Hello AWN friends,
I was wondering if you all could throw in your two cents on a topic I'm studying for a research paper! I'm learning about the way that character designers give "gender" to non-human characters.
In other words, how do you make a non-human character look like a boy or a girl?
Do some methods seem more legitimate to you than others? (For instance, just slapping a bow on on a buffalo's head to make it female might seem lazy...or not-- it could be "ironic" haha)
I've also noticed that in a lot of cartoons, only the female animal characters have distinguishing "gender" markers, (like eyelashes), whereas if there aren't any gender markers, people automatically assume the character is male.
Does this make designing female characters different, since they carry an obligation to differentiate themselves as feminine? Or is it pretty much the same?
Thanks for your input! I should have wrote sooner, since the paper is due in four days, but I thought I'd ask anyway.
Happy Memorial Weekend :)
T. King
The most subtle and effective way to my mind is translating the more delicate, curvaceous and slender forms of the human female's body to a different kind of anatomy. It's an elusive thing and all but impossible to put into a formula.
My advice is to have a close look at, for example, Pongo and Perdita in 101 Dalmatians. Perdita doesn't have many of the cliché characteristics of cartoon females, like pronounced eyelashes or lips that look like painted on, and her general construction is very similar to Pongo's. Yet by applying subtle qualities of the "standard" human female's built to her dog anatomy, and also animation, it becomes quite clear she is female.
So in short, it's about built, posture and mannerisms. It's very easy to get caught in clichés when trying to draw and animate females. I sure got the hang of drawing males earlier. It might be worth to ask among women animators and designers whether they have or had an easier time learning to draw male or female characters.
My theory is that our understanding of gender and our feeling comfortable about drawing either males or females is directly related to our own gender and ways we were brought up feeling about the other.
Do some research on Dr. Stuart Sumida He's worked with all the major studios on character & gender designs. Think link isn't the greatest but you'll get a sense of who he is and what he does. Check out youtube for his videos.
Keep in mind, it isn't just "genderizing" but adding human characteristics as well.
Some other characteristics - leg length (women have longer legs). Check out Tarzan for short/male legs.
Jaw line - females tend to have softer jaw lines (check out Pocahontas for a female with a square jaw - done on purpose to differentiate her from other Disney female characters).
Shoulders - wider in males
hips - wider in females.
I have a pet peeve in todays 3D animation female characters. IMO they all look alike.
Hey, thanks a lot!
To Jabberwocky:
I'd never noticed those differences in Pongo and Perdita's designs. That's quite ingenious.
I wonder how many physical cartoon gender traits are conventions that we've just learned to read by exposure, and how many are actually inherent to the way we read gender in real life? (Like, most girl characters have bigger eyes, whereas I'm pretty sure that's not true in real life).
I wonder where these conventions came from? Aah! Too many questions for a paper that's due in just 2 days :)
Also, even though I'm a woman, I've also found genderizing female characters to be harder. I think part of the problem is that characters tend to read as male by default, and trying to add feminine features without falling back on stereotypes is tricky. The other part of the problem is--I need to take more life drawing classes!
To b'ini:
Thanks so much for this link! I wish I'd found out about this guy sooner. I had no idea such people existed.
Thank you both for your thoughtful answers!
T. King
That I'd like to read. May you publish it or pass it on to others?
Sure Jabberwocky!
I'm not entirely satisfied with it though--I think that the class placed way too much emphasis on making a grand statement about society, and I think this pressure ended up skewing my paper a lot. (Though I care about the topic, I really wasn't as interested in making a huge social statement as I was in learning about the gendering techniques character designers use).
I'm not sure if I actually agree with some of the conclusions I drew. (For instance, I know I wrote off Christmas Carol and Polar Express way too quickly). Also, I didn't have very much input from actual character designers, which would have helped.
But still, I definitely learned a lot, which was the point I guess. Yay college!
My paper is pasted below.
Thanks again for your help.
