By Anna Kreider
The Inspiration: Simon’s request and a 2007 LiveJournal Article
At GenCon this year, I was asked by Simon Rogers to write an article about sexism in gaming for a new issue of Page XX. It didn’t take me more than a few seconds to say, “Sure, okay”. But in pulling together notes on what I could write about, I realized that “sexism in gaming” is just too big a topic to cover well in one short article without focusing on only one aspect.
This realization made me think of an excellent article I had seen a few years ago on livejournal. Back in 2007, a blogger who goes by the handle Morgue on livejournal did a detailed analysis of the covers of the first 350 issues of Dragon Magazine and wrote an excellent series of posts about the results therein. It’s an excellent set of articles, and you can find them here: http://community.livejournal.com/gametime/19820.html
Briefly, Morgue found that Dragon covers displayed an overwhelming trend of sexist imagery on its covers. (Again, see the linked article for a detailed analysis.) I was curious to know if this was a trend that would be repeated in other areas of gaming, or if Dragon would prove to be anomalous, and this seemed like a good focus for this article.
I had a hunch that Dragon would prove to be pretty typical when examined in the context of gaming as a whole, but I wasn’t willing to trust that hunch since I’ll admit to being rather sensitive to sexist imagery in gaming material. Since I fully acknowledge my own bias, as a female gamer who loathes chainmail bikini porn, I had to set out some ground rules for myself.
The first problem I had to deal with was making sure that I examined images for a consistent set of variables. While the variables considered in the Dragon cover survey made perfect sense in that small context, they didn’t generalize too well to gaming outside of D&D. The other problem I found is that things like “focal” versus “non focal” and “submissive” versus “not submissive” were a bit too subjective for my tastes. Furthermore, there was a trend that I wanted to analyze that was not examined at all – the overall archetype breakdown by gender of depicted figures.
After some thought, I decided to track the following data points for each media source:
Ratio of male to female figures
In each set of images I examined, I recorded the number of male figures and the number of female figures. Since I wasn’t sure how to easily differentiate between focal and non-focal figures in a way that wasn’t entirely subjective, I simply counted each figure that had an easily discernable gender and did not count those figures where gender was ambiguous.
A specific note pertaining to D&D: While the handbook states that it is not easy to tell the difference between male and female Dragonborn, one of the figures on the Dragonborn page quite clearly has breasts. Furthermore, it is quite easy to find more images in official books such as the DMG and the PHB2 of Dragonborn with breasts. As such, I included all Dragonborn figures in my counts since, I considered the explicit gendering of the Dragonborn in the art to override the description of visual gender ambiguity in the text.
Active poses versus neutral poses
In Morgue’s survey, he classified poses as “active”, “neutral”, or “submissive”. This was a useful starting point, but was a bit too subjective for my needs. After some thought, I removed the “submissive” category and decided to simply look for active poses versus neutral poses:
Fully clothed and suggestively attired figures
In the original survey conducted by Morgue, the author purposefully took a conservative view of what constituted suggestive – counting figures only showing a moderate amount of skin as not suggestively attired. Morgue’s theory was that they wanted to err on the side of caution when evaluating images this way so as to eliminate possible bias. I largely took this view, with the caveat that any figure showing cleavage or not having covered legs was automatically counted as suggestive.
Fully-clothed and suggestively attired are not opposite ends on a spectrum. Some figures that were not fully clothed were not counted as suggestively attired while some figures that were fully clothed were also counted as suggestively attired. A few examples:
It was not always applicable, but when possible I looked at what class archetype a figure was depicted as: fighter, rogue, or mage. I counted all archers as rogues, as well as thieves. I counted anyone casting a spell as a mage, even if they had a sword. Fighters were any characters wielding only melee weapons and not casting spells.
Selection of Sources
I wanted to examine images from what I saw to be the major areas of gaming: tabletop roleplaying, CCGs, MMOs, and console games. The problem was determining where I could look to find images that would be representative of the hobby. Board games have not been considered as a segment of gaming, since – unlike with tabletop or CCGs – there is no one definitive game that all board gamers have played. As such, I’ve chosen not to include board games in this survey since I couldn’t figure out a way to include board gaming in a fair and representative manner.
Tabletop: For tabletop roleplaying, it was hard for me to find hard and fast numbers as to what is the most popular tabletop system in use, so I decided to simply look at 4th Edition D&D, since D&D is the franchise that spawned tabletop roleplaying to begin with as well as arguably having the largest name cachet.
I limited my examination to the Dungeon Master’s Guide, the Player’s Handbook, the Player’s Handbook 2, and the Adventurer’s Vault – since these are all books that one could reasonably expect a 4th Edition group to own. Because I was curious to see if the contents of the book would differ significantly from the promotional material, I also downloaded the D&D publicity kit from the Wizards of the Coast website and considered those images as well, though as part of a separate set from the art in the gaming texts.
CCGs: There is an overwhelming variety of CCGs released each year, so I stuck with the archetypal CCG that everyone knows and is familiar with – Magic. Picking Magic as my representative for CCGs complicated matters, since Magic periodically disallows old expansions from tournament use. Also, the nature of CCGs means that there is no base set that every player can reasonably be expected to have access to.
I decided to use the Magic publicity kit from the Wizards of the Coast website as my primary image set. I recognize that I will be unable to account for any difference between the promotional art Wizards selects for the publicity kit and art found on the cards themselves, but I had to make some sort of a compromise.
MMOs: Looking at art from MMOs posed a similar difficulty. The market for MMOs is very fragmented, and while World of Warcraft is very popular, there are many other different types of MMO out there. Also, there are many Asian-originated MMOs that are very popular, like Lineage II, Final Fantasy XI, and Final Fantasy XIV. Trying to analyze sexism in games created by non-Western cultures presented a challenge, so I settled for looking only at art from the top 7 North American MMOs which command the lion’s share of the market: World of WarCraft, Guild Wars, EVE Online, City of Heroes/City of Villains, EverQuest/Everquest II, Warhammer Online, and Dark Age of Camelot.
Since there is such a variety of armor and model options within games, I chose instead to look at promotional and concept art posted on each game’s official website. (Interestingly, this criteria immediately removed two games from the list: EVE Online, which has no pictures of characters at all on their website, and Dark Age of Camelot – which features models who wear exactly the same armor regardless of sex and equally represents both men and women.)
Consoles: For console games, I wanted to look at the top selling console games within a given period. As such, I looked at the cover art of games on the lists of top games for the Xbox 360, PS3, and Wii on Gamespot.com.
Now, I think I’ve talked enough. Onto…
Dungeons and Dragons
As is clearly shown in both graphs, both the art in the 4th Edition books and in the promotional kit displays sexist trends. Women comprise only a little more than a third of the figures depicted and are more likely to be depicted as neutral than their male counterparts. Also, in instances where a figure is suggestively attired, the odds are overwhelming that that figure will be female. However, it should be noted that there was an encouraging diversity of class archetypes among female figures with depictions of fighters being almost evenly split between men and women.
Interestingly, the art in the publicity kit is marginally more sexist. Across every category, the art in the publicity kit is more sexist than in the game books. The difference is small, often only a few percentage points, but consistent. So it seems reasonable to posit that Wizards is taking a conscious “sex sells” stance in regard to its marketing, even if they don’t allow it to skew their art too much away from what is in the game books.
(As the class archetype does not apply, this data was not included.) The art in the Magic publicity kit also displays sexist trends. As with the D&D publicity kit, women comprise little more than a third of the figures. The Magic Fankit has about the same number of female figures that are fully clothed, and even has a lower proportion of suggestively attired women – though women still account for the great majority of suggestively attired figures.
However, the art in the Magic publicity kit also depicts women as quite a bit more passive, with a significantly lower instance of active female figures and a higher instance of neutral female figures.
MMOs (WoW, GW, Warhammer, CoH/CoV, EQ/EQII)
When you examine the data from all five MMOs together, the trends don’t look too different from those of the D&D promotional kit, though the D&D promotional art is demonstrably less sexist. Women make up slightly less than one third of the figures, and are less likely to be depicted with active poses as female figures in the D&D promotional kit. There is, however, pretty much the same proportion of suggestively attired female figures, which is to say that the majority of suggestively attired figures are female.
The difference, however, is in the instance of fully clothed figures and in class archetype. While the proportion of suggestively attired figures is similar to that of the Magic and D&D artwork, the instance of fully clothed female figures is actually lower. Also, women are overwhelmingly likely to be depicted either as a mage or a thief archetype rather than as a melee fighter. (This obviously does not include CoH or CoV, since they do not ascribe to traditional fantasy archetypes for classes in their artwork.)
However, there are some interesting things that happen when you consider the data from each MMO separately:
World of Warcraft is fairly similar to the above graph, though the female figures in their artwork come close to approaching parity with male figures when it comes to active poses. The proportion of female to male figures is lower, however, at only 27% of all figures counted. And again, across all categories, WoW’s art is more sexist than the D&D art.
Again, suggestively clad figures are overwhelmingly female, and female figures are overwhelmingly likely to be depicted as thieves and mages rather than fighters. Furthermore, among female figures that were depicted as fighters, not a single one counted as fully clothed, with all but one counting as suggestively attired.
An interesting thing happens with the Guild Wars artwork. Female figures actually account for more than half of the figures counted, clocking in at a whopping 55%. However, the female figures depicted are also more sexist. Like women in WoW’s art, female figures are most likely to be depicted as thieves or mages. However, they are twice as likely to be depicted in neutral poses as active ones. Furthermore, 95% of suggestively attired figures are female. So while women get more screen time than men in Guild Wars’ art, they are also hyper-sexualized in a way that men are not.
Women are a bit of an afterthought in Warhammer Online’s art, it seems, which is a bit strange considering that it is published by Mythic – the same publisher responsible for Dark Age of Camelot. Although there were no poses that counted as neutral in any image, the only mage depicted out of eighteen figures was female, and both of the female figures depicted were suggestively attired. So when the artists of Warhammer Online remember about women, they only think to draw them in suggestive ways, it seems.
Unlike the other MMOs, the sample size of promotional art on Warhammer Online’s website is very small. However, there is sufficient supporting material on their website that makes me think that this data is still representative. In particular, there are two classes of Dark Elves that stand out: the Witch Elf and the Sorceress. Both classes are female only and both classes wear only pretty classic chainmail bikinis. Also, when looking at the class information on the Warhammer site, only three of sixteen classes (not counting dwarves or greenskins) are depicted as female, with two of those depictions being half-naked Witch Elves and Sorceresses.
For City of Heroes, the criteria had to be evaluated slightly differently, since everybody wears spandex. Fully covered was considered to be any character whose costume didn’t show skin. Suggestively attired was considered to be figures who costumes did not cover suggestive areas – cleavage, thighs, etc.
In City of Heroes artwork, women again punch in at around a third of the figures overall. Interestingly, they are actually slightly more likely to be depicted as active than male figures. Also, they are more likely to be depicted as fully covered and slightly less likely to be suggestively attired, though the trend holds true that the clear majority of suggestively attired figures are female.
Everquest does better than any other MMO except for Guild Wars in terms of its percentage of figures being female – women account for 43% of the figures counted. However, they score at the bottom of the pack in almost every other aspect. Female figures are twice as likely to be depicted as neutral than as active. More than 80% of mages were depicted as females, with less than 20% of fighters being females. And again, women make up close to 90% of all suggestively clad figures while accounting for none of the figures counted as fully clothed.
Overall Ranking of MMOs
If I had to rank them, I’d say that City of Heroes/City of Villains and World of Warcraft come out on top, Everquest/Everquest II and Warhammer Online are at the bottom of the heap, and Guild Wars fits in somewhere in the middle.
