by Nick Yee
The female players in these narratives make clear that female avatars are disturbing to them not only because of the sexual exaggeration but also because the avatars are a constant reminder that they have stumbled into some sort of digital peep show. As game designer Sheri Graner Ray puts it, female avatars are designed “as male players would like them to be—young, fertile, and always ready for sex.”11
One common male defense to this argument is that male avatars are exaggerated, too, and thus that the unfair treatment is equal on both sides. Although it is true that male avatars are exaggerated, it is in a very different way. In female avatars, the exaggeration tends to be sexual in nature—large busts, low-cut clothing, sheer or almost nonexistent pants. In male avatars, the exaggeration tends to center on strength or athleticism, not sexual features. Male avatars have unrealistically big muscles, yes, but we don’t see male avatars with prominent bulges in their pants or tight stripper shorts that reveal the top quarter of their behinds, and most of their pants selections do not consist of briefs and thongs. In fact, as one female player noted, there is a very definite bulge problem in male avatars.
I find it somewhat disturbing that while the female avatars all have very prominent breasts, none of the male avatars have anything visible at the crotch at all. Their clothes are cut in a male style, but there’s no “bulge” where things ought to be. If female avatars are made to approach some “ideally attractive” or “sex-specific” model, then male avatars ought to be the same, rather than being de-sexed. [World of Warcraft, female, 31]
Thus, even though both male and female avatars are distorted, very different features are being exaggerated. And if you are a heterosexual male gamer who is made even the slightest bit uncomfortable reading about prominent male bulges in thongs, then perhaps that gives a hint of what women gamers feel when they play a female avatar in most online games.
The sexualized female avatars also encourage sexual harassment. The immediate visual of a scantily clad, voluptuous female is the only point of reference available in the game of who another person is and makes it easy to attract unwanted attention.
One thing that pushes many women away (or in my case into playing a male character) is the ogling and cat-calls that can go on in games. I was astounded that people would hit on a cartoon in a lewd manner. [EverQuest II, female, 42]
Sexual harassment is something that many male players experience for the first time when they play a character of the opposite gender in an online game.
I never realized how irritating it can be to have to put up with unwanted advances. [EverQuest, male, 38]
I’m amazed how thoughtless some people can be, how amazingly inept men are at flirting and starting a conversation with a female, and how it really does take more effort to be taken seriously as a female versus a male. [EverQuest, male, 24]
The same piece of armor often looks different when it is worn by a male avatar compared with when it is worn by a female avatar. Not only is the armor tailored differently for the different body silhouettes, but in the case of many high-level or rare pieces of armor, the cutting and design may be entirely different as well. More often than not, male armor pieces tend to cover the body, whereas female armor pieces tend to reveal the body. Oddly, this differentiation tends to intensify with the increased rarity or higher level of the armor. On the official forums for Diablo III, one female gamer articulates her concern over her high-level wizard armor:
I realize this is a controversial topic among many gamers—particularly male gamers who want female characters to look like prostitutes regardless of how the women who play them feel—but I’ve come here to post about the level 63 pants on a female wizard. . . . The pants are “crotchless” and have no thighs to them; they’re essentially high-riding underwear with no pant around them but on the outer legs. I think I’d prefer just to wear underwear at that point. . . . I want to be a wizard, not a pornstar. . . . I don’t even really mind the highly impractical stilettos on my demon hunter or the crazy cleavage on some of her cloaks, but Archon Faulds are just ridiculous sex shop attire.12
Although this asymmetrical design has been in place for more than a decade—it has almost become an inside joke among gamers—there is an insidious logic embedded in this asymmetrical armor progression. As male characters level up and become more powerful, their bodies become better protected and covered. In contrast, as female characters level up and become more powerful, their bodies are uncovered and made more vulnerable. Thus, as women gain power, they are disempowered in another way. Unfortunately, this is the only logical endpoint of a male fantasy that uses female bodies as sexual objects to be controlled. Whether it is Firiona Vie chained and held at swordpoint or a powerful female wizard in crotchless pants—two visual references that bookend roughly twelve years of online games—the message is that all women, no matter how successful, are vulnerable sexual objects.
Perhaps the greatest irony of this male fantasy is that women are simultaneously highly desired and shunned. Idealized female body parts are put on display and ogled, but the moment a real woman steps into an online game, her presence is deemed suspect and her body parts are questioned. Women are worshipped and idolized as long as they are not real; it is in this sense that online games reveal their function as a male fantasy. And perhaps another reason why games deny women access is because the male fantasy can be sustained only by presuming a male audience.