T. King
Tamarind King
Donna Hunter
PWR 2-Section 2
6/5/10
RBA Final Draft
Female identity in non-human character design
A few years ago I was working on an animated project called ESM. It was a short film set to music featuring three animated characters—a little boy, a rooster, and a burro. Late one night I got a call from P.G., a friend of mine and director of the project. He wanted to talk about my character designs. The kid was okay. The rooster was okay. But when we started discussing the burro, we hit problems.
“He looks okay,” P.G. said.
“Wait—he? I thought the burro was a girl.”
“He doesn't look like a girl, the way you've drawn him.”
“I mean, she looks like a burro. All burros look like that.”
My friend went on to explain to me that if I wanted the burro to be a girl, I'd have to make her gender more obvious. “Put a flower behind her ear,” he suggested helpfully. “And add some eyelashes.”
The conversation bothered me, though I wasn't quite sure why at the time. It seemed unfair to me that a boy burro could just be a burro, while a girl burro had to distinguish herself as being distinctly feminine. Not that I had any problem with distinctly feminine characters. My favorite childhood characters tended to be girl animals from the Disney films—Nala from The Lion King, and Rita from Oliver and Company. Instead, I was disturbed that the burden of gender distinction seemed to always land on female characters. Why couldn't I design the burro to look however I wanted? It also bothered me that I was expected to gender her with clichéd markers like flowers and eyelashes. Because I didn't want to rely on such gags, I was faced with a second challenge. How could I depict gender without relying on old stereotypes of what is masculine and what is feminine? Are there more objective or accurate ways of depicting gender? Where can character designers find material for such an objective depiction? Sex recognition studies? Motion capture technology? Personal observation?
Though there's been a recent push for more realistic gender representations, it's possible that there is no objective way to gender characters, especially non-human ones. At the same time, designers face the challenge of gendering characters without falling back on old tropes. While outsiders criticize the stereotypes that persist in modern cartoons, many artists argue that character design should be a matter of creative freedom, and that policing by feminist critics will stifle artistic creativity and only result in more token girls and bland “role model” characters.
This paper will specifically explore the problems surrounding the depiction of femininity in non-human animated characters, and attempt to address the issues mentioned above. There is already plenty of writing on gender roles in cartoons, so I won't be addressing that in this paper. Instead, this paper will focus on the topic of gender in character design (the visual aspects of a character). The characters that we'll be analyzing are all female, non-human, animated characters. There are reasons for being so specific. Since characters are assumed to be male unless gender markers are added to denote otherwise, designers pay special attention to gendering female characters. Thus, this topic is especially salient to females. Animated characters are good subjects for analysis because they are completely synthetic, and provide clearer indicators of how gender is imagined and constructed. Finally, non-human characters are important to explore because gendering a non-human character requires that designers strip gender down to what they consider its most basic qualities, then map those qualities onto non-human bodies.
One might wonder—if gendering in cartoons is so controversial, then why designers bother to gender non-human characters in the first place? In the past, animators have chosen to use non-human characters in order to get around the issue of not having different ethnicities properly represented. So why can't animators use androgynous non-human characters to get around gender representation conflicts as well?
It actually comes down to audience identification, defined by psychologist Eleanor Maccoby as “the process by which a viewer shares a character's perspective and vicariously participates in his/her experiences during the program” (Hoffner). Designers gender nonhuman characters for the same reason that they anthropomorphize them—in order to make them more relatable to a human audience. With real animals, people don't usually care about gender. However, most animals in cartoons (especially the talking ones) are meant to be related to as people. Because gender is such an important part of the way people relate to one another, it becomes an important part of non-human character identity as well.1
Which brings us to the challenge. How do designers get audiences to read gender in non-human characters?
One method is to use what media analysis website Tvtropes.org calls tertiary sexual characteristics.2 Tertiary sexual characteristics refer to the costume gags, such as clothing and props, that designers often use as shorthand to denote gender. (Minnie Mouse's bow and dress, for example). However, while tertiaries are quick and efficient gender indicators, they still often rely on extremely shallow stereotypes of femininity and masculinity (for instance, all girls wear bows, and only boys smoke cigars). There's also the obvious incongruence of male and female characters who look exactly alike, save a few costume differences (again, looking at Minnie), which suggests that feminine identity needn't be very carefully thought out when there's a male counterpart to compare it to.
