Book Read Free

Wordslut

Page 11

by Amanda Montell


  Pronouns aside, there are also some languages that are essentially gender-free, containing very few words that make reference to a person’s “natural” gender at all. Yoruba, a language spoken in Nigeria, has neither gendered pronouns nor the dozens of gendered nouns we have in English, including son, daughter, host, hostess, hero, heroine, etc. Instead, the most important distinction in Yoruba is the age of the person you’re talking about. So, instead of saying brother and sister, you would say older sibling and younger sibling, or egbun and aburo. The only Yoruba words that make reference to a person’s gender (or sex, as it were) are obirin and okorin, meaning “one who has a vagina” and “one who has a penis.” So if you really wanted to call someone your sister, you would have to say egbon mi obirin, or “my older sibling, the one with the vagina.” When you get that specific, it makes our English obsession with immediately identifying people’s sexes seem just plain creepy.

  As a high school student, I took Italian as my foreign language (this is a language with grammatical gender), and while learning which words were feminine and masculine, I always wondered how exactly each noun got its gender assignment. Why were the words for table, chair, and fork feminine, while napkin, food, and knife were masculine? On the surface, it seemed totally random. But linguists agree that pairing nouns and genders is actually quite a complex process. Morphologist Greville G. Corbett once wrote that the rationale behind gender assignments varies considerably across languages. For some, the assignment is based on the sound or structure of the word; for others, it’s based on the meaning; and for many, both the structure and the meaning are involved. Furthermore, it’s very possible for a word’s gender assignment (and the motivation for it) to shift over time.

  Throughout history there have been a few male scholars who’ve tried to manipulate the relationship between grammatical gender and human gender to reflect their personal views of men and women. In the 1800s, a German grammarian named Jakob Grimm saw grammatical gender assignments as direct extensions of biological sex, a notion he deemed necessary for making sense of the world. “He spoke of the concept of grammatical gender as an extension of a ‘natural’ order onto each and every object,” Romaine explains. “Things named by masculine nouns were, in Grimm’s opinion, earlier, larger, firmer, quicker, more inflexible, active, moveable, creative; those which were feminine were later, smaller, softer, quieter, suffering/passive, receptive.”

  Grimm had a pal named Karl Lepsius, a linguist from Prussia, who agreed with this on-the-nose interpretation. Lepsius went so far as to claim that only the most civilized “leading nations in the history of mankind” distinguished between gendered nouns. To him, speakers of grammatically gendered languages had an overall more sophisticated understanding of the two human sexes. Languages without grammatical gender? They were “in decline.”

  As far as grammatical gender theory goes, Grimm and Lepsius were boneheads—if for no other reason than the fact that English seems to be doing pretty okay (oh, to have the confidence of a nineteenth-century Prussian man). And not everyone shares their extreme views about how “natural” gender should or shouldn’t be reflected in language. Many scholars have actually decided that there is no connection, no “leakage,” between them at all. But Grimm and Lepsius were right about one thing: grammatical gender and speakers’ attitudes toward men and women are not always separate. They just aren’t objective or inherently truthful either.

  Sometimes they can be pretty darn sexist.

  In languages with grammatical gender, there are entire ideas about men and women that cannot be communicated in a way that’s “grammatically correct” by prescriptive standards. In French, for instance, most prestigious jobs are masculine: the French words for police officer, doctor, professor, engineer, politician, lawyer, surgeon, and dozens of others all have masculine gender. (The words for nurse, caretaker, and servant all happen to be feminine, though.) Thus, in French, if you wanted to say something like, “The doctor is brave,” but the doctor is a woman and you want to indicate that, you’re out of luck. Grammatically speaking, the noun le docteur, meaning doctor, and the adjective courageux, meaning brave, would both have to be masculine.

  French feminists have tried to come up with alternatives—la docteur, la docteure, la doctoresse—but in France, there is a real-life grammar police, an official language council called the Académie française, which is reluctant to recognize such words or add them to the dictionary. (At the time I’m writing this only four of the current thirty-six members of the Académie are women. Yet somehow, six of the members are men named “Jean.”)

  In Italian, the masculine noun segretario refers to the prestigious position of a political secretary, like a secretary of state (a role traditionally held by men), while the feminine noun segretaria refers to a low-paid receptionist (a role traditionally held by women). Today, if a woman starts out as a segretaria in a government office and moves her way up to become a secretary to a politician, she would have to change the suffix of her title to a masculine one. For her, moving up professionally would literally mean having to masculinize her own title.

  How much of an effect can these grammatical gender assignments truly have on the way people see the real world? A major one, and there’s research to back this up. In 1962, scholars conducted an experiment where Italian speakers were presented with a series of made-up gibberish nouns that either ended in o or a (typically masculine and feminine suffixes in Italian, respectively). The speakers were asked to imagine what these faux nouns might represent and describe them using a list of adjectives: good, bad, strong, weak, small, large, etc. Then, they were asked to describe men and women using the same adjectives. The results? The feminine nouns, just like the women themselves, were described as good, weak, and small. The masculine nouns and men were bad, strong, and large. This study proves there’s no way grammatical gender isn’t leaking into speakers’ worldviews.

  There are also plenty of languages with noun classification systems that aren’t explicitly gendered (and these speakers do not have a “lower consciousness” of the sexes, as Lepsius argued, whatever that even means). In the Tamil language of South India, nouns are divided by caste—either high caste or low caste. A Native American language called Ojibwa categorizes its nouns on the basis of “animate” or “inanimate” (which, to me, seems way more logical than the masculine-feminine distinction).

  Tamil and Ojibwa are also examples of languages whose classification systems are not at all arbitrary but instead directly rely on what each word means (or meant at some point) to its speakers. People, animals, trees, and spirits belong to the animate class of Ojibwa nouns, but because of the speakers’ cultural outlook, so do snow and cooking pots. Where cultural significance in noun classification starts to get problematic is when it happens in languages with masculine and feminine grammatical distinctions—there are several of these languages. Because grammatical gender and cultural perceptions of human gender are not separate for these speakers, theoretically, their impressions of men and women aren’t able to escape the language’s influence and would inevitably start to solidify as soon as they learn to talk.

  One very untheoretical example of this comes from the Dyirbal language spoken in Aboriginal Australia. Dyirbal has four noun classes: the first is masculine, the second is feminine, the third is specifically for edible fruits and vegetables, and the fourth is for anything that doesn’t fall into the first three. That might sound straightforward enough, but here’s where it gets weird: an animal will always be assigned masculine gender in Dyirbal, unless it is notably more harmful or dangerous than the others in its category—in that case, it gets moved to the feminine class. For instance, while fish belong to the masculine category in Dyirbal, dangerous fish, like stonefish and garfish, are marked as feminine. So are all other potentially deadly creatures, as well as anything having to do with fire, water, or fighting. “The rationale for this categorization tells us something about how Dyirbal people conceive of their world and
interact with it,” Romaine says.

  The Dyirbal system also serves as another example of treating masculine gender as the default. Sort of like, everything in the world of this language is male until given a good reason not to be. This default male pattern shows up in the grammar structures of hundreds of languages—in ways as understated as the French rule that only female nouns are marked with an e, or the fact that in Italian, a group of both male and female kids would automatically be called a group of “boys” (only when referring to a group of all girls would you use the feminine noun). Meanwhile, default maleness can be as overt as the Dizi language of Ethiopia, where almost every noun is classified as masculine, except for things singled out for being “naturally” female (girl, woman, cow), as well as things that are small in size (small broom, small pot). Ultimately, language can serve as a rather blatant means of otherizing all things feminine.

