions cause heroes to act out during
battles in ways that players cannot control, such as cowering and refusing to fight, or doing extra
damage in their attacks or self-healing. And stress keeps rising for Afflicted heroes until it hits
200 points, at which time the hero suffers a heart attack, and can be permanently killed with a
2
There is only one instance of such a choice in the game (that we know of). In the final boss battle, the final form of
the boss is Heart of Darkness and it has a particular attack called Come Unto Your Maker, which deals an instant
killing blow to one of your party’s heroes. However, the player must make the choice of which hero will be
sacrificed.
15
single blow. The stress and affliction system and how players choose to engage with it forms the
core of what makes Darkest Dungeon so interesting from an ethical standpoint. Players can
choose to engage with the system in ways that keep all their heroes healthy, or drive them to the
brink of madness and then dispose of them, or even nurture a particular set of heroes throughout
the game, while using other heroes as means toward that end. The game doesn’t reward or punish
any of those approaches – it instead is up to the player to decide what kind of overseer they wish
to be.
Indeed, as the town’s overseer, the player is never asked or directed by the game to be
particularly altruistic or ruthless in their handling of heroes, and heroes themselves are never
portrayed as inherently good or bad. Instead, similarly to Firewatch, players make multiple
seemingly mundane choices about how to treat their heroes– as potentially valuable assets to be
nurtured, or as cannon fodder to be disposed of when most convenient.
Case in point: when provisioning a group for a dungeon crawl, the player can purchase as
much or as little food and torches as they deem necessary to complete the event. Provisioning
many or few torches (which are fairly cheap) may seem ethically neutral, but it can have a
powerful impact on the mental health of one’s heroes – both positively and negatively. At full
light levels heroes have better chances of scouting ahead for traps or enemies, but as light levels
decrease, its more probable that the party will be taken by surprise by enemies, that heroes will
suffer more stress damage, and that enemies make stronger attacks. For some players then,
keeping the lights on becomes a priority, no matter the potential cost involved. However, in
lower light levels heroes can also gain greater loot and increase their chances of striking critical
hits, possibly pushing players in the other direction, of keeping their heroes in the (literal) dark.
And individual heroes may also have afflictions that greatly increase their stress levels in low
light, or even react poorly to full light conditions, exacerbating those original decisions. Players
must take all such conditions into account when deciding how many torches to purchase,
potentially confronting ethical cognitive dissonance as the ludic imperatives of optimally
managing available resources (including the human variety) conflicts with the cultural
significance of intentionally torturing people for profit.
16
Figure 3. A hero suffering multiple afflictions. Source: Darkest Dungeon (Red Hook
Studios, 2016)
Another series of micro-choices that the player must confront involve how to deal with
afflictions that heroes accumulate during adventures. While stress can be reduced or eliminated
by trips to the Abbey or Tavern (with treatments that cost varying amounts), certain afflictions
can only be removed through a visit to the Sanitarium’s Medical and Treatment Wards. Some
afflictions aren’t cheap to remove. For example, a hero in one of our games (Bose, a level 5
Leper) currently suffers from Necromania (fascination with corpses), Cove Phobe (+20% stress
in the Cove dungeon), Resolution (won’t drink while in town), Enlightened (will only meditate
while in town) and Fear of Eldritch (+15% stress versus Eldritch type enemies; -10 Accuracy
versus Eldritch). It would cost 30,100 gold to remove all those afflictions. Similarly, Watteau, a
level 5 Man-at-Arms suffers from several similar afflictions, but also has three Diseases: Bulimic
(-20% healing in Camp), The Red Plague (-75% Bleed resist) and The Runs (-20 Dodge and 10% maximum HP). His diseases are a relative bargain – only 1,138 gold to remove each one.
As a point of reference, missions pay out various amounts for successful completion – ranging
from around 3,000 to more than 12,000 gold coins, and heroes will often find more gold and
17
valuables while exploring. However, it costs several thousand gold to provision a party for a
mission, and if they fail there is no reward given – only an increase in their stress levels.
