The Washington PostDemocracy Dies in Darkness

Explaining the gender gap

For our first contribution to the gender gap symposium we are delighted to welcome Jane Mansbridge, Charles F. Adams Professor of Political Leadership and Democratic Values at Harvard University, 2012-2013 president of the American Political Science Association, and author of numerous groundbreaking scholarly works, including the award winning Why We Lost the ERAProfessor Mansbridge’s contribution provides the broader context necessary to understand why biases against and disadvantages for women persist and what can be done about them.


Since the “second wave” of the women’s movement in the late 1960s, feminists have been pointing out that the discrimination against women today is mostly structural and implicit.  You don’t usually see the “smoking gun.”  It is therefore highly resistant to legal action.

The structural variables have to do primarily with 1) women’s primary responsibility for childcare and 2) the unconscious replication of social networks.

Some universities have parental leave policies that give both parents who take 50 percent or more responsibility for childcare a semester of paid leave and a year off the tenure clock.  Some have subsidized on-site childcare.  Many have neither of these supports.  Even those that have the supports, however, do not begin to make up for the time spent on childcare by any parent with 50 percent or more responsibility in the years from one to 21 as the child grows up, gains skills, has the inevitable minor or major health problems and a busy or troubled social life, and relies on the parent for emotional and physical support.  Accordingly, women who have no children do better in academia than women with one child, and women with one do better than women with two or more.  Not until all men take 50 percent or more of the responsibility for childcare will the playing field be equal for women on this front.

People also tend to socialize by gender.   Men tend to have more men among their friends; women tend to have more women.  So when someone is asked to recommend someone for a lecture or a prize, or when people are sitting around discussing others’ good articles or books, men are likely to think of people they know, who are more likely to be men, and women are more likely to think of people they know, who are more likely to be women.   If the group is not 50-50 male and female, the numerically dominant group (in political science usually men) will tend to produce more mentions of others in that group.  Those who hear those mentions will underscore in their minds their memories of those people.  Social networks reinforce themselves.

Psychologists interested in group dynamics and in the preconscious have studied the implicit variables that produce discrimination by gender.

Studies of group dynamics reveal in-group positive biases and out-group negative biases, even when the “groups” are formed overtly and explicitly by the throw of a dice, a mechanism chosen to underscore the randomness of the selection.  Groups with deeper routes produce deeper biases, and gender has deep roots in every society.  Societies engage in much “gratuitous gendering,” giving male and female noun forms to neutral items such as table and armchair, or dividing all the world into Yin and Yang.  Gender is one of the first pieces of information given and asked about a newborn.   Mothers treat their boy and girl babies differently from the first days of birth.  Boys and girls separate into somewhat different social groups in the preschool years and remain in different groups thereafter, although are later brought together again by school activities, courtship, the raising of a family, and, in some societies, by relatively gender-neutral work.  In most societies work itself is gendered, although not with the strong taboos of some.

Anthopologists report that among the Suku of Africa, only the women can plant crops and only the men can make baskets. But among the Kaffa of the Circum-Mediterranean, only the men can plant crops and only the women can make baskets. Among the Hansa of the Circum-Mediterranean, only the men can prepare skins and only the women milk. But among the Rwala of the Circum-Mediterranean, only the women can prepare skins and only men milk.

Studies of implicit associations show that gendered attributes are part of almost everyone’s preconscious.  Go to this Harvard Web site and you will be able to take a test online that, whether you are a man or a woman, will almost certainly reveal your own implicit association of women with the humanities and men with science.  The test relies on our brain’s ability to register congruent items more quickly than incongruent ones (so that you can read the word “red” faster when it is written in red than when it is written in green).

Preconscious assumptions regarding gender also affect hiring and promotion patterns, as when male managers respond with a lessened desire to work with an employee who has a woman’s name but not when the employee has a man’s name if that employee has made a number of irritating negotiating demands.

How to handle these structural and preconscious biases?  First, acknowledge their existence.

Second, set in place mechanisms to diminish them, recognizing that it is hard, if not impossible, to eliminate them.  On childcare, the more time a university can give for paid leave and on its tenure clock to parents who are genuinely involved in 50 percent or more of their children’s care, the better.  We have not even begun to approach the point at which the time given could make up for the time taken away.  On networks, make sure that all search committees begin by phoning women faculty for suggestions of outstanding colleagues and students.  Make affirmative efforts to tap female networks to give out and collect every form of important information.  Recognize that citations are not neutral measures of the importance of a piece of work but derive from gendered social networks.  Recognize that fields are not neutrally “important” or “marginal,” but gendered, with fewer citations occurring in those fields that are gendered female.  On implicit biases, notice the faculty with whom deans and chairs get irritated and those who are the stars, and try to identify the possibility of gender biases.  When female colleagues complain of bias, try to see the ways in which they might be right.

This is a tough nut to crack.  The problem is larger than any of us and will take generations to solve.  But justice, care, and the full use of our society’s resources demand that we try.