  In Dyirbal’s system, a noun is considered masculine unless it could literally kill you. This gender distinction between dangerous and nondangerous nouns might seem far-flung, but English speakers unwittingly do something similar. Think of all the instances where the pronoun she instead of it is used to refer to nonhuman things—namely, cars, boats, planes, oceans, countries, the Loch Ness monster, and hurricanes. (In the 1950s, the National Weather Service made the formal decision to assign solely female names to hurricanes, borrowing from naval meteorologists’ tradition of naming their ships after women. Fortunately, in 1979 they got their gender equality act together and revised the system to include both male and female names.) All of these things are big, challenging, and have a long history of being conquered by men.

  In 1995 United Airlines magazine printed an article titled “Boeing Beauty,” penned by a pilot who described his Boeing 727 like this: “My wife JoAnna met her competition today. She has known about the ‘other woman’ for most of our 31 years of marriage . . . of course, I’m talking about an airplane . . . but what an airplane it is.” A dad shows off his freshly waxed beamer—“Isn’t she a beauty?” An IT guy stops by your desk at work to look at your malfunctioning laptop: “Open ’er up!” In 2011 a group of linguists published a study in the Journal of Popular Culture analyzing sexist souvenir T-shirts sold in the wake of Hurricane Katrina: slogans included “Katrina, That Bitch!”; “I Got Blown, Pissed On, and Fucked By Katrina: What a Whore”; and “Katrina Can Blow Me: She Won’t Keep Me Away From Mardi Gras 2006.”

  In the 1920s male linguists came up with a name for this process of giving an object a human pronoun; they called it “upgrading,” as if by calling these things “she,” they are elevated to human status. They couldn’t quite see that it simultaneously downgrades women to the status of toys and property.

  In practice, what these metaphors of women as nature, territories, and technologies do is place feminine gender in that same distant category of “other.” According to Romaine, by comparing her to things like storms and seas, “woman is symbolic of the conflict between nature and civilization, tempting men with her beauty, attracting men with her charms, but dangerous and therefore in need of conquest.” Woman is a continent to colonize, a fortress to siege. These sentiments are reflected not only in English; in languages all over the world, from Italian to Thai, a nation’s government is labeled as having “founding fathers,” while the land itself (“Mother Nature,” “virgin territory”) is perceived as a feminine entity. In grammar as in allegory as in life, women are considered reckless places outside the civilized male world—wild things meant to be tamed into the weak, delicate flowers we’ve traditionally wanted women to be.

  Some scholars believe that languages with grammatical gender have a more blatant impact on the attitudes of their speakers. A 1982 study suggested that children who grow up speaking a grammatically gendered language, like Hebrew, start to grasp their gender identities earlier than children who grow up speaking English or Finnish. One interpretation of these findings is that a language like Hebrew might be ideologically trapping its speakers into gender stereotypes sooner. But Romaine doesn’t think of it that way. Because when a problem is more visible, it is often more discussed too. “In languages with grammatical gender, like French and Italian, speakers’ attention is constantly drawn to the issue of gender in a way that it is not in a language like English,” Romaine says. With more attention, a solution might be closer within reach.

  In France especially, language is one of feminists’ most powerful tools of resistance. French women often appropriate feminine-gendered terms in place of masculine ones, like using the feminine word la personne, meaning “person,” instead of the masculine le sujet, to describe the subject of a story or conversation. “Although theoretically . . . [le sujet] supposedly encompasses both males and females, one of the tenets of French feminist theory is the argument that patriarchy constructs the subject as masculine and effectively excludes women,” Romaine explains. “Paradoxically, its apparent grammatical inclusion of women guarantees their exclusion.”

  In English, we’ve introduced the more inclusive term person as a substitute for man in words like chairperson and salesperson—but the difference is that while English attempts to remove gender from the word entirely, the French la personne is explicitly feminine. English person doesn’t make the same dramatic statement, as it doesn’t actively shift the position of the subject to feminine soil. “The political significance of the use of personne by French women is lost when translated into English,” comments Romaine. And while it is possible for feminists to draw attention to the English language in a similar way—it’s often done through wordplay and respellings, like herstory, womyn, and shero—it just doesn’t seem to gain as much traction.

