Navigation – Plan du site

Ontology of Gender in Computer Games

L'ontologie du genre dans les jeux vidéo
Ivan Mosca


Cet article vise à illustrer comment l'ontologie sociale peut remplacer l’approche traditionnelle constructiviste pour l'analyse des médias. En particulier, on utilise les outils de l'ontologie sociale pour classer la relation entre l'identité sexuelle et les jeux vidéo. L'évolution des jeux vidéo résume exactement l'évolution de l'identité de genre ces dernières années, et ce d’autant qu'ils ont largement influencé la libre représentation des catégories sexuelles non traditionnelles (jeux japonais) et le libre choix de l'identité sexuelle (jeux américains).

Haut de page

Texte intégral

1. Over Constructivism

1During the Sixties, structuralism was criticized because it described institutions as reified social relations, creating the basis for a metaphysical justification of the status quo. Constructivism therefore took its place, becoming the most important scientific grand narrative of the last fifty years. But today things are changing. The political events at the turn of 21th century, such as the rise of Italian and North American populisms (which programmatically deny reality in favour of its representation), have shown that even constructivism can be used for the justification of social powers. The current field of human science is becoming more and more a reality-based community, against the Weltanshauung expressed by the Bush administration with these famous words (Suskind 2004):

We're an empire now, and when we act, we create our own reality. And while you're studying that reality — judiciously, as you will — we'll act again, creating other new realities, which you can study too, and that's how things will sort out. We're history's actors…and you, all of you, will be left to just study what we do.

2This clarifies that the role of scientific knowledge in the battle for rights and justice cannot be a simple criticism of the notion of truth – like some postmodern branches of constructivism seem to suggest, but rather the affirmation of truths against the falsity of the reifications proposed by the social powers. The Nineties, as witnessed by the affaire Sokal, led to the end of the 'narrative turn', the postmodern incarnation of the philosophical 'linguistic turn' which spawned constructivism. So a 'new realism' has arisen (Ferraris 2012), which sometimes relies on a biological Marxist-materialism, sometimes on a 'documental' institutionalism called 'social ontology'. The recent debate between these new realisms is well represented by the diatribe about historical researchers Jared Diamond (1997) and Daron Acemoglu (2012). In any case, postmodernism is strongly criticized or completely dismissed. Even Bruno Latour, one of the most aggressive postmodernists, stated that the constructivist point of view became a strategy of political and economic powers, which exploit uncertainty to deny natural and social facts, like global warming, the absence of weapons of mass destruction in Iraq, Greek national debt and/ or the injustice of the measures adopted by the Greek government to pay this debt (Latour 2004). In another text, he strongly defended the emerging discipline of 'ontology', probably not recognizing he was talking about 'social ontology' (Latour 2005).

3Postmodernist constructivism has followed Nietzsche's claim “there are no facts, only interpretations”. Ontology instead starts from the common and realistic supposition that there are facts before interpretations. The branch named social ontology treats these same interpretations as facts.

Table 1. From subjects to objects


No facts under interpretations


Facts under interpretations

Social Ontology

Interpretations are facts

4As a discipline, social ontology was re-discovered by the philosopher John Searle during the Nineties. According to Searle, social ontology studies objects that exist only insofar as people believe they exist, such as marriage, money, debts, laws, games, gender, etc. According to social ontologists, these institutions are objective and cannot be denied by the individual. Today in many fields, from history to gender studies and from psychology to sociology, social ontology is taking the place of constructivist and linguistic approaches which have characterized the last fifty years of cultural research. So objective properties, causes and effects, hierarchies, realist structures and other non-linguistic explanations of phenomena are returning to the core of explanations.

5Game studies began with a constructivist approach, ‘narrativism’, which viewed games as texts (types of text, ways to narrate stories, linguistic phenomena, communication systems, etc.), but the most important authors of this discipline have since contested narrativism, introducing a ludological approach which is more object-oriented. Today’s game researchers, such as Ian Bogost (2006), use ontology instead of the early adopted semiotics, a discipline strongly influenced by constructivism.

2. The Ontology of Games is a Social Ontology

6Social ontology can be helpful in investigating the nature of gender in games and the common sense of game researchers about this topic. Game studies are the result of many disciplines, from informatics to philosophy, from media studies to semiotics. All these disciplines have different approaches regarding their formal elements, what philosophers call “Peirce's types”. Every type refers to a multiplicity of different but similar tokens. There are three logical levels of types: the first level is tied to Epistemology, that conceives of the type as a name for a set of tokens (e.g. the name “game” applied to a set of different objects or activities); the second level is tied to ontology, that conceives of the type as a property of a set of tokens (e.g. the property of “games” is “to be an interactive fiction”); and the third level is tied to Social Ontology, that conceives of the type as a rule regulating some practices in relation to a set of tokens (e.g. the definition of “game” is strictly related to the meta-rule “you cannot play at work”). All these concepts of type are involved in the definition of 'gender' in computer games.

7Perhaps a little experiment can illustrate how the socio-ontological approach can influence game studies research. What is the result of the dice roll below?

