Against the Theory of "Sexist Language"

It is common today in public discussion, whether the context is academic, political, or even legal, to take it for granted that using the word "man," in isolation or as a suffix, to refer to all of humanity, or using the pronoun "he" where any person, male or female, may be referred to, is to engage in "sexist language," i.e. language that embodies, affirms, or reinforces discrimination against women or the patriarchal subordination of women to men. Not everyone agrees with this view, and "he" and "man" often seem to creep inappropriately into the speech of even those who consider themselves above such transgressions; but the ideology that there is "sexist language" in ordinary words and in the ordinary use of English gender rarely comes under sustained criticism, even in the intellectual arenas where all things are supposed to be open to free inquiry. Instead, the inquiry is usually strongly inhibited by quick charges of "sexism" and by the other intimidating tactics of political correctness. Such defensiveness accompanies the widely held conviction that the theory of "sexist language" and the program to institute "gender neutral" language are absolutely fundamental to the social and political project of feminism. The theory of "sexist language," however, is no credit to feminism, for it is deeply flawed both in its understanding of the nature of language and in its understanding of how languages change over time. Since the ideology that there is "sexist language" seeks, indeed, to change linguistic usage as part of the attempt to change society and forms of thought, the latter is particularly significant.

First of all, the theory of "sexist language" seems to say that words cannot have more than one meaning: if "man" and "he" in some usage mean males, then they cannot mean both males and females in other usage. This view is absurd enough that there is usually a more subtle take on it: that the use of "man" or "he" to refer to males and to both males and females means that maleness is more fundamental than femaleness, "subordinating" femaleness to maleness, just as in the Book of Genesis the first woman, Eve, is created from Adam's rib for the purpose of being his companion. Now, the implication of the Biblical story may well be precisely that Adam is more fundamental than Eve, but the Bible did not create the language, Hebrew, in which it is written. If we are going to talk about the linguistic structure of Hebrew as distinct from the social ideology of the Bible, it is one thing to argue that the system of grammatical gender allowed the interpretation of gender embodied in the story of Adam and Eve and something very much different to argue that such an interpretive meaning necessarily underlies the original grammar of Hebrew--or Akkadian, Arabic, Greek, French, Spanish, English, Swahili, etc.--or that such a system of grammatical gender requires such an interpretation.

What a language with its gender system means is what people use it to mean. It is an evil principle to think that we can tell other people what they mean by what they say, because of some theory we have that makes it mean something in particular to us, even when they obviously mean something else. Nevertheless, there is now a common principle, in feminism and elsewhere (especially flourishing in literary criticism), that meaning is only in the response of the interpreter, not in the mind of the speaker, even if the speaker is to be sued or charged with a crime for the interpreter having the response that they do. There is also on top of this the Marxist theory of "false consciousness," which holds that "true" meaning follows from the underlying economic structure, today usually just called the "power" relationships. Most people are unaware of the power relationships which produce the concepts and language that they use, and so what people think they mean by their own statements and language is an illusion.

The implications of these principles are dehumanizing and totalitarian: what individual people think and want is irrelevant and to be disregarded, even by laws and political authorities forcing them to behave, and speak, in certain ways. But they are principles that make it possible to dismiss the common sense view that few people speaking English who said "man" in statements like "man is a rational animal" were referring exclusively to males, even though this usage was clear to all, from the context, for centuries before feminism decided that people didn't "really" mean that. But even if some speakers really did mean that, it is actually irrelevant to the freedom of individuals to mean whatever they intend to mean through language in the conventionally available forms that they choose. What was meant by the gender system in the languages that ultimately gave rise to Hebrew is lost in whatever it was that the speakers of those languages were saying to each other; but what we can say about the functioning of gender systems and about language in general is very different from the claims that the theory of "sexist language" makes.