  Part of that might be because not everyone is cool with “feminizing” words that are already technically gender neutral. Many women comedians have said they detest when people call them comediennes. “I don’t like it,” Broad City star Ilana Glazer told Elle magazine in 2016. “Do they call them doctresses?” Margaret Cho agreed: “I like stand-up comic better. Just comic.”

  Psycholinguistic studies show that in English, excessively “girly” suffixes like -ette and -ess possess actively negative or at least diminutive connotations. After all, -ette did not start out as a feminine suffix, but as a way to refer to something smaller or of lesser value (kitchenette, cigarette). Words like actress and waitress are still in everyday use, but there used to be way more of these gendered nouns: neighboress, singeress, servantess, spousess, friendess, farmeress, and indeed doctoress were all real words in Middle English that have faded into obscurity.

  Many women object to these feminized suffixes, but others adore them: just like some women actively like their gender to be highlighted in phrases such as “female writer” or “woman scientist,” there are women who delight in and identify with explicitly “girly” grammatical structures. Not long ago, I posted on social media that the Italian language distinguishes between male and female elephants—elefante means “boy elephant;” elefantessa means “girl elephant”—and while half of the comments I received expressed that this seemed odd and gratuitous, I also got several messages from women who thought the -essa suffix was enchantingly femme and not belittling at all.

  Plenty of women find charm in feminine suffixes. A UK organization encouraging young girls to pursue science, technology, engineering, and math boasts the company name STEMette. I recently came across the website of a female business owner who calls herself an “entrepreneuress.” There’s no conclusive right answer to whether or not these words are sexist. But it is worth thinking about why we perceive them the way we do.

  Women are certainly not the only people who benefit from critiquing grammatical gender—it is both politically meaningful, and just plain practical, for trans and gender-nonconforming folks too. You might think that a person who doesn’t identify as either a man or a woman would be screwed in a language like French, but queer speakers are coming up with some pretty creative ways to solve these problems. “People
can actually use a binary grammatical gender system to position themselves outside of that binary,” says Santa Barbara linguist Lal Zimman, who leads workshops for schoolteachers on how to give gender-inclusive instruction in languages with binary systems.

  Queer speakers of Hebrew, for example, often use a mixture of masculine and feminine forms, or invent entirely new ones, to express their queer identities in ways that English speakers don’t have the opportunity to. In 2016 news circulated about a Hebrew-speaking summer camp in Maryland that allows kids to modify gendered suffixes so that everyone can feel included. In Hebrew, like in Italian, you would use the masculine term for “kids” to describe any group that contains at least one boy. The Hebrew word for “kids” ends in the masculine suffix -im—the feminine version of the word ends in the suffix -ot—but at Camp Moshova, groups of both boys and girls use a mix of the two: the newly invented suffix -imot. Even the word for camper is gendered in Hebrew—chanich means male camper, and chanichah means female camper—leaving out any folks who don’t think of themselves as either. At camp, however, a nonbinary kid can use a new, gender-neutral form of the word: chanichol.

  Sociolinguists tend to think of these new language usages as fascinating and exciting, but not everyone is quite as pumped. Many people I know (or see on Reddit)—people who never think or care about “proper grammar” in their everyday lives—suddenly get very unsettled at the idea of someone changing a word just because they want to. Grammar, as they see it, is a supreme, stable authority that you shouldn’t be able to challenge so freely.

  In the United States, a lot of this drama has to do with pronouns. As transgender and nonbinary identities become more and more visible, so does the discussion of “preferred pronouns,”* and many people who identify as neither male nor female are choosing to go by singular they. But not everyone is on board yet. Folks often defend their resistance to using they to describe one person by arguing that the word as they learned it is plural; using it any other way, they contest, would be grammatically incorrect.

 

‹ Prev