Figure 1. Epistemology, Ontology, Social Ontology

8At a certain epistemological level of granularity, these dice show the very same face, because we identify them with a single name ('six'). But for an ontologist, the different spatial, temporal, causal and effectual properties of these dice show that their upper faces are different. Finally, social ontology allows to understand that we can count two different tokens as a unique type because we join a collective intentionality regulated by the following constitutive rule: “the ideal number which the concrete upper face of a die refers to... counts as the result of that die roll”. John Searle (1995) has defined the formal string of constitutive rules:

9X counts as Y in C

10So a X-physical object is counted as a Y-status function (the social object) in a C-context, the frame. Also the interpretation (but not the functioning) of informatics (* ?) processes is based on cultural entities which lead to count a physical (analogical) object (such as a die – or an electrical exchange) just as one of the possible states of a discrete and ideal system (such as output – or the natural numbers from 1 to 6). According to the inventor of computer, Alan Turing, people treat computers as discrete states machines, but “strictly speaking there are no such machines. Everything really moves continuously” (Turing 1950, p. 439). Thus, by following a certain constitutive rule, people can conceive continuous (analogical) machines as discrete (digital) machines. The constitutive rules of social ontology also regulate the commonsensical division between hardware and software (Moor 1978) and the commonsensical conception of computer processes as governed by rules. All these pre-assumptions/ understandings are fading today, in parallel with the constructivism that determined them. Constructivism in turn was influenced by the second Wittgensteinian theory of language:

11Wittgenstein (1953) → Constructivism → Games and Computers are governed by Rules

12According to Wittgenstein, culture is a set of language-games, which are the rules that define our possibility to interact with society. This theory has contributed to the constitution of a particular common sense, widespread in the scientific environment of game researchers, who think that games are necessarily defined by rules. But exactly what is a rule?

13John Rawls (1955), inspired by the Kantian categorical and hypothetical imperatives, divided rules in two types. Prescriptive rules are the generic type of token commands (“in order to exit from the museum, visitors must use only the exit door”): they imply duties and prohibitions, which require the free will of the subject who must obey them. Descriptive rules are instead objective representations of how a system works (“in order to exit from the museum, visitors must use a door”): they are not directed to subjects and primarily are used by scientists. The ontology of informatics shows that computers do not involve prescriptive rules. Computers are machines which operate by deterministic causes and without intentional states. Their physical functioning can be profitably described using descriptive rules, but this does not explain how a computer actually works. Prescriptive rules are directed to subjects that understand them as prescriptions, they are part of social reality; whereas computers cannot disobey or cheat program instructions. Neither do users obey rules: they just interact with affordances. For example, computer chess moves (exactly like in Tetris) cannot be invented or 'disobeyed'. Unlike board chess, in computer chess invalid moves are simply impossible to make, like the sprite of a Tetris block cannot pass through another (Mosca 2013).

14These ontological considerations are not only applicable to computer functioning, but also to game structures and gender identities. Analysing common language, Ludwig Wittgenstein (1953) noticed that there are no objective properties shared by all games. This has incited many game researchers to oppose Wittgenstein, trying to find at least one property shared by all games. For example one of most famous of them, Bernard Suits (1978), has interestingly changed Wittgenstein's empirical motto ('don't think but look !') into a participative one ('don't look but play !'). However, Suits probably has played only one type of game, because he reduces 'gaming' to an autotelic 'overcoming of useless obstacles'. Many researchers have individuated some 'necessary' properties of all games, but ontological analysis shows their lack of necessity or alternatively their insufficiency. For example, a necessary but insufficient property of games is interaction: effectively all games (not only computer games – Aarseth 1997), share interactivity; but many other non-gaming activities also have interaction. Specifically, the interaction of games is material, whereas the interaction of literature and other media is just symbolic, as the Constance School has shown (Iser 1976). In any event, most game researchers share the belief that the most important property common to games is their supposed necessary relation to rules. This is an inheritance from the constructivist semiotics where game studies originated (and from which they want to detach, but with many difficulties). Many researchers, such as Salen and Zimmermann (2003), have argued that games are a medium totally constituted by rules. But, as we have seen, normal computer games have no rules, just like many non-computer games: playing with puppets, building sand castles, and all Ilinx or Mimicry games (Caillois 1958). Not every game is a rule system and not every rule system is a game (e.g. Law). So the formalist approach that reduces games to rule systems is neither sufficient nor necessary.

15How to build an ontology that does not deny the Wittgensteinian evidence that there are no objective and shared properties in all the activities that a single subject can call 'games'? The only possible answer comes from social ontology. Indeed we can interpret Wittgenstein's theory using the filter of Saul Kripke (1982): if the subject can use a single name (epistemology) for different things (ontology) then the shared property is the subject herself (social ontology). Therefore the ontology of games, which are social objects, is principally directed to investigate how different subjects define 'games': thus it is a social ontology. This discipline emphasizes the fact that people behave following constitutive rules: prescriptions with the appearance of descriptions. So the constitutive rule “this piece of paper is a 20-dollars banknote” is a prescription directed to subjects, but it hides itself under a shape of objective prescription. All socio-ontological performative speech acts hide a prescription (an 'ought') under a description (an 'is'). The entire social system is supported by constitutive rules, from the law system (“la France est une République indivisible, laïque, démocratique et sociale”), to fashion (“this dress is cool”), customs (“the color of the funeral is black”), economy (“the interest rate of Greek bonds is higher than the German”), religion (“this bread is the body of Christ”), games (“this wooden token is a Bishop”) and sexuality (“homosexuality is a perversion of natural behavior”). Social Ontology studies the relation between 'is' and 'ought', geographical territory and political map, in the social environment made by beliefs.

