Language and Woman’s Place Revisited
ROBIN TOLMACH LAKOFF
1975
It is hard to remember just how different the world was when Language and Woman’s Place (LWP) was first published in 1975 and harder still to return (even in imagination) to that world. Rereading the book, I am struck equally by how much has changed and how much remains essentially the same. While the knowledge available to me (as a linguist and a feminist) back then was much sparser than what we have at our disposal today, work done then still has bearing on the ways we think now.
The original essay was situated at a revolutionary moment, in both linguistics and women’s history (not to mention American history). There was the youth revolution against the Vietnam War and the pieties of the “Establishment.” There was women’s liberation, born out of the civil rights and antiwar movements, but by about 1968 taking off on its own. The third (and most obscure) revolution occurred within transformational linguistics, the creation of generative semantics. Each of these contributed to LWP.
I entered linguistics as transformational generative grammar (TGG) was being developed at MIT by Noam Chomsky and his students and collaborators. It is difficult now to remember what linguistics was like prior to the advent of TGG: an obscure field, hyperspecialized, hardly capable of providing Big Answers or even asking Pretty Big Questions. American structural linguistics, the dominant paradigm in the United States until the mid-1960s, discouraged questions that could be explored through the medium of the investigator’s intuitions. Language had to be analyzed as an astrophysicist might examine a distant galaxy. There were good historical reasons for this stance; but by the 1960s it had made linguistics a sterile discipline.
Besides being a powerfully charismatic figure in linguistics (and other fields), Chomsky played an active role in the antiwar movement. It was impossible to do linguistics at MIT and remain neutral about events in the larger world. Chomsky’s growing status as public intellectual grew out of his contributions to linguistics, and brought that previously obscure field to public notice. It was thrilling: in our small way, we, the young and powerless, could play a role, however tiny, in making the revolution. We could change the language, the discourse, of a field—and of a nation.
But it is often true of revolutions—certainly of all of the ones I have personally known—that, once they succeed, they factionalize. By the late 1960s it had become clear to several of us that Chomsky’s linguistic revolution wasn’t the revolution in which we had enlisted. Chomsky had promised us a theory and method that would make language a “window into the mind.” But within standard transformational theory that possibility could be realized only to a very limited degree, if at all. While investigators could use their minds as interpretive instruments—to judge the grammaticality or semantic similarity of sentences—they were not permitted to investigate meaning, much less a speaker’s intention in uttering a sentence in a particular form, or the effect of that utterance on an addressee. TGG permitted us to posit a formal relationship between active and passive sentences but not to talk about why English has a passive construction, or why a speaker would be moved to use it. Particles like well, Imean, or like were beyond analysis and beneath contempt, examples of “performance” rather than “competence” and hence not rule-governed. (Yet speakers know how to use them and knowledge of this kind is what the linguistic grammar represents.) Such cases began to multiply. Chomskyan theory eliminated the possibility of examining those parts of language that revealed the most about its users’ minds and social relationships. If transformational theory was a window into the mind, that window was in need of cleaning. Some of us began to think about ways to build depth into TGG, in which we still counted ourselves true believers. We devised rules and representations to relate externally accessible linguistic forms to mental states—for example, desires, assumptions, and personal identities—while retaining the Chomskyan belief in the primacy of the syntactic component of the grammar. Deep structure simply got deeper, wider, and more complex.
Far from accepting these innovations as moves in the right direction, Chomsky and his loyal adherents objected strenuously, creating an irrevocable breach. The new theory, generative semantics, was impossible, they argued. It was—the ultimate dismissal— “too powerful”: rather than being the most economical way to relate surface form to deeper structures, it forced into the theory more complex, messy, barely formalizable constructs; it required that there be many more—and more kinds of—transformational rules relating deep and surface structures than were required in the simple and general formulations of the standard theory. Chomsky’s concept of “explanatory adequacy” required the linguist to choose the theory that related deep to surface in the simplest, most economical, way.
We riposted: simple and economical TGG might well be if the criteria for adequacy were minimal. But to relate language and mind, the grammar had to be more complex than was possible in the standard theory. That complexity, while messy, would permit deeper, more gratifying explanations: real explanatory adequacy. It would allow linguistics to relate form to meaning and intention, demonstrating (something notoriously missing in the Chomskyan systematics) a precise, rigorous linkage between thought, culture, and language—an ambitious but essential project.
Much too ambitious, the other side (by then known as Extended Standard Theory) groused. If you followed generative semantics to its logical conclusion, everything speakers know about the world would have to be included within the transformational component, which therefore would become infinite. And while the output of the grammar, the sentences of a language, constitute an infinite repertoire, the grammar itself could not be infinite, or it could not be learned, and no one would be able to internalize the grammar and speak a language.
Not necessarily, said the generative semanticists. The linguistic grammar need only include those aspects of the extralinguistic world that have direct bearing on grammatical form: just a small subset of everything. So for instance, the grammar would have to include some representation of the social status of participants, at least in those languages that force speakers to make distinctions between intimate and formal second person address. But it would not have to incorporate specific mention of participants’ occupations, or heights, or religions. The project was achievable. But we still had to answer, at least to our own satisfaction, the question that these claims raised: What parts of our psychological and social reality did require linguistic encoding, in at least some languages?
While this brouhaha was going on in the rarefied world of linguistic theory, other fights were proceeding in the larger universe. By around 1968 the feminist movement (then most commonly called “women’s liberation,” or, often disparagingly, “women’s lib”) was coming into its own. It is often hard to locate the moment at which a revolution begins: the storming of the Bastille, the Boston Tea Party, or the moment at which the women’s movement caught fire. The 1963 publication of Betty Friedan’s Feminine Mystique was critical. Also important was the training many women received in revolutionary theory and practice in the civil rights and antiwar movements: not only the practical experience of distributing leaflets, striking, sitting-in, and otherwise protesting; but the less gratifying realization that they were being left to make the coffee and run the mimeograph machines while the men got credit for the intellectual heavy lifting. Rising out of these experiences, feminism emerged in the late 1960s as an articulate locus of protest of its own. What women had learned from participating, if often behind the scenes, in the earlