Moving Beyond the Bechdel Test
Last year, four Swedish cinemas introduced a new rating system designed to alert moviegoers to possible gender bias in films. Under the new system, only those films passing the Bechdel test will receive an A rating.
Named for its creator, American cartoonist Alison Bechdel, the three-part test requires a film to a) feature at least two (named) female characters who b) speak to one another c) about something other than a man. While the test has been effective in raising moviegoers' awareness of the under- and misrepresentation of female characters, Bechdel never intended it to be used as an actual measure of the quality of a film's portrayals. Others, perhaps enamored with the test's simplicity and lacking other tools, have now systematically applied it to hundreds of films, often with unreliable and curious results. Films such as Gravity, with a sole but prominent female protagonist, fail the test, while films such as American Hustle pass, in spite of their heavy sexualization of female characters. The test does not require that females serve as protagonists or even as major characters who exert some influence in the story. As such, the test sets absurdly low standards for portrayals of women.
Although any set of more accurate measures or questions is unlikely to be as easy to apply as the Bechdel test, the following three questions may ultimately be more helpful in assessing the quality of a film’s portrayals of women:
•Are the female characters central to the story?
•Do the female characters have agency?
•Are the female characters multidimensional?
Centrality, agency, and dimensionality are key factors in determining the quality of film portrayals of women and girls. I recently released a study, titled It’s a Man’s (Celluloid) World, considering the centrality and agency of characters appearing in the top-grossing films of 2013. Conducted by the Center for the Study of Women in Television and Film, the analysis found that females accounted for 30 percent of all speaking characters (major and minor), 29 percent of major characters, and only 15 percent of protagonists. Last year, the vast majority of films featured male protagonists, meaning the stories were told from a male perspective. Moreover, males with speaking roles outnumbered females by more than two to one.
The study also used two measures of a character’s agency or ability to act and influence not only his or her own outcome but the outcomes of others. The measures included the presence (and type) of goals and whether characters acted as leaders. Overall, 75 percent of speaking characters had an identifiable goal. A larger proportion of male characters had an identifiable goal (79 percent) than female characters (67 percent). Further, male characters were much more likely to have work-related goals than personal life goals (75 percent vs. 25 percent). In contrast, the goals of female characters were split more evenly between work-related and personal life goals (48 percent vs. 52 percent).
We also examined the gender of characters portrayed as leaders. A character was considered to be a leader if others followed his or her behavior and/or directives. Leaders could occupy formal positions of power in corporations, politics, criminal organizations, or the military, or more informal positions of power, serving as leaders in social groups. Seventeen percent of all characters were portrayed as leaders, but films portrayed a larger proportion of male characters (21 percent) than female characters (8 percent) as leaders. Broken down by type of leader, males comprised 97 percent of blue-collar leaders, 89 percent of political leaders, 88 percent of criminal leaders, 86 percent of military leaders, 84 percent of white-collar leaders, and 77 percent of social leaders.
Studies of film and television currently underway will also consider the dimensionality of female characters by examining the number and types of roles female and male characters assume in their interactions. Generally speaking, characters who interact with many individuals in a variety of settings are more likely to be multidimensional than characters who interact with one or two other characters in a single setting. For example, someone who plays a supportive wife or girlfriend in one setting is likely to have limited dimension. However, if that supportive wife is also a high-powered attorney in the workplace and a concerned friend in another setting, she is more likely to be multidimensional.
Centrality, agency, and dimensionality form the foundation for a thoughtful analysis of female characters. Any serious discussion of film portrayals requires those concerned with this issue to move beyond the minimum standards set by the Bechdel test to imagine a new set of measures that ask more of Hollywood than the very least it can muster.
For new research on women in all media platforms, see the Women's Media Ccenter's updated report The Status of Women in the U.S. Media 2014.