3. The Social Ontology of Gender

16Searle's formula for social objects “X counts as Y in C” is also profitably applicable to gender issues. For example a X-physical object (a body, a person) counts as a Y-social object (a woman) in the C-context (creating or recognizing documents, programming or selling games, etc.). Therefore we can try to understand how common sense intends gender. This does not imply that the socio-ontological categories about gender identities are right or that they cannot be changed. A scientific description of some ontological properties is not their socio-ontological prescription.

17According to many scholars, gender is not a discrete category. Alice Dreger (2001) has written that “the sex spectrum is like the color spectrum” and according to Anne Fausto-Sterling (1993) sexual difference is a continuum rather than a binary, just as we distinguish between the analogical functioning of computers and the discrete states which users adopt in order to interpret them. Common sense prescribes some behaviours by describing a binary categorization (male-female), but we cannot simply reject it: in order to describe common sense, we have to understand the implicit prescriptions in its explicit descriptions.

18According to one of the most important authors of constructivism, Michel Foucault (1976), 'sex identity' is a cultural script, not a biological property, so it is relative and without consistency. Although respecting the political intentions of Foucault, an ontologist cannot completely agree: sex identity is a social construct that is part of social reality, just like any other identity. Id-entities are always epistemic thoughts, doppelgangers of ontological entities. What is socially constructed is the social Gender identity, not the biological Sex entity. Judith Butler (1990), another constructivist, also treats Sex and Gender as a same thing, but there is an important difference between the socially constructed Y-Gender and the biological X-Sex. Like many other postmodernist thinkers, Butler claims that X-entities are not under Y-identities. However, we have seen that cultural objects cannot be explained if their material substrate is ignored.

19A clearer explanation comes from Carrie Hull (2006) and Borghini and Casetta (2011), who have developed an ontological approach to gender very different from the constructivist views that dominated gender studies in the last decades of 20th century. Hull maintains “a distinction between sex and gender, and sexuality and behavior” (Hull 2006, p. 5). Thus, investigating the social reality of common sense, we can extend Hull's approach by distinguishing Sex, Orientation and Gender.

20Sex (e.g. female) is a biological type related to a multiplicity of anatomical properties, Orientation (e.g. homosexual) is a psychological type that relates to the the direction(s) of erotic preferences, and Gender (e.g. woman) is a cultural type that contributes to define the social identity of a person and her legal status. Obviously, Sex, Orientation, and Gender are not necessarily bound in a determined structure, e.g. there are females (Sex) who are heterosexual and females who are homosexual (Orientation), but the traditional common sense binds them in a fixed structure that could be called the 'traditional triad'.

Table 2. The traditional triad










21So, according to the traditional triad, if you are of, for example, female Sex you should have a heterosexual Orientation and a woman Gender. But there are many other combinations, which include Homosexuality, Bisexuality (a variation of the relation between Sex and Orientation), Transsexualism (a variation of the relation between Sex and Gender), Transgenderism (a variation of the relation between Orientation and Gender) and Intersexualism (a variation of the relation between primary, secondary and chromosomal phenotypes).

Table 3. Other cominations














Males and Females










Classic gender









Not-male and not-female



22In this table, question marks specify where the datum is not important for the definition. In order to better understand why the traditional triad is inadequate, we have to take a closer look at our categories.

23Sex (e.g. female) can be considered as a pheno-type, the description of some empirical properties. Obviously Sex descriptions can be used as a prescription, becoming a social object. The traditional Sex categories are 'male' and 'female', but today science individuates primary phenotypes (genitals and other organs linked to reproduction), secondary phenotypes (pummel, beard, tone or timbre of voice, etc.), chromosomal phenotypes (XX, XY, XXY, etc.), and hormonal phenotypes (related to the quantity of certain hormones produced by the organism). Although these descriptions have a sincere scientific purpose, they leave some issues open. For example, sometimes doctors carry out surgery on the genitals of a newborn, thereby choosing the Sex category of the child. Sex identities (Y-status) are applied to physical phenotypes (X-objects), but they depend on the collective intentionality adopted in a specific cultural frame (C-context), which can vary. For example, the Olympic committee oversees the division of the athletes into different Sex categories. During the Sixties the parameters switched from secondary to primary phenotypes, but many problems arose in the subsequent twenty years which led the committee to use chromosomal phenotypes and finally a method based on hormones. Still today it is hard to determine if an athlete is male or female, because of the vagueness of the boundaries of human body. But as we have seen, when the common sense counts an analogical boundary (such as a natural river) as a discrete boundary (such as a cultural frontier) the individual cannot change it simply by disobeying the prescriptive rule hidden under the descriptive one (Smith and Varzi 2000). In order to do this, significant institutional means are required. The Olympic committee needs to find a way to mark a discrete boundary between sexes in order to respect the analogical boundaries of the majority of people, who normally have different levels of performance dependng on whether they are male or female. Indeed, without a discrete categorization, almost all the people currently considered female would be excluded from international competitions because of their inferior results, as in the first 1896 Olympics.