Historically, if a language possesses a gender system and distinguishes between "he" and "she," then one or the other will also tend to be the common gender for when both genders are involved. In English, and most other languages with gender, that falls to "he," and the feminist argument is that this reflects patriarchal dominance and so sexism--a hierarchy in which the masculine is more fundamental. That may even be true in many cultural contexts; but interpretation is separate from the grammatical structure, and the structure allows for interpretation that cuts both ways. Logically, English "he" stands to "she" as "number" stands to "prime." Number, in a sense, is more "fundamental" than primeness, just because it is more general; but prime numbers are certainly no less numbers than any other numbers. Prime numbers are simply marked with a certain property that other numbers do not have. Calling prime numbers "prime" represents the traditional sense that the distinguishing property of prime numbers--that they cannot be evenly divided by any numbers besides one and themselves--is particularly striking and salient.

If "she" is logically subsumed under a more general "he," it may then be because the female was regarded as more "marked" than the male. Feminists sometimes notice this, to their irritation, in the structures of the words "female" and "woman" as compared to "male" and "man": each simply adds a syllable. Similarly, Afro-Asiatic (or Hamito-Semitic) languages from Ancient Egyptian and Hebrew to Modern Arabic have added the syllable -at as the mark of feminine nouns (where the t is usually silent and the a often later pronounced as e or i). More subtly, French may represent the same thing through the quality of the vowel in the definite articles: The feminine singular article, la, contains a full and pure vowel, /la/, while the masculine article, le, actually contains a reduced vowel, the indistinct and indefinite "schwa" sound. The full feminine vowel can easily be interepreted as more "marked" than the reduced masculine schwa.

Such superadded distinctness, properties, or syllables, of course, could represent something either positive or negative--femaleness could be either more valuable or less valuable than humanity in general. Or the property could be just salient and distinguishing, without being relatively more or less valuable. In any case, this means that the gender system of English is just as amenable to a feminist interpretation that it reflects a primaeval matriarchy as it is to the interpretation of Old Testament patriarchy. Since the gender systems of Indo-European and Afro-Asiatic languages certainly go back to the prehistoric periods where speculation about matriarchies proliferates, it is surprising that such an alternative interpretation has not been advanced by such theorists.

The actual positive markedness of the feminine gender could be argued on the basis of the gender systems of Greek and Latin, which display a general characteristic of complete Indo-European gender systems: the most common regular nouns display endings that are mostly identical for the masculine and neuter genders (-o- themes in Greek, like ho oîkos, "the house," masculine, and tò biblíon, "the book," neuter) but quite different for the feminine (-e- themes in Greek, like hee epistoleé, "the letter"). We might interpret this to mean that things with masculine gender are the most like inanimate objects, while things with feminine gender are unmistakably different from inanimate objects. This could mean that the feminine is more markedly human than the masculine. The similarity between the endings of masculine and neuter nouns still occurs in German. On the other hand, other noun endings in Greek and Latin (consonant stems, etc.) do group masculine and feminine together, contrasting them with the neuter, so there is also obviously a sense that both masculine and feminine actually are animate or human. A gender system that distinguishes femaleness as having a salient property, whether positive, negative, or neither, might still be regarded as a kind of sexism, whichever way the property goes; but it is a rather different matter from the usual feminist complaint about the patriarchal conception that we find all the way from Genesis to Aristotle to Freud: that the male is more "marked" and valuable because of the presence of a phallus, while the female is less "marked" and valuable, indeed envious, because of the absence of a phallus. It looks to be essential to the feminist theory of "sexist language" that a gender system where the masculine gender doubles as the common gender causes or reinforces "phallocentrism" and a patriarchal society. The feminine as merely the more "marked" gender, however, makes that unlikely.

But all this as a theory can actually be tested: We would expect that if linguistic gender were a correlate of social form, an engine for the enforcement of patriarchy or a reflection of the existence of patriarchy, then we would find it present in sexist or patriarchal societies and absent in non-sexist or non-patriarchal societies. In fact, the presence of gender in language to bears no relation whatsoever to the nature of the corresponding societies. The best historically conspicuous example is Persian.

Old Persian, like Greek, Latin, and Sanskrit, had the original Indo-European genders of masculine, feminine, and neuter. By Middle Persian all gender had disappeared. This was not the result of Persian feminist criticism, nor was it the result of the evolution of an equal opportunity society for women. It just happened--as most kinds of linguistic change do. Modern Persian is a language completely without gender. There are not even different words for "he" and "she," just the unisex un. (There are not even different titles for married and unmarried women: Persian khânum can be translated as "Ms.") Nevertheless, after some progress under Western influence, Iran has retreated from the modern world into a vigorous reëstablishment of mediaevalism, putting everyone, especially women, back into their traditional places. So the advice could be: If someone wants "non-sexist language," move to Iran. But that probably would not be quite what they have in mind.

Why didn't the "gender free" Persian language create a feminist utopia? This goes to show us that gender in language is completely irrelevant to the sexual openness of society. And one of the greatest ironies for us is that a feminist attempt to produce a gender free "non-sexist language" in English could only be contemplated in the first place because grammatical gender has already all but disappeared from English. Feminist complaints must focus on the meaning of words like "man," even though words can mean anything by convention, because the pronouns "he," "she", and "it" are all that remain grammatically of the three Indo-European genders. Getting gender to disappear in German or French or Spanish (etc.), on the other hand, would be a hopeless project without completely altering the structure of the languages. Occasionally feminists say that they are personally offended by people referring to ships or aircraft as "she"; and manuals of "non-sexist" language usually require that inanimate objects be "it" without exception. Good luck in French. Since every noun is either masculine or feminine, not only would this feature have to be abolished, but an entirely new gender, the neuter, presumably with new pronouns, would have to be created. Then there would have to be decisions about words like livre, which is differentiated into two words by gender alone: le livre is "book," from Latin liber, while la livre is "pound," from Latin libra. French doesn't even have English's happy refuge from inclusive "he" in "they," since you still have to decide in the third person plural between ils and elles. Only on ("one") allows for a gender free pronoun, just as "one" does in English.

It is now hard for people to quote Aristotle's famous dictum, "Man is a rational animal," without gratuitously adding that this is a "sexist" remark because, presumably, Aristotle didn't say "human beings." This goes to show the silliness of this whole kind of exercise and the willful know-nothing-ism of many writers when it comes to linguistic history. Even if we think that English "man" is "sexist," Aristotle was, of course, not speaking English. And in contrast with English, Greek and Latin both "mark" the male as well as the female in their vocabulary: aneér in Greek and vir in Latin both mean "man=male"; gyneé in Greek and femina in Latin both mean "woman"; and ánthroopos in Greek and homo in Latin both mean "man=person." Aristotle said "ánthroopos," not "aneér; and Classics scholars are usually happy to point out the inclusiveness of the former term. However, ánthroopos and homo are both in the masculine gender. Since Greek and Latin are languages were every noun has gender, like French, Hebrew, etc., there is actually no grammatically "gender neutral" expression possible, as there is in English. So was Aristotle sexist after all? If so, then we are still using a sexist expression in "human beings" because "human" is from homo, which had masculine gender to start with.

I often notice this kind of tangle over languages with much more complete gender systems than English since the politically correct term for people of Hispanic derivation or identity these days is "Latino," which is of the masculine grammatical gender but of course embraces both men and women. The feminine term "Latina" is never used unless only women are referred to. That sounds like it should make for a cause célèbre in the non-sexist language world, but of course no feminist would want to be labeled ethnocentric or culturally imperialist by applying their critique of English to Spanish. And then, unlike French, where gender specific word endings have been lost, Spanish still has a lot of nouns whose gender can be predicted from this o/a alternation of endings. A non-sexist Spanish presumably would have to pick some other vowel, or none, to replace these fossil Latin endings. And while some activists seem to have lately begun using the expression "Latino/Latina" more carefully, they are unlikely to be amenable to "reforming" the morphology of Spanish so that it would be as gender free as, of all things, English.