A second gendering strategy that designers use involves adding human gender traits to a non-human character's body (for instance, adding the curve of hips or bust to a squirrel). This is the most common form of gendering for highly anthropomorphized animals, like those in PBS's Arthur, and Disney's Robin Hood—but it's arguable that these characters are really just people with animal heads. Anthropomorphic gendering is far more subtle for characters that are less human. For instance, Disney's 101 Dalmations features dogs that look like dogs, but are given subtle, human gender clues. Pongo, the male, has more pronounced shoulders, a squarer snout, and straighter lines than his mate Perdita. In contrast, Perdita has a more delicate chin, rounder shoulders, and a more curved, slender body. However, though anthropomorphic gendering is more subtle than tertiary gendering, it still risks falling into stereotype. After all--not all women are curved and slender with delicate chins.
So far we've looked at the two of the most common methods of gendering—through tertiary sexual characteristics, and through anthropomorphism. However, both of these methods rely on the designer's imagination of what is feminine or masculine, and are thus subject to stereotypes and the designer's personal biases.
How then can character designers find more objective ways of portraying gender? Gender recognition studies and motion capture present one possible solution. In 1977, Lynn Kozlowski and James Cutting conducted a study testing participants' ability to recognize the sex of a person walking by in the dark with small lights affixed to his or her major joints. While participants were able to guess the sex of the walkers at a rate higher than 50-50 chance (men were correctly identified 72% of the time, and women 67%), the researchers concluded that, “Walking appears to be a wholistic act. Information as abstract as the sex of the walker appears to be distributed through all the movement.” (Kozlowski and Cutting). In other words, there is no one clue that people go by to determine sex in a walker. This suggests that even though movement does carry sex information, it would be difficult to condense that information to a single trait which could be applied to cartoon characters.
Besides the difficulty that the objective approach presents, there's also the argument that people's willingness to identify with a character doesn't depend on the realism of the design, but on its believability. Realism refers to how closely a character resembles reality. Believability refers to how easy or difficult it is to relate to the character as an actual person. In cartoons especially, believability is completely independent from realism. Few people would argue that Bugs Bunny looks like a real rabbit, or that Elmer Fudd looks like a real man, but that doesn't keep the audience from suspending their disbelief and engaging with the characters.
Virtual reality theorist Joseph Bates claims that in character design, “believability will not arise from copying reality.” He argues that instead, a character's believability depends on its ability to show emotion convincingly (Maldonado and Roth). Bates' theory seems to be confirmed in practice. Upon the releases of animated features Polar Express (2004) and A Christmas Carol (2009), audiences complained that the motion capture animation seemed jarring and cold, despite the fact that the creators had used real life motion for their sources. If Bates' claim is true, and if one of the main goals of character design is audience identification, then gender recognition and motion capture may not be an ideal place for character designers to derive better representations of femininity.
Thus, the original challenge remains unsolved. Subjective gendering is difficult to do without falling back on stereotypes, while objective gendering methods are not yet practical for application in character design. Fortunately there is another remaining option—namely, outsourcing the job of gendering to other non-visual character elements.
For instance, a designer may gender a character through voice-over only, and not worry about adding physical gender clues at all. The movie series Land Before Time (1988-2007) uses this method almost exclusively. The main characters are all dinosaurs, and none of them have gender clues, except for their voices. With the exception of a few eyelashes (displayed only by a small number of young females), the males looked like the females, or else differed only in biologically accurate ways.