24Orientation (e.g. homosexual) is a psychological attitude, an attraction, a desire. There is a difference between Orientation and sexual behaviour, because not everyone acts in accordance with their Orientation. Orientation has many subsets: there is an Orientation towards a type of Sex (e.g I like females), an Orientation towards another type of Orientation (e.g. I like homosexuals) and an Orientation towards a type of Gender (e.g. I like women). These subsets can be simple Orientations, complex Orientations and supercomplex Orientations (excluding other types of attraction, such as the attraction for blonde hair, for rich people, for protective people, etc...). Below are some examples from the point of view of a male, heterosexual, man (you may want to substitute his entries with your own):

Table 4. Simple Orientations for a Male, Heterosexual, Man.




I like females

I like males


I like homosexuals

I like heterosexuals


I like women

I like men

25Abélard is male, heterosexual and a man. His Orientation can be driven by the Sex, Orientation or Gender of his partners. E.g. if Abélard is Sex-driven, he is attracted by Héloïse because she is female. If Abélard is Orientation-driven, he is attracted by Héloïse because she is heterosexual (or, in the case of complementarity, because she is homosexual). Finally, if Abélard is Gender-driven, he is attracted by Héloïse because she is a woman (or, in the case of symmetry, because she is a man). It is important to note that a Sex-driven complementary person can like both homosexual females and heterosexual females, women female and men females, etc. People do not usually present a simple Orientation, but rather a complex one, which includes not one, but many categories. So, in the case of the Abélard (a male, heterosexual, man), he could like females who are both heterosexual and homosexual but only females who are women. The number of complex Orientations is very high.

26A subset of complex Orientations is supercomplex Orientations, as in the case of a man who likes women who behave like men but that phenotypically appear as females. Supercomplex Orientations can present potentially endless combinations. Some examples:

(A) males who like people who like them,
(B) males who like females who like them,
(C) males who like females who like other females,
(D) males who like people who are attractive to other people
(E) males who like (A)
(F) males who like (E)

27Gender (e.g. woman) is a cultural type that contributes to defining the social identity of a person and her legal status. Gender can be based on Sex or Orientation, or without any link with them. Gender can be constituted in a variety of ways:

Table 5. Types of Gender

Belief about a social object

E.g. I believe that men are those who are like their father, or those who have moustache, etc.

Behaviour as object of the belief

E.g. I believe that men are those who are rude, or those who are heavy drinkers, etc.

Behaviour of other people

E.g. I believe that men are those who are respected, or those who attract women, etc.

Sex as object of the belief

E.g. I believe that men are those who have a penis, or a big penis, or a working penis, or a fruitful reproductive apparatus, etc.

Orientation as object of the belief

E.g. I believe that men are those who like women, or beautiful women, etc.

28In the majority of cases, individuals share their beliefs about Gender with the majority of other people (what is called common sense), but in some cases individuals have different beliefs. For example I could believe myself to be a man, but many people could believe that I am a woman. The conflict between individual beliefs and shared beliefs can be solved by education, which is becoming more and more structured by the use of media. Today computer games are probably the most used medium by children and teenagers, so it is important to understand how they treat gender topics.

4. The Social Ontology of Gender in Computer Games

29Games are normally designed for the market with a Gender target ('gamer boys', 'party player girls', 'casual player women', 'retired old men', etc.). Marketing managers do not take into account the differences between Sex, Orientation and Gender, which are well distinguished, however, by designers. To discover the Gender of buyers, marketing takes into account only their Sex. Recently an important study (ESA 2011) demonstrated that 42 % of players and 48 % of buyers are female.

Figure 2. Sex or Gender?Image 2000000900002CF600000FB72D54CF83.wmf

30The researchers innocently treat Sex-identity as Gender-identity. This is not an exception, but the normal way to do marketing. For example, another study (Winn 2009) has found a relation between play time and free time. The research has also individuated a relation with Gender: women play less and in shorter sessions than men. The reason is that women have more duties so they have less free time, which is also divided in shorter time chunks, than men. Even in these excellent studies, Gender is divided only into two types (man-woman) and detected by the answers to a « M or F » question. We don't know if interviewees answered this question in relation to their primary biological phenotype, their psychological Orientation or their cultural Gender. This gap in the knowledge cannot be ignored: neither ‘common sense’, nor scientific research makes distinctions between Sex, Orientation and Gender.

31Research data could be better interpreted using socio-ontological categories, which can be applied not only to players, but also to characters (avatars), defined by Taylor (2003) as 'intentional bodies', so fully social objects. Does Pac-Man's avatar have a Sex? An Orientation? A Gender?