Personally, this result is irritating since the difficulty of finding pants that are actually pants and not bikini bottoms for my female WoW characters has been a pet peeve of mine for many years now.
Consoles: cover art of Top Games in a given week, as rated by Gamespot.com
When comparing the three top consoles, it’s not surprising that the cover art for Wii games is far more family-friendly than those of other consoles, with only one male figure and one female figure counting as suggestively attired. The Wii had the highest proportion of male to female figures as well, with female figures making up 36% of the total. However, there was not a single female figure counted as actively posed, which is a bit dismaying when compared with the 42% of male figures that were actively posed.
I had expected to find that the art for Xbox 360 and PS3 games would be largely similar, since many of the same games appear on both lists, but I was surprised to find that the opposite was true – owing to different covers for games released on both platforms.
On Xbox 360 covers, women made up only 25% of figures, as compared to 32% on PS3 covers. However, on the Xbox covers, women actually had a higher percentage of active poses than male figures, while on PS3 covers female figures constituted only around a third of all actively posed figures. Women also made up around 70% of suggestively attired figures on Xbox 360 covers, whereas they comprised almost 90% of suggestively attired figures on PS3. So while women were depicted more on PS3 game covers, those depictions were also more sexist.
Conclusions and Personal Impressions
It is with some amusement that I note that my perceptions did not align with reality in regards to the results I expected. Having been a long-time D&D player, I perceived 4E to be less sexist in its imagery than it actually was by the numbers, simply I had played both 3rd Edition and 3.5, which I had seen as so much more sexist by comparison.
Conversely, my experiences with playing female characters on WoW for several years had led me to believe that WoW would come out on top as the MMO with the most sexist imagery. I was quite chagrined to discover that the opposite was true and that most of the other top MMOs were actually more sexist in their imagery than WoW. Even though EVE Online and DAoC exempted themselves from my survey, both games together still comprise less than 5% of WoW’s current membership. As such, I’m forced to consider them as an anomaly. Dark Age of Camelot is sadly a small island of gender parity in an ocean of chainmail bikinis, while EVE Online is just a deserted island.
I have to say that doing the research for this project was pretty enlightening, although I can’t say that it was fun. As a female gamer, there are a lot of things that I turn a blind eye to out of love for gaming. I always tell myself getting angry all the time about the sexist imagery used in gaming wouldn’t do much besides make me miserable. But forcing myself to look at each image objectively, or at least as objectively as I could manage, and really critically examine the imagery that is being used…
I’ll be honest. It’s been disheartening. It’s hard to look at the endless parade of cleavage and not equate those images with myself. It’s hard not to ask myself – is this how they see me? Is my worth to the gaming community judged on how attractive I am, or how willing I am to dress suggestively? Am I valued only as a pair of breasts and not as an artist, gamer, and designer?
I’m not saying that gamers are all sexist, or that gaming as a hobby is terrible, or that the companies who use these images in their products are evil, or that the artists who create the art are bad people. Don’t get me wrong here, my intent is not to point fingers. If you’re reading this article, then I’m sure that you’re a very enlightened and egalitarian person. I’m simply trying to put forward this issue in such a way that might make people think about the implications of using demonstrably sexist imagery in gaming might be. There are certainly unintended consequences of these sexist depictions that are rampant in our hobby that I don’t think a lot of people stop to examine.
(And yes, this is hardly a scientific survey, and I’m not a professional researcher, and D&D might not representative of all tabletop games, and, and… I’m sure there’s a thousand reasons one could find to entirely discount this, but that’s hardly my problem.)
Lastly, I’ll state that sexism in gaming is such a monstrous topic that I haven’t even begun to scratch the surface with my little survey here. If you’re interested in doing more reading on your own, you can check out this post on my blog with the list of links I made for myself while researching this article.
[Editor's note - Anna is analysing comments on this article over on this entry on her sexism in gaming blog. ]