A second method of non-visual gendering is by “commentary.” This means creating a character with no gender clues, and allowing the viewer find out the character's gender when some other character happens to mention it. This method is used in the Nick Jr. show Blue's Clues, where the main character is a dog who barks instead of speaking. Apart from the fact that her caretaker calls her a “she,” Blue's character design doesn't have gender markers. In fact, the creators consciously went against traditional gender associations by making Blue blue, a traditionally male color (Nick Jr. UK). Most of the Blues Clues characters lack visual gender markers, and are instead gendered by their voices or by other characters' comments. It's also interesting to note that the show's website explicitly states each character's gender in the character descriptions. For instance, Blue's description reads thus: “The star of Blue's Clues, Blue, is a girl puppy who communicates to Steve and Joe through barks, which they understand” (Nick Jr.). Gendering by commentary or voice-over is nice because it allows designers more creative freedom with the female designs—they are no longer obligated to include gender clues that could be stereotypical (like costume gags or pink), and they aren't limiting themselves to realistic representations (like motion capture). Unfortunately, with both of these techniques, people will still read the characters as male until someone says something. While researching Blues Clues, I came across numerous web disputes over whether or not Blue was a girl or a boy.3 Ultimately, however, this is a problem with audience perception. The tendency for audiences to assume that characters are male isn't something that designers can change, except perhaps by consistently omitting gender stereotypes, and hoping that audiences will adjust their views over time.
A third gendering solution is to find a good middle ground where traditional female markers are optional, and are used as they fit the character, but are in no way required. Adventure Time (a new show created by character designer Pendleton Ward) is a good example of a cartoon that uses the “middle ground” method. The show has plenty of female characters, and a huge variety of creative character designs for both males and females. Some females have feminine markers, and some don't. For instance, Lumpy Space Princess is a purple cloud creature with a crown and a humorously gutteral masculine voice4. Princess Bubblegum on the other hand wears a pink dress and crown, and has a friendly female voice that fits her personality. In these cases, gender traits aren't treated as “markers” (merely methods of providing information). Instead they are used creatively as expressions of character, like any other design element. This allows female characters the same amount of design freedom as male characters without eschewing feminine traits completely. It evens out the design constraints—just as beards and mustaches aren't required for all male characters, bows and eyelashes aren't required for females.
Unfortunately, this kind of design equality is rare—and the reason why it is rare has to do with audience identification. Because audience identification is a major factor in attracting viewership, one of the main goals of character design is to create main characters that the audience can relate to, and secondary characters that will elicit the same feelings in the audience that the main character feels towards them. (For instance, if a villain is intimidating to the main character, chances are that the artist will design the villain to be intimidating to the audience as well). However, since the main character is often male, females are often gendered depending on their relationships to male characters. (For instance, available females often have more sexualized character designs than unavailable females like mothers, wives, and teachers). This places a considerable restriction on female design freedom. But what about cartoons with female leads? Do the designs of male characters follow the same rules? Are male viewers willing to watch a world filtered through a female's eyes?
The animation industry is notorious for its lack of female leads. Most main characters, especially of the non-human variety, tend to be male, and male secondary characters vastly outnumber females. One of the most famous documents to point out this discrepancy was “Dear Pixar, from All The Girls with Band-Aids On Their Knees,” written by NPR contributor Linda Holmes. The letter, a sincere and admiring request that Pixar “please make a movie about a girl who is not a princess,” sparked much debate among and workers in the animation industry and feminist media critics. Holmes was responding specifically to Pixar's movie The Bear and The Bow—a princess movie, and Pixar's first film featuring a female lead.5 She wrote:
I have nothing against princesses. I have nothing against movies with princesses. But don't the Disney princesses pretty much have us covered? If we had to wait for your thirteenth movie for you to make one with a girl at the center, couldn't you have chosen something — something — for her to be that could compete with plucky robots and adventurous space toys? (Holmes)
The letter caused much speculation on why female leads are so rare in animation. One of the biggest explanations is that most of the people working in the animation industry are men, and thus gravitate more toward creating male-centered content. Another explanation is that industry executives have long held the belief that girls will watch shows with boy leads, but boys won't watch shows with girl leads. Thus, by making a show with a girl lead, they would risk losing half their audience. While this belief seems to be confirmed by wishful identification studies (Hoffner), media critics have argued that the only reason why female-led shows have failed with boy audiences is that they tend to be flat and poorly made, or else they focus on stereotypically feminine activities (like shopping and fashion) rather than featuring thoughtfully written stories. Another argument is that in American culture it's more acceptable for girls to engage in traditionally male activities than it is for boys to engage in traditionally female activities.