Figure 3. Pac-Man

32The first Pac-Man version (Iwatani 1980) is abstract and without any characterization. But the first sequel introduced a wife (McRae 1981), so that players learned that Pac-Man is a male heterosexual and that the final word of his name did not just signify 'human' but also that he is 'man'. Initially, the new character was called 'Mrs Pac-Man', but female programmers protested, trying to change it to 'Miss Pac-Man'. Unfortunately there was a Baby Pac-Man character too, so the main character became a neutral 'Ms. Pac-Man'. To change Pac-Man to Ms. Pac-Man designers used a single attribute, a red hair-ribbon:

Figure 4. Ms. Pac-Man

33This simple attribute gives a Gender to Ms. Pac-Man, because red ribbons are used only by the culturally defined woman-Gender. According to the traditional triad, if a character has the Gender ‘woman’, then she is also female (Sex) and heterosexual (Orientation). The majority of games presents the traditional triad: for example in World of Warcraft (Pardo 2004), secondary phenotypes (like muscles) directly generate Gender types (like men). This means that the choice of the body appearance automatically entails the choice of Gender:

Figure 5. Races and Sexes

34In many fantasy settings, Sex directly affects Gender: for each 'race' (elves, orcs, etc.) or 'class' (mages, warriors, etc.), the female armor has significantly less surface area than their male counterparts. There, women are just naked females.

35Over the last 30 years, many game characters have contributed to changing the traditional triad, starting with Orientation. Indeed, many mainstream games, unlike mainstream movies, emphasize non-traditional variations of Sex-Orientation-Gender combinations. In the beginning, homosexuality was not directly represented, but just suggested. For example, the first game with a clear gay target, Stroker (MCS 1983), depicted a solitary guy in the act of masturbating: sexuality - but not homosexuality - was explicit. After the 1983 market crisis, the publisher Nintendo built its success by clamping down on production and using many restrictive rules in order to broaden the marketing target without putting off the parents of the players with explicit sexual representations (Schwartz 1994). This led many minor publishers, like Sega, to encourage its developers to include sexual themes without explicit representations: homosexuality was the main theme, both in a gay and an anti-gay view. The initial introduction of non-traditional sexual themes was dominated by homophobia, which became more and more homophilia. According to De Santis (2013, p. 36):

over the years, the role of LGBT characters follows the morphological classification of fables, passing from the Guardian to the Shadow to the Hero. This change in role marks first negation, then refusal, then full acceptance.

36For example, in Vendetta (Konami 1991) homosexual characters are evil henchmen of the villain (Guardians) who rape the avatar of the player. But soon, in Streets of Rage 3 (AM7 1994) the gay character becomes the direct boss of a level (a sort of Half-Shadow). Only in Go Go Ackman 3 (Aspect 1995) is there an openly-gay final boss (Shadow). The 1vs1 beat'em up genre has been always full of gay characters, since the forefather Street Fighter 2 (Nishitani 1989), which presented a character, Zangief, who “likes fighting against polar bears and dancing with cossacks”, but he “dislikes nice girls”. We had to wait for Star Ocean 2 (Norimoto 1998) for an openly-gay playable character (although it is not a fixed script, but the player can choose the plot).

37Today most games have characters which are probably gay, but not openly so, in order to not 'disturb' the heterosexuals, which constitute a large proportion of buyers. Final Fantasy XIII (Toriyama 2009) has inspired fans to debate whether two of the main characters have a relationship or not. The steps toward acceptance have been small but substantial. In The Sims (Wright 2000) two homosexual characters can live together (“move in”); The Sims 2 (Wright 2004) introduces common-law couples (“joined union”), and The Sims 3 (Wright 2009) allows homosexual wedding (“get married”). Today in mainstream games like Fable 2 (Molyneux 2008) 10 % of non-playable characters are gay, as in real life.

38The Sims (Wright 2000) in particular has introduced the possibility to choose Sex, Orientation (by the enactment of a certain behaviour) and Gender (limited to the use of certain clothing) of the main character, the avatar used by the player. Today this tool is present in many American-style RPGs, which are focused more on the exploration, personalization and projection of the self into a new environment, in contrast to the Japanese-style RPGs, which are mainly focused on the plot and on the introjection of a new self (Mosca 2013). In general the relation between player identity and avatar identity can be structured around one of these axis: what the player feels, what social reality prescribes, or what the player is interested in trying.

Table 6. Axis of relationship between player identity and character identity


Symmetry between player feelings and avatar features (or behaviours).

Social Identity

Symmetry between common sense and avatar features (or behaviours).


Asymmetry between common sense (or player feelings) and avatar features (or behaviours).

39Following this scheme, in Fable II (Molyneux 2008) players can freely choose Sex, Orientation, and Gender, so that avatars can have sons, contract sexual diseases, marry and divorce. Few games are entirely dedicated to these themes. Rare examples are A Closed World (Stein 2011), an indie game that literally puts the avatar into the wood of doubts over Orientation, and Grand Theft Auto IV: The Ballad of Gay Tony (Houser 2009), whose title makes explicit that the main character is a homosexual-male-man (still non-playable: just a friend of the avatar).