A Wrinkle in the Numbers
One caveat in this male territory discussion is that studies have consistently identified statistical differences between male and female gamers. In particular, many studies have found that women are less interested in the achievement and competitive aspects of games than men are. And in the interest of full disclosure, I’ve found and reported these findings in my own research. These data seem to suggest that a gender partitioning of games or game genres is sensible. But there’s actually a wrinkle in these data.13
In psychological research, two samples are “significantly” different from each other if the difference between them is greater than chance alone would predict. In psychology journal papers (and those in related social science fields using quantitative methods), the term significant is reserved for this statistical meaning; however, whether the difference is substantively meaningful is an entirely different issue. This is because the ability to detect differences is a direct function of the number of participants in a study. In studies with large samples, even very small differences can become statistically different. For example, in one survey of over a thousand World of Warcraft players, I found a statistically significant difference between players of the two factions—players who prefer Horde were on average 27.5 years old, whereas players who prefer Alliance were on average 28.7 years old—but this marginal age difference isn’t very meaningful. Thus, you might have a statistically significant difference that is substantively trivial.14
We can also turn the statistics of difference on their head and instead calculate the statistics of similarity. In my own data on gameplay motivations, the largest difference was found in the mechanics motivation (that is, rules and optimization), but even here the overlap between men and women was 67 percent. The average gender overlap across all the gaming motivations listed in chapter 2 was 82 percent. The findings from another study of EverQuest II players show a similar pattern. Even though the difference in terms of how strongly men and women were motivated by achievement in online games was significant, the overlap was 70 percent. What these numbers show, when we look at statistical overlap instead of statistical difference, is that the majority of male and female players in online games actually like the same kinds of play. In fact, psychologist Janet Shibley Hyde has found this same pattern of gender similarity across a broad range of psychological variables outside of gameplaying. When we look at gender similarities instead of gender differences, we find that claims of dramatic differences between men and women are often inflated. Attempting to identify gaming motivations that appeal
to the “female brain” might be attempting to solve a problem that doesn’t really exist.15
Not only are the gender overlaps large, but the gender differences are actually inflated at the outset. One danger of studying gaming populations (or any natural community) is that of bringing underlying biases into the data. The gender difference in achievement motivations is a good case in point. It turns out that age influences the achievement motivation more than gender; older players are much less interested in goals and competition in online games compared with younger players. In fact, the relation between age and the achievement motivation dwarfs the gender difference. Age explains almost twice the statistical variance in this motivation compared to gender. It also turns out that, on average, women in online games are older than men by almost six years. Thus, when researchers compare the superficial gender difference without factoring in the age difference between men and women, they exaggerate the observed difference in the achievement motivation.16
Moreover, women gamers are perfectly capable of saying what they want and don’t want in a game.
I think that by marketing specific games for women, game companies are patronizing women and missing the point of the problem entirely. They don’t need to make games specifically for women (games which usually involve shopping, or other stereotypically feminine things). They need to make current games less sexist so more women will be interested in playing them. Lots of women enjoy MMOs, FPSs [first-person shooters], and other popular kinds of games. But when these games give them the message that women are best when they’re all T&A [tits and ass] all the time, they are not appealing. [World of Warcraft, female, 27]
There isn’t really a concept of a “game for boys”: men I know play everything from Japanese RPGs to tactical simulations to FPSs, yet women are expected to ALL play The Sims, as if there aren’t just as wide a variety of tastes amongst women as there are amongst men. [Dungeons and Dragons Online, female, 26]
As we’ve seen here, although there are numbers that show gender differences in gaming motivations, these statistical differences are often not as straightforward as they may appear.
Gender-Bending and Gendered Bodies
We’ve seen how gender expectations are encoded into and sustained as male technological fantasies in both the design of avatars and the powerful rhetoric, which allow women into these spaces but as second-class citizens and sex objects. By allowing players to have fluid bodies, virtual worlds actually go one step further. These online games are uniquely powerful tools for perpetuating stereotypes.