In their game design discussions, the developers have repeatedly made the point that they
wanted to ‘toy with player agency’ and also to ‘capture the human response to stress’ (Sigman &
Bourassa, 2015). In our analysis, Darkest Dungeon succeeds in capturing the ‘human response’
to (and by extension: moral significance of) stress partly because it toys with player agency,
systemically limiting the player’s ability to resolve the game’s implicit ‘wicked problems’
(Sicart, 2013) with Mankind Divided-style optimal solutions. Adventurers cannot avoid trauma,
and as harm accumulates over the game’s slow grind this trauma becomes harder and harder to
ignore, giving the player plenty of time to appreciate it from an ethical perspective.
We believe the system could be even more successful in this regard if the game’s
semiotic layer did more to humanize the adventurers under your control, exacerbating the friction
that results from treating them poorly. Sicart’s (2011) extended discussion of the Sorrow
sequence in Metal Gear Solid 3: Snake Eater (Konami, 2004) points to a way this might be
achieved. In this sequence, the player must negotiate a river swarming with the restless spirits of
fallen foes. If the player has been liberal dispensing lead, the river will be dense with ghosts that
hold back the avatar, hindering progress and providing space for the player to reflect. Sicart
writes, ‘this gameplay sequence is one of the most accomplished translations of the ethical
possibilities of games into actual game design.’ (Sicart, 2011, p.107). Perhaps if, as reviewer
Joseph Anderson suggests (2016), Darkest Dungeon did something similar, confronting the
player with all their dead heroes at the game’s conclusion, a similar effect could be achieved.
The game’s graveyard, where players can go to view tombstones of fallen heroes, is along the
same lines, but the effect is muted by the identical headstones and procedural descriptions.
By drawing on the familiar rhetoric of mental health to frame stress and ailments,
Darkest Dungeon already makes an implicit plea for the player to sympathize with their suffering
heroes, but perhaps this plea would be more persuasive in a game with a more serious, less jokey
tone. While most of us can sympathize to some degree with mental health issues like depression
and paranoia, afflictions like Necromania and Fear of the Eldritch are, we hope, far less relatable.
As an example of how tone might reframe the systemic logic of the stress/ailment system,
imagine a similar system implemented in a game about managing staff in an emergency ward or
a platoon of soldiers in Afghanistan, where ailments like ‘PTSD’ are not inflicted by giant
18
spiders but by the loss of a patient or an encounter with a roadside IED. While far easier said
than done, recontextualizing gameplay systems in this fashion may be a sensible, efficient
approach to ENG design for academics, indies, and other resource-limited developers.
Conclusion
In this article we have examined three recent, popular examples of ethically notable
games (ENGs): Firewatch, Deus Ex: Mankind Divided, and Darkest Dungeon. Perhaps more
than anything else, what our analysis indicates is that commercial game designers are exploring
more diverse, sophisticated approaches to representing morality and moral choice in their games.
Where ‘big choices with big consequences’ once dominated ENGs, we now find a multitude of
games – particularly within the indie sphere – exploring micromoral scenarios, refocusing on the
small everyday decisions that, in aggregate, paint a picture of our moral character. Where
‘morality meters’ once proliferated we now see games like Darkest Dungeon and Papers, Please
(Pope, 2013), featuring core gameplay loops that are intentionally problematised at the semiotic
level (Formosa, Ryan, & Staines, 2016) provoking ethical cognitive dissonance and compelling
the player to reflect on choices they’ve made and must continue to make if they wish to keep
playing. In stark contrast to heroic epics like Mass Effect, these games indicate that selectively
restricting the player’s moral agency is an effective means of ‘slowing down’ gameplay and
providing space for a reactive, instrumental player to focus on the broader significance of their
actions. Conversely, absolute moral agency may have the opposite effect, diminishing the
humanity of NPCs and reducing dialogue to a series of diktats delivered by the godlike player.