Figure 6. Gay Tony

40It is possible to individuate some paradigmatic characters who represent different combinations of Sex-Orientation-Gender combintions:

Table 7. Sex-Orientation-Gender combinations in game characters

Sex does affect Orientation and Gender

World of Warcraft (Pardo 2004)

Sex does not affect Orientation and Gender

Birdo (Tanabe 1986)

Sex does not affect Orientation but it does affect Gender

Vamp (Murata 2008)

Sex does affect Orientation but it does not affect Gender

Bridget (Ishiwatari 2002)

Sex does not affect Orientation but Orientation does affect Gender

Poison (Yasuda 1989)

No Sex or Gender, just Orientation

Garrus (Hudson 2010)

41Let's examine them more closely. In Super Mario Bros. 2 (Tanabe 1986) a sort of pink dragon character, Birdo, is male 'but' he believes himself to be a girl, enjoys dressing as a girl and finally he likes males. In this case, phenotype does not affect Orientation or Gender. Birdo has evolved in many other sequels, alternatively becoming male woman, a female woman, and then returning back to its origins. In Metal Gear Solid 4 (Murata 2008) there is a non-playable character, Vamp, who is male, bisexual and man:

Figure 7. A real man (male and bisexual)

42In this case the phenotype does not affect Orientation and Orientation does not affect Gender. In Guilty Gear XX (Ishiwatari 2002) there is a playable character, Bridget, who is male, heterosexual and woman, so in this case (unlike the previous one) the phenotype does affect Orientation but Orientation does not affect Gender:

Figure 8. A real woman (male and heterosexual)

43The plot of Guilty Gear XX is interesting: Bridget has born male in a little town where the birth of twins of the same Sex is considered bad luck. His family named and raised him with a girl Gender. Becoming adult, Bridget decides to explore his Gender. During the course of the plot, he is deceived by his friends and many men do not understand that he is male: Bridget refuses their attentions and is attracted by a female homosexual woman. The Bridget character has widespread success among teenagers and he is one of the most played during cosplay gatherings, where only a handful of those using the mask of Bridget are male.

44Another famous case of variation of the traditional triad is Poison, a non-playable character of Final Fight (Yasuda 1989) with a female Sex and a woman Gender:

Figure 9. A real woman (male and homosexual) or a real woman (female and heterosexual)?

45In the porting of the coin-op for the Japanese version of the NES home console, because of Nintendo censorship that forbade representations of violence against females, the programmers simply changed the wording of the character description (maybe to avoid rewriting the code of the game) so that Poison was transformed from a female woman into a male woman. But the transsexualism of Poison was very bizarre. In the American version of the game for NES, the graphic design changed him into a male man, called Billy. So in the first coin-op version Poison was a female woman who, after censorship, became a transsexual: in this case the traditional triad was changed only due to lack of time to re-design the character. Sheff (1993, p. 225), however, gives another interpretation of the facts: for the first story-board described Poison as a 'newhalf' (the Japanese word for 'transsexual'):

Figure 10. Story Board

46Sheff's version seems confirmed by the fact that the Japanese subsequent versions of the character (in other consoles and sequels), use the original graphic design of the character and the transsexual description. Yoshinori Ono, the Capcom producer of the game, said in an interview that “in North America, Poison is officially a post-op transsexual. But in Japan, she simply tucks her business away to look female” (Leone 2008), i.e. a non-operated transsexual. This is a case where phenotype does not affect Orientation but Orientation does affect Gender: in America Poison is male, homosexual and woman, and in Japan Poison is female, heterosexual and woman. Ontological categories do not consider post-op people as transsexuals, because they are not variants of the traditional triad (having a female sex and a woman gender). Only a four-dimensionalist ontology, which consider entities in their relation with time, can consider a post-op person as a transsexual. But Social Ontology cannot be a four-dimensionalist ontology, otherwise the social objects which start to exist in time after a performative speech act as “I now pronounce you Man and Wife” or “I now pronounce you Male” in absence of documents could not be considered as 'objects' but only as 'occurrents' (Smith and Varzi 2000).

47Almost every game cited has been developed in Japan, where homosexual Orientation is considered acceptable only if it goes with transgenderism, which is considered as naturally linked to transsexualism. Indeed the change of Sex is a classic topic in J-Pop culture (e.g. Ranma ½), represented in all its variants. In Art of Fighting (Yasuda 1992), there is a transsexual female-to-male, whereas in King of Fighting XI (SNK 2005) there is a transsexual male-to-female. The circulation of J-Pop in Western culture has also incited American developers to explore the various possible combinations of Sex-Orientation-Gender configurations. In particular, a Canadian mainstream developer named BioWare seems very attracted to the Gender theme. In one of their recent market-breaker titles, Mass Effect (Hudson 2007), the player can choose Sex (which directly affects Gender) and Orientation (which does not). The avatar can have a flirt with an alien woman (with hermaphrodite primary Phenotype), changing Orientation if the player uses a male or female avatar.

Figure 11. A real woman (bisexual and hermaphrodite)

48In the sequel Mass Effect 2 (Hudson 2010), the player can flirt with another alien, but in this case the alien does not seem to be either male (or female), man (or woman). It is very big and strong, but there are no other signals about its Sex or Gender.

Figure 12. A real ungendered lover (bisexual and intersexual)


49In the recent past, the battle for the free choice of Sex, Orientation and Gender has used the constructivist argument that there is no objective property that can categorize our belonging to one type or to another. Carrie Hull has shown that this misunderstanding can be solved by dividing biological Sex from cultural Gender. Social ontology adds Orientation, so that these three institutional axes constitute a larger set of crossing categories applicable to the study of sexual identities in computer games, both for characters and for players.

50Computer game characters show a wide range of Sex-Orientation-Gender combinations, very well described by game studies, whose level of detail can be increased using a socio-ontological approach. Social ontology considers Sex identity, Orientation identity, and Gender identity as social objects constituted by a particular type of rules which combine both the descriptive and the prescriptive form (constitutive rules). Ontological analysis leads us to discover that there is a lack of research on the actual sexual identity of players, and this affects marketing.