On its surface, gender-bending seems like the perfect counterpoint to my argument that these games provide a false freedom. At the click of a button, men and women can switch their biological sex and experience life from an entirely different perspective. What could be more liberating than that? Indeed, among World of Warcraft players, 29 percent of men report having a main character that is female. On the other hand, only 8 percent of women report a main character that is male. When looking across all the characters that a player has, the same pattern remains. Fifty-three percent of men have at least one character that is female, whereas only 19 percent of women have at least one male character. On average, 33.4 percent of men’s characters are female, whereas only 9 percent of women’s characters are male. No matter how we slice it, men gender-bend roughly three to four times more often than women. On the Daedalus Project, my post on this gender disparity in gender-bending has elicited more than two hundred comments from players trying to explain this phenomenon. By far, the most widely adopted male explanation is that the third-person perspective in these games means that players spent a great deal of time looking at the back of their character. As one male player put it, “If I am going to stare at a butt all game it might as well be a butt I’d like to look at.” Data my colleagues and I gathered at the Palo Alto Research Center provide empirical support for this. When men gender-bend in World of Warcraft, they tend to play races with attractive female characters: Humans, the Draenei, and Blood Elves. Races with unattractive females—the diminutive dwarves or the giant, muscular, cowlike Tauren—are seldom selected by men who gender-bend. In short, gender-bending among men is often an artifact of the sexualized female avatars, rather than an explicit attempt to explore gender roles.17
Virtual worlds also allow false gender stereotypes to be made true. In a study of World of Warcraft players in which my team combined survey data with in-game log data, we first asked players whether they thought male or female players preferred certain activities in the game. These included different combat roles, such as healing, tanking, and damage-dealing classes, as well as noncombat activities such as crafting. The most strongly stereotyped female game activity by far was healing; players believed that women have a much stronger preference for healing compared with men. We found that this was not the case. We calculated the ratio of total healing output compared with total damage output for each player in the study. This healing ratio allowed us to get around the noise of some players playing more hours a week than other players and get to a comparable measure of healing preference. We found that male and female players had almost exactly the same healing ratios—33 percent for men, and 30 percent for women. Thus, the stereotype that women prefer to heal in online games is false. Men and women have the same preferences for healing.
Where we did find a statistical difference was in character gender. Female characters had a much higher healing ratio compared with male characters. This disparity was a direct consequence of how players behave when they gender-bend. When men gender-bend and play female characters, they spend more time healing. And when women gender-bend and play male characters, they spend less time healing. In other words, when players in World of Warcraft genderbend, they enact the expected gender roles of their characters. As players conform to gender stereotypes, what was false becomes true. Thus, when players interact in the game, they experience a world in which women prefer to heal.18
Jesse Fox, a communication scholar at the Ohio State University, has found that the design of female avatars can elicit dangerous stereotypes. In her study, students were put into a virtual world and interacted with a variety of female avatars. She found that female avatars that conformed to gender stereotypes—a coy, conservatively dressed women or a dominant, suggestively dressed women—increased sexist beliefs and rape myth acceptance. In short, participants exposed to scantily clad female avatars were more likely to believe that women who get raped deserve it because of their perceived promiscuity. In virtual worlds, false stereotypes are being made true via play.19
Rethinking Utopias
The things we build often become contaminated with unspoken preconceptions and prejudices, whether this is a technological artifact or something as mundane as a road:
Anyone who has traveled the highways of America and has become used to the normal height of overpasses may well find something a little odd about some of the bridges over the parkways on Long Island, New York. Many of the overpasses are extraordinarily low, having as little as nine feet of clearance at the curb. Even those who happened to notice this structural peculiarity would not be inclined to attach any special meaning to it. In our accustomed way of looking at things like roads and bridges we see the details of form as innocuous, and seldom give them a second thought.
The highway developer Robert Moses designed these low overpasses for the same reason that he vetoed a proposed railroad to Long Island: his decisions ensured that the twelve-foot-high buses used by the lower class would be unable to navigate the parkways, reserving Long Island’s beaches for white, privileged New Yorkers. Political theorist Langdon Winner argues that human artifacts embody politics—the things we build can implicitly regulate who does and does not belong.20
The fantasy worlds we build also have these unspoken rules. From the moment a woman steps into a gaming store or enters an online game, she receives cues that she doesn’t belong. Virtual worlds and online games are crafted by people with certain mindsets and
biases that are too often hidden and unquestioned. As with the color of our skin or the national boundaries we happen to fall into, our offline biological sex still matters in virtual worlds. In this instance of the Proteus Paradox, false beliefs and stereotypes of women are not only being perpetuated in virtual worlds, they are being made true via play.
I’ve focused on online games in this chapter, but sexism isn’t a gaming problem; it’s a social problem. Although having more female game designers would likely lead to more gender-inclusive games, the dramatic statistics on awarded degrees in the computer sciences illustrate that as a society we have very different expectations of what careers men and women should pursue. In addition to career stereotypes, there are also striking differences in how men and women think of their free time. Studies have consistently shown that women have less free time and that their free time is more likely to be infringed on by gendered expectations of housework and child care. Because these gender-stereotyped household responsibilities (some ever-present, others unpredictable) are co-located with leisure spaces in the home, women often feel less relaxed and more pressured even when they are ostensibly free. Thus, women are more likely to experience guilt when they engage in leisure activities in the home. This conflicted sense of leisure is exploited quite effectively by advertisers. From body lotion to chocolates, from yogurt to spa treatments, products are often marketed to women as guilt-free indulgences—that just this once, they can indulge in something special without feeling guilty about it. Advertisements for men almost never employ guilt. But this trope reveals an important social message: women are normally expected to feel guilty about leisure and pleasure. The stereotype of gaming as a waste of time likely exacerbates this expected guilt and further lowers women’s desire to game.21