Attitudes seem to be changing, though, especially in television. Since 1990, Nickelodeon has released several female-led shows, like My Life as a Teenage Robot, and The Wild Thornberries that found success with both boy and girl audiences. The Disney Channel has also done its share with Kim Possible, Brandy and Mr. Whiskers, and The Proud Family.
Unfortunately, female led films and television series are still only a small minority, and media critics continue to point this out. This kind of criticism is nothing new. For years, feminist critics have done exactly what this paper has attempted to do—pointing out the problems in female media representation and suggesting solutions. The hope is that by pointing out unfair representations in the media, we can promote a more just view of women in society. Unfortunately, while such criticism is well-meaning, it can sometimes have negative effects.
Cartoon Brew, run by animation historians Jerry Beck and Amid Amidi, is a blog that posts news relating to the animation industry, and gives members an opportunity to respond to the articles that are posted. A 2007 article called “Geena Davis on Cartoons” prompted several frustrated responses. (Davis had criticized the animation industry for creating flat, stereotypical female characters, citing Snow White and Daisy Duck as examples). One member (Inkan1969) wrote:
All that the rabble rousers produce are token female characters who are entirely defined by being rolemodel admirable; they owe their existence only to political correctness. I say let creatives create female characters that have actual significance in the story, and have character depth that goes beyond “I’m a PC female character” ( Chris Sanders created two such characters indeed. ). (Cartoon Brew)
This member argues that the problem isn't just that there aren't any female characters; it's that studios are creating female characters for the wrong reasons—to be politically correct and create role models. As a solution, the member suggests allowing creatives to make their own female characters with greater depth and story significance. While this argument has some merit (Pendleton was given the freedom to create his own characters, and thus came up with really creative female designs), it also assumes that artists will naturally create good female characters if allowed the creative space to do so. Is this true? Most of the artists in animation are male, and it seems plausible that they would gravitate towards creating male leads. On the other hand, two of the more respected female animated leads (Lilo and Nani from Disney's Lilo and Stitch) were created by male artist Chris Sanders.
Another member (John Sanford) alludes to a second possible reason why female characters are rarer in animation:
I worked on a movie with 3 female leads. Every time we had any of the characters do anything remotely interesting, someone would jump all over us and say “You can’t do that, you’re making her look mean/selfish/stupid/unreasonable etc.” The male leads are ALWAYS easier because you can make them as flawed as you want and everyone loves it. You even think of loading some flaws into a female character and a world of hurt comes down on you. (Cartoon Brew)
Sanford claims that it's difficult to create interesting female characters, because supervisors in the industry dislike having flawed female characters. This might be an unfortunate result of the demand for positive female role models in cartoons. It's possible that male leads are easier to write simply because they are far less scrutinized. Thus, instead of increased criticism bringing about better female characters, it prevents female characters from being made at all.
Perhaps these negative effects are due to the fact that most criticism focuses on the flaws of characters (they aren't perfect, so they must be poor role-models, and bad for kids), rather than focusing on the lack of thoughtfulness, depth, and creativity of female character designs. Based on what I have read thus far, it seems that the latter kind of criticism would be more helpful to getting rid of stereotypes, and creating believable female characters that both boys and girls could relate to. This, in turn, would help female-led cartoons break the myth that boys won't watch cartoons with female leads, and hopefully encourage studios to produce more of them.
In conclusion, there are good reasons to represent gender in non-human characters. Gender figures largely into the way people relate to characters, even non-human ones. Even if characters' genders are never mentioned, viewers will automatically attribute gender to them, and they are likely to assume that ungendered characters are male. I think girls want distinguishably female characters to identify with. However, the identification issue doesn't stop at gender. Girls also want to identify with characters that are interesting and well thought-out, not hollow role models of what girls should be, or flat stereotypes of how males imagine them.
Though objective methods of gendering exist, objective traits are often far too complex to condense and apply to cartoon characters. Furthermore, a character's success depends more on believability than on realism. This believability comes from a well-written interior—the character's emotions and reactions—as opposed to realistic body movement.