51The ontological description of sexual categories can combine design and marketing instances, but above all can lead to a better understanding of the cultural evolution of players, which are in part mirrored by the cultural evolution of characters. Computer games show very well the contradictions in the conflict between the traditional triad (e.g. male-heterosexual-man) and a more liberal point of view. For example, today the majority of Western people accept diversity in sexual Orientation, but not diversity in Gender. So male characters (Sex) can be heterosexuals or homosexuals (Orientation), but only exceptionally women (Gender). Computer games present a more liberal approach to the topic than other mainstream media, such as cinema and television. Indeed games history shows a tendency to abandon the traditional triad in favour of the acceptance of new combinations of Sex, Orientation and Gender.

52The greater liberality of games is due to many factors: on the one hand computer games constitute a mature medium that only recently has become mainstream, so it is still influenced by underground culture. On the other hand computer games is a medium that for technological and industrial reasons has been strongly characterized by a double cultural influence, today completely merged: Japanese culture (in particular J-Pop), which legitimises the free representation of sexual categories, and American culture (Yankee), which emphasizes the free choice of everything that has been legitimated.

SSI Ludography

53Aspect Co. 1994 Go Go Ackman 3, Aspect Co, Banpresto: Japan, [SNES].

54Houser, D., Humphries, L., Garbut, A. 2009 Grand Theft Auto IV: The Ballad of Gay Tony, Rockstar North, Take Two Interactive: UK-USA, [XBox 360].

55Hudson, C., Watamaniuk, P., Karpyshyn, D. 2007 Mass Effect, BioWare: Canada, [Xbox 360].

56Hudson, C., Walters, M., Karpyshyn, D. 2010 Mass Effect 2, BioWare: Canada, [Xbox 360].

57Ishiwatari, D. 2002 Guilty Gear XX. The Midnight Carnival, Arc Systems Works, Sammy Studios: Japan, [Coin-Up].

58Iwatani, T. 1980 Pac-Man (Pakkuman), Namco: Japan, [Coin-Up].

59SNK 2005 King of Fighting XI, SNK, Sega: Japan, [Coin-Up].

60Macrae, D. and Curran, K. 1981 Ms. Pac-Man, General Computer Corporation, Midway Games: USA, [Coin-Up].

61Matsumoto, H. 1992 Art of Fighting, SNK: Japan, [NeoGeo].

62Molyneux, P. 2008 Fable II, Lionhead Studios: UK, [XBox 360].

63Murata, S., Kojima, H. 2008 Metal Gear Solid 4, Konami: Japan, [Playstation 3].

64Nishitani, A., Yasuda, A. 1991 Street Fighter II: Way of the Warrior, Capcom: Japan, [Coin-Up].

65Norimoto, M. 1998 Star Ocean 2: The Second Story, tri-Ace, Enix: Japan, [Playstation].

66Pardo, R., Kaplan, J., Chilton, T. 2004 World of Warcraft, Blizzard Entertainment: USA, [PC].

67Stein, A. 2011 A Closed World, MIT Gambit Lab: Singapore, [PC].

68AM7 1994 Streets of Rage 3, AM7, Sega: Japan, [Sega Mega Drive].

69SMA 1983 Stroker, Magic Carpet Software: Australia, [Commodore 64].

70Tanabe, K. 1986 Super Mario Bros. 2, Nintendo : Japan, [NES].

71Toriyama, M., Watanabe, D. 2009 Final Fantasy XIII, Square Enix: Japan, [Playstation 3].

72Konami 1991 Vendetta, Konami : Japan, [Coin-Up].

73Wright, W. 2000 The Sims, Maxis, Electronic Arts: USA, [PC].

74Wright, W. 2004 The Sims 2, Maxis, Electronic Arts: USA, [PC].

75Wright, W. 2009 The Sims 3, The Sims Studio, Electronic Arts: USA, [PC].

76Yasuda, A., Pon G, Nishitani, A. 1989 Final Fight, Capcom: Japan, [Coin-Up].

Haut de page


Aarseth, E. 1997, Cybertext: Persectives on Ergodic Literature. Baltimore: Hopkins University Press.

Acemoglu, D. and Robinson, J. 2012 Why Nations Fail: The Origins of Power, Prosperity, and Poverty, New York: Criwn Publishers.

Bogost, I. 2006 Unit Operations. An Approach to Videogame Criticism, Cambridge: MIT Press.

Butler, J. 1990 Gender Trouble: Feminism and the Subversion of Identity, New York: Routledge.

Caillois, R. 1958, Les jeux et les hommes. La masque et la vertige. Paris: Gallimard.

Borghini, A., Casetta, E. 2011 Tratti, sessi e generi, in “Bioetica. Rivista Interdisciplinare”, 1: 35-53.

Consalvo, M. 2003 It's a queer world after all: studying The Sims and sexuality, GLAAD center for the study of media and society. Http://

De Santis, L. 2012 Videogaymes. Omosessualità nei videogiochi tra rappresentazione e simulazione (1975-2009), Milan : Unicopli.