Of the three games we’ve analysed, Firewatch stands out for its consistent and
accomplished implementation of the design techniques discussed. Nevertheless, it’s worth
remarking that, on our viewing of numerous playthrough videos on YouTube, we found players
still eager to fulfill what they perceived to be the game’s ‘traditional’ ludic goals. Many players,
for example, had Henry put all collectible items in his pack, irrespective of who they belonged to
or what the implications of taking them might be. This isn’t surprising: the imperative
underpinning many videogames – particularly of the adventure/RPG variety – is to be a packrat
and collect absolutely everything and anything in the off chance it may come in useful down the
track. Disabusing players of these long-ingrained habits and encouraging them to view
gameworlds as more than purely instrumental is no mean feat, and as we can see even games that
19
do a superior job of facilitating ethical play falter in the face of someone determined to play
unreflectively. This is both a drawback of ENGs generally and a necessary, crucial component of
their design. The player must choose to be reflective because free choice is the fundamental basis
of ethical responsibility: it isn’t enough to simply tell players when their choices are morally
significant, or to force ethical reflectiveness with reductive metrics like morality meters and
karma points. In short, if we want players to play reflectively, we must convince them to play
reflectively. Design features described in this article go some way toward achieving this goal, but
there is still ample room for improvement and we look forward to seeing how ENG design
continues to evolve in the future.
References
Anderson, Joseph. (2016). Darkest Dungeon Review and Critique [YouTube]. Retrieved from
https://youtu.be/_Buwei6ZWqU
Aristov, M. (2012, May 29). An alternative to the Karma Meter.
Bethesda Game Studios. (2007). Fallout 3.
Bioware. (2003). Star Wars: Knights of the Old Republic.
Bioware. (2007). Mass Effect.
Birch, A. (2014, March 28). The Problem of Morality in Videogames. Retrieved January 9,
2019, from https://www.denofgeek.com/games/29837/the-problem-of-morality-invideogames
Campo Santo. (2016). Firewatch.
Carr, D. (2005). On the contribution of literature and the arts to the educational cultivation of
moral virtue, feeling and emotion. Journal of Moral Education, 34, 137–151.
https://doi.org/10.1080/03057240500127053
Carr, D. (2006). Moral education at the movies: on the cinematic treatment of morally significant
story and narrative. Journal of Moral Education, 35, 319–333.
https://doi.org/10.1080/03057240600874448
20
Consalvo, M., Busch, T., & Jong, C. (2016). Playing a Better Me: How Players Rehearse Their
Ethos via Moral Choices. Games and Culture, 1555412016677449.
Crawford, C. (1985). Balance of Power.
Eidos Montreal. (2011). Deus Ex: Human Revolution.
Eidos Montreal. (2016). Deus Ex: Mankind Divided.
Flanagan, M., Nissenbaum, H., Belman, J., & Diamond, J. (2007). A method for discovering
values in digital games. Presented at the Situated Play, Proceedings of DiGRA 2007
Conference.
Formosa, P. Ryan, M., Staines, D. (2016). Papers, Please and the Systemic Approach to
Engaging Ethical Expertise in Videogames. Ethics and Information Technology, 18.3,
211–225
Galactic Cafe. (2013). The Stanley Parable.
Heron, M., & Belford, P. (2014a). ‘It’s only a game’ — ethics, empathy and identification in
game morality systems. The Computer Games Journal, 3(1), 34–53.
https://doi.org/10.1007/BF03392356
Heron, M., & Belford, P. (2014b). ‘Do you feel like a hero yet? Externalized morality in video
games. Journal of Games Criticism, 1(2), https://doi.org/10.1007/BF03392356
Infinity Ward. (2009). Call of Duty: Modern Warfare II.
Konami. (2004). Metal Gear Solid 3: Snake Eater.
Krellian. (2017, September 11). Not Recommended. Steam. Original Post. Retrieved July 30,
2019, from https://steamcommunity.com/profiles/76561197964117895/recommended/383870/
Lange, A. (2014). “You’re just gonna be nice”: How players engage with moral choice systems.