Subjective gendering methods have their own problems. Subjective methods are more likely to fall into stereotype, since they require designers to condense gender into a few traits that they consider to be “feminine” or “masculine.” Because each artist will choose to include or omit different traits, every rendering of gender will be colored by the artist's own imagination and biases. By outsourcing the job of gendering to the character's voice over or to commentary, however, artists can free up character design considerably. This way, gender traits can be used as creative design elements rather than obligatory “markers.”
While it's important to think critically about the media, forcing restrictions on artists in the name of political correctness can have negative effects. Well-meaning (but misdirected) criticism stifles the creativity of industry artists, and can actually make the lack of female representation worse. At the same time, it is helpful to look critically at why we make the decisions we do when creating media, instead of simply accepting conventions because they've become culturally familiar. As a kid, I knew what girl characters I liked; I knew that I wanted cool girl characters to relate to; and I got annoyed with seeing the same flat stereotypes repeated in girl characters from show to show, while the boy characters seemed to be so much more interesting. There's something to be said for giving girl characters the same amount of thought and creative energy as guy characters. Perhaps then, we'll be able to see more female leads in animation, and more egalitarian character designs.
Works Cited
101 Dalmations. Perf. Rod Taylor, Cate Bauer, Betty Lou Gerson. Walt Disney Productions, 1960.
Arthur. PBS. Television.
Beck, Jerry. "Geena Davis on Cartoons." Cartoon Brew. 8 Mar. 2007. Web. 05 June 2010. .
Blue's Clues. Nickelodeon. Television.
A Christmas Carol. Dir. Robert Zemeckis. Perf. Jim Carrey. Walt Disney Productions, 2009.
Hoffner, Cynthia. "Children's Wishful Identification and Parasocial Interaction with Favorite Television Characters." Journal of Broadcasting & Electronic Media (1996). Communication & Mass Media Complete. Web. 5 June 2010. .
Holmes, Linda. "Dear Pixar, From All The Girls With Band-Aids On Their Knees : NPR." National Public Radio. 1 June 2009. Web. 05 June 2010. .
Kozlowski, Lynn T., and James E. Cutting. "Recognising Sex of a Walker from a Dynamic Point-light Display." Perception & Psychophysics 21.6 (1977): 575-80. Web. 5 June 2010. .
Land Before Time. Dir. Steven Spielberg and George Lucas. Perf. Gabriel Damon. Sullivan Bluth Studios, 1988.
Maldonado, Heidi, and Barbara Hayes-Roth. "Toward Cross-Cultural Believability in Character Design." Agent Culture: Human-agent Interaction in a Multicultural World. By Sabine Payr and Robert Trappl. Mahwah, NJ: Lawrence Erlbaum, 2004. Print.
"Meet the Characters on Blue's Clues." Nick Jr. Nickelodeon. Web. 05 June 2010. .
"Meet the Creators." Nick Jr. UK. Nickelodeon. Web. 5 June 2010. .
Nowak, Kristine L., and Christian Rauh. "The Influence of the Avatar on Online Perceptions of Anthropomorphism, Androgyny, Credibility, Homophily, and Attraction." Journal of Computer-Mediated Communication 11.1 (2005). Web. .
The Polar Express. Dir. Robert Zemeckis. Perf. Tom Hanks. Warner Brothers Entertainment, 2004.
Robin Hood. Perf. Phil Harris, Brian Bedford. Walt Disney Productions, 1973.
Seibert, Fred, and Pendleton Ward, dirs. Adventure Time. Cartoon Network. Television.
"Tertiary Sexual Characteristics." Television Tropes & Idioms. Web. 05 June 2010. .
my answer is short and simple:
take everything everything Image Comics "artists" ever did, and everything Peter Chung ever did, combine them into a big pile, defecate on it, and burn it.
Then watch about 4 hours of Bakshi that Network and Pay TV won't touch with a bottle of good merlot by your side.
You'll be a wiz at gender destinction in no time.
[I]I'll work 10 hours a day for $350.
Andreas does the photoshop posters for Paramount, gets $700 per day at 7 hours plus an hour off for lunch.
You do the math as to which is a better deal.[/I]