Diamond, J. 1997 Guns, Germs and Steel, New York: Norton Co.

Dreger, A.D. 2001 Top Ten Myths About Intersex, ISNA News, February: 3-5.

ESA (Entertainment Software Association) 2011 Essential Facts about Computer and Video Game Industry,

Fausto-Sterling, A. 1993 The Five Sexes: Why Male and Female are Not Enough, in “Sciences”, 33: 20-24.

Ferraris, M. 2012 Manifesto del nuovo realismo, Rome-Bari : Laterza.

Foucault, P.M. 1976-1984 Histoire de la sexualité, Paris : Gallimard.

Hull, C. 2006 Ontology of Sex. A critical inquiry into the deconstruction and reconstruction of categories, New York: Routledge.

Iser, W. 1976, Der Akt des Lesens. Theorie ästhetischer Wirkung. In The Act of Reading: A Theory of Aesthetic Response, Baltimore: Hopkins University Press, [1978].

Kripke, S. 1982, Wittgenstein on Rules and Private Language. Oxford: Basil Blackwell.

Latour, B. 2004 Why Has Critique Run out of Steam? From Matters of Fact to Matters of Concern, in “Critical Inquiry”, 30, 2: 225-248.

Latour, B. 2005 Reassembling the Social: An Introduction to Actor-Network Theory, Oxford: Oxford University Press.

Leone, M. 2008 The Big Comeback, in “Electronic Gaming Monthly”, 224: 68.

Moor, J.H. 1978, Three Myths of Computer Science, in “British Journal for the Philosophy of Science”, 29, 3: 213 – 222.

Mosca, I. 2013 The Social Ontology of Digital Games, in Agius, A. and Angelides, M. (Eds) Handbook of Digital Games, Hoboken: Wiley-IEEE.

Pearce, C. 2007 Communities of Play. Emergent Cultures in Multiplayer Games and Virtual Worlds, Cambridge: MIT Press.

Salen, K. and Zimmermann, E. 2003, Rules of Play: Game Design Fundamentals. Cambridge, MA: MIT Press.

Schwartz, S. and Schwartz, J. 1994 The Partent's Guide to Video Games, New York: Prima Lifestyles.

Searle, J.R. 1995 The Construction of Social Reality, New York: Free Press.

Sheff, D. 1993 Game Over Nintendo's battle to dominate an industry, London: Coronet Books.

Smith, B. and Varzi, A.C. 2000 Fiat and Bona Fide Boundaries, in “Philosophy and Phenomenological Research” 60, 200: 401–420.

Suits, B. 1978, The Grasshopper: Games, Life and Utopia. Peterborough: Broadview Press, [2005].

Suskind, R. (2004) Faith, Certainty and the Presidency of George W. Bush, in “The New York Times Magazine”, 10-17-2004.

Takahashi, R. 1987 Ranma ½, Weekly Shonen Sunday, Tokyo: Shogakukan.

Taylor, T.L. 2006 Becoming Player: Networks, Structured and Imagined Futures, in Proceedings of the “Beyond Barbie and Mortal Kombat: new perspectives on Gender, Games and Computing conference, NSF Workshop, Los Angeles: University of California.

Turing, A. 1950, Computing Machinery and Intelligence, in “Mind”, 59: 433-460.

Winn, J. and Heeter, C. 2009 Gaming, Gender, and Time: Who Makes Time to Play?, In “Sex Roles”, 61, 1-2: 1-13.

Wittgenstein, L. 1953, Philosophical investigations. London: Basil Blackwell.

Haut de page

Table des illustrations

Légende Figure 1. Epistemology, Ontology, Social Ontology
Fichier image/png, 32k
Légende Figure 2. Sex or Gender?
Fichier image/png, 17k
Légende Figure 3. Pac-Man
Fichier image/png, 79k
Légende Figure 4. Ms. Pac-Man
Fichier image/png, 41k
Légende Figure 5. Races and Sexes
Fichier image/png, 516k
Légende Figure 6. Gay Tony
Fichier image/png, 959k
Légende Figure 7. A real man (male and bisexual)
Fichier image/png, 470k
Légende Figure 8. A real woman (male and heterosexual)
Fichier image/png, 215k
Légende Figure 9. A real woman (male and homosexual) or a real woman (female and heterosexual)?
Fichier image/jpeg, 12k
Légende Figure 10. Story Board
Fichier image/jpeg, 60k
Légende Figure 11. A real woman (bisexual and hermaphrodite)
Fichier image/png, 566k
Légende Figure 12. A real ungendered lover (bisexual and intersexual)
Fichier image/jpeg, 24k
Haut de page

Pour citer cet article

Référence électronique

Ivan Mosca, « Ontology of Gender in Computer Games », Mise au point [En ligne], 6 | 2014, mis en ligne le 01 mai 2014, consulté le 24 mars 2017. URL : ; DOI : 10.4000/map.1631

Haut de page


Ivan Mosca

Researcher in Philosophy at the University of Torino, collaborating with the LabOnt Research Centre and Consulta di Bioetica Onlus. His fields are Social Ontology, Game Studies and Bioethics. He has recently published “From Fiction to Reality and Back. Ontology of Ludic Simulations” and “Ontologia della pirateria informatica”.

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus de la revue Mise au point sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page