Subscribe to The Brown Daily Herald Newsletter

Sign up for The Brown Daily Herald’s daily newsletter to stay up to date with what is happening at Brown and on College Hill no matter where you are right now!


Science & Research

Language linked to gender inequality, research suggests

Studies indicate countries with gendered languages are more likely to have wider gender gaps

Senior Staff Writer
Monday, October 27, 2014

Countries where citizens speak gendered languages — in which nouns are masculine or feminine — display a higher rate of gender inequality than countries with languages that do not ascribe gender to nouns, said Jennifer Prewitt-Freilino, the only full-time psychology professor at the Rhode Island School of Design.

She presented her research about the relationship between language and gender inequality at a lecture hosted by Brown’s Department of Cognitive, Linguistic and Psychological Sciences Friday in Metcalf 305.

Though women are now overtaking men in areas like college enrollment, many inequalities still exist between men and women, Prewitt-Freilino said.

“In no country is there even equal pay,” she said. “Women are paid about 16 percent less than men.”

But the amount of inequality “is not uniform across cultures,” she added, citing language as a possible contributing factor in such differences in equality.

Language is “usually seen as not that important, but research over the last decade has suggested that how we speak also influences how we think,” she said.

Prewitt-Freilino hypothesized that “countries that speak predominantly gendered language should evidence less gender equality relative to countries with natural gender and genderless language countries.” Gendered languages are those with masculine and feminine words, such as Spanish. Natural gender languages are those in which most nouns are not gendered, but pronouns like “he” or “she” are gendered, such as in English. Genderless languages are those in which both nouns and pronouns are not gendered.

Prewitt-Freilino took data from 134 countries, of which 111 had primary languages that fit into one of the three categories. She identified 26 genderless, 12 natural gender and 73 gendered language countries and then looked at the Global Gender Gap index, a measurement of national gender gap, for each of the countries. The GGG index “benchmarks national gender gaps of 136 countries on economic, political, education- and health-based criteria,” according to its website. Each country is given a score between zero  — denoting absolute inequality — and one, for absolute equality.

Prewitt-Freilino found that the average scores for countries with genderless, natural gender and gendered languages were 0.68, 0.74 and 0.67, respectively.

The data supported her hypothesis that gendered languages have the most inequality, she said.

Since countries with similar languages often have other common links, Prewitt-Freilino looked at human development, religious tradition, geographic location and system of government as covariates that might affect the data. She then examined the data after accounting for these covariates, and found the average scores for countries with genderless, natural gender and gendered languages to be 0.70, 0.72 and 0.67, respectively: Gendered language countries still had the highest average inequality.

Prewitt-Freilino said previous studies found the same phenomenon that she noted.

In one study she cited, German-speaking and Spanish-speaking participants were asked to describe qualities of a key, a word that is masculine in German and feminine in Spanish. The German speakers often used words like “hard, heavy, jagged and metal” to describe the key, whereas Spanish speakers often used words like “golden, intricate, little and shiny” to describe the key.

In another study she cited, children were asked to write a story in response to a prompt with one of the three pronouns: “When a kid goes to school, (he/they/he or she) often feels excited on the first day.” The researchers found that only 12 percent and 18 percent wrote about female characters when given “he” or “they,” respectively. But 42 percent wrote about female characters when “he or she” was used in the prompt. Prewitt-Freilino suggested that this may be because people do not think about writing a female character unless “she” is explicitly mentioned.

Though she has identified an initial correlation, Prewitt-Freilino continues her work on gender differences in language use and perception, studying how the way men and women describe success and failure affects what others think of them.

She hypothesized that “women would view a candidate more positively when they shared success (by using “we”) and took personal responsibility for a failure (by using “I”) whereas men would view a candidate more positively when they took credit for success and (deflected) blame for loss.”

Prewitt-Freilino then conducted a new study, in which participants were asked to read a quotation from a student government candidate. The quote described a fundraising goal, where the candidate either failed or succeeded and used either “I” or “we” when discussing the fundraiser. She found that women liked “we” more if the candidate succeeded, whereas men valued personal success more and tended to not support the candidates who said, “I failed.”

She concluded the lecture by describing how subtle language differences can shape thought processes, which can affect social interaction.

“Not only is language a source for conveying current systems of hierarchy, but (it) might also be a way of reproducing them,” she said.

Xuan Zhao GS, who attended the lecture, said she enjoyed the lecture, as it was interesting to think about how language influences gender equity.

She expressed surprise to learn that some school kids have a male-dominant bias, as evidenced by one of the studies mentioned.

To stay up-to-date, subscribe to our daily newsletter.

  1. Patricia May says:

    This is truly interesting! Years ago I did my thesis research on gendered language and its effect on the stress levels of managerial females. At that time I dealt only with English. I am delighted to see that others are interested in the same types of issues!

  2. Christopher Burd says:

    The effect is pretty small, though. With only 134 data points, could it be due to random noise?

  3. In fact, one would need to ensure that the effect is not being driven by the preponderance of countries in the study whose dominant language is either Spanish or Arabic, both of which happen to have grammatical gender. As I understand it, the statistics were done by countries, not by languages – so with many countries falling into these two groups, this could skew things quite a bit. Moreover, as I unde
    rstand it, languages with so-called “natural gender” (like English) displayed LESS gender inequality (relevant to Patricia May’s comment, as she was looking at English). This too may be an accident of the number of countries for which English was the dominant language (English happens to be a ‘natural gender language’ and not a ‘grammatical gender language’). I read the paper and attended the talk and teach related work in my courses, and I think one needs to approach the results extremely tentatively for the reasons mentioned above, among others. (given the presumed logic behind the correlation, the countried which speak so-called ‘natural gender’ languages should not have shown the least amount of gender inequality, but should have come out in the middle. The researcher did try to explain this by saying that the ‘genderless’ languages do still have some hidden ‘natural gender’ in their system (if I understood her claims correctly) but the actual work there showed rather that the participants had gender biasses, not that there was anything relevant about the linguistic systems.

  4. I am interested to see when Canada & USA make laws against French and Spanish and force all the gendered nounds to be gender neutral. Otherwise it is discrimination to be forcing that there are only 2 genders when there are now over 72 recongised gender identities.

Comments are closed. If you have corrections to submit, you can email The Herald at