To reform a natural language like that, we would have to set up some political authority to decide what changes to make and then spend many decades coercing people into following the preferred forms: all to produce something that often happens spontaneously anyway, has progressed almost completely to the loss of gender in English already, and never in the past with the slightest effect on the structure of society. So why bother with all the grief and recriminations of trying to impose a feminist New Speak? But perhaps that is the point. All the grief gives ideologues something else with which to browbeat people and a completely phony issue through which to claim political authority over how people speak, in all innocence and good will, in natural languages. It can even translate into the introduction of virtual political commissars, often with punitive powers, into schools, workplaces, churches, etc. to monitor incorrect speech. And that is the kind of power that ideologues like.

But the conceptual error underlying this kind of thing didn't originate with feminism; it is the heritage of once popular but now discreditable theories about the nature of language--that how we talk determines how we think (to paraphrase something the semanticist S.I. Hayakawa actually said--a kind of linguistic behaviorism) and that the structure of language creates the structure of the world (promoted by the philosopher Wittgenstein and his recent followers). If we talk with grammatical gender, so this goes, then this determines not only that we think in exactly the same way but that the grammatical structure is projected into the world.

In fact, as the counterexamples indicate, such linguistic structures as gender determine little about thought and nothing about the world. Grammar is usually just grammar, nothing else. It is used to express meaning--it does not determine meaning. But the most significant assumption and the greatest hybris in the theory of "sexist language" is just that language and linguistic change are controllable, and so can be controlled by us, if we wish to. But language is not anything that can be planned or controlled. Languages grow and change spontaneously. The kind of theory that properly can describe the development of language is one that credits events with the capacity for developing spontaneous natural order. Theorists of such order range from the great naturalist Charles Darwin, to the great economist F.A. Hayek, and to the great philosopher Karl Popper.

Those who traditionally have wanted to control linguistic usage for one reason or another, and who believe that it can be controlled, are always ultimately frustrated. Literary or sacred languages can preserve ancient or elevated usages--as with ancient Hebrew, Greek, Latin, Arabic, Sanskrit, Chinese, etc.--but real spoken language goes off on its own merry way, exuberantly evolving new meanings, words, usages, and even new languages, always to the chagrin of the priests, scholars, and traditionalists. Nobody ever plans that. As feminism has wanted to control, mainly to abolish, the use of gender, it thus puts itself into the pinched shoes of the traditional grammatical martinet--leaving us with the image of a fussy schoolmarm swatting knuckles with a ruler rather than of the heroic revolutionary woman leading the way to a better future.

In the end, gender, in any language, is just an expression of the affinity of our understanding for logical divisons and hierarchies; and since logical divisions and hierarchies are essential to thought, the principle of eradicating gender (or "hierarchy") is absurd. Even if the feminine gender is usually more "marked" than the masculine, this can really mean anything, depending, indeed, on what we intend to mean. Instead of gender systems compelling patriarchy or, obviously, matriarchy, the whole idea of sexual equality was conceived in languages (English, French, German) with strong or remnant gender structures, while other languages with gender structures (Sanskrit, Arabic, Swahili) or without (Persian, Chinese, Malay) produced nothing of the sort. Serious intellectual dispute on any issue always must focus on what the speaker means by what is said, not on theories about how it is said compels certain unintended meanings, especially when such theories are clearly mere features of certain political and ideological systems of interpretation.

This essay was expanded, from a brief description of gender in languages, at the personal urging of Christina Hoff Sommers, the courageous author of Who Stole Feminism? How Women Have Betrayed Women [Simon & Schuster, 1994]. Dr. Sommers teaches philosophy at Clark University.

Copyright (c) 1996 Kelley L. Ross, Ph.D. All Rights Reserved

Return to Main Page