Journal of Games Criticism, 1.
21
Melenson, J. (2010). The Axis of Good and Evil. In K. Schrier (Ed.), Designing Games for
Ethics: Models, Techniques, and Frameworks (pp. 57–71). Hershey, PA: IGI Global.
Nay, J. L., & Zagal, J. P. (2017). Meaning Without Consequence: Virtue Ethics and
Inconsequential Choices in Games. In Proceedings of the 12th International Conference
on the Foundations of Digital Games (p. 14:1–14:8). New York, NY, USA: ACM.
https://doi.org/10.1145/3102071.3102073
Origin Systems. (1985). Ultima IV: Quest of the Avatar [Apple II game].
Pope, L. (2013). Papers, Please.
Rest, J. R., Bebeau, M. J., & Thoma, S. J. (1999). Postconventional moral thinking: A neoKohlbergian approach. Psychology Press.
revanchisto. (2016, August 31). CASIE Enhancer makes conversations TOO EASY. Reddit.
Original Post. Retrieved January 9, 2019, from
https://www.reddit.com/r/Deusex/comments/50fd4o/casie_enhancer_makes_conversation
s_too_easy/
Ryan, M., Staines, D., & Formosa, P. (2016). Four Lenses for Designing Morally Engaging
Games. In Proceedings of The 1st International Joint Conference of Digra and FDG,
Abertay University, Scotland
Schrier, K. (2015). EPIC: a framework for using video games in ethics education. Journal of
Moral Education, 44, 33.
Sicart, M. (2010). Values between systems: Designing ethical gameplay. In Karen Schrier (Ed.),
Ethics and Game Design: Teaching Values through Play (p. 1–15). Hershey, PA: IGI
Global.
Sicart, M. (2011). The ethics of computer games. MIT Press.
22
Sicart, M. (2013). Beyond Choices: The Design of Ethical Gameplay. Cambridge, MA: The MIT
Press.
Sigman, T. (2016, November 3). 1 Million Copies and the Road Ahead. Retrieved October 30,
2017, from
https://www.gamasutra.com/blogs/TylerSigman/20161103/284863/1_Million_Copies_an
d_the_Road_Ahead.php
Sigman, T., & Bourassa, C. (2015, May 28). Game Design Deep Dive: Darkest Dungeon’s
Affliction System. Retrieved October 27, 2017, from
/view/news/244480/Game_Design_Deep_Dive_Darkest_Dungeons_Affliction_System.p
hp
Smelthurt T, & Craps, S. (2015). Playing with trauma: Interactivity, empathy, and complicity in
The Walking Dead video game. Games and Culture, 10(3), (p. 269–290)
Some_Guy_87. (2016, August 31). CASIE Enhancer makes conversations TOO EASY. Reddit.
Reply. Retrieved January 9, 2019, from
https://www.reddit.com/r/Deusex/comments/50fd4o/casie_enhancer_makes_conversation
s_too_easy/d73oel4
Staines, D. (2010). Videogames and moral pedagogy: a neo-Kohlbergin approach. In K. Schrier
& D. Gibson (Eds.), Ethics and Game Design: Teaching Values Through Play (Vol. 2, p.
35). Hershey, NJ: Information Science Reference.
Stevenson, J. (2010). A framework for classification and criticism of ethical games. In K. Schrier
& D. Gibson (Eds.), Designing Games for Ethics: Models, Techniques and Frameworks
(Vol. 1, p. 36). Hershey, NJ: Information Science Reference.
The Fullbright Company. (2013). Gone Home.
23
Weaver, A. J., & Lewis, N. (2012). Mirrored morality: An exploration of moral choice in video
games. Cyberpsychology, Behavior, and Social Networking, 15, 610–614.
White, O. (2016). Firewatch took away our ability to be good people, and that’s where it shines.
Retrieved January 11, 2017, from
https://www.polygon.com/2016/2/12/10966494/firewatch-agency-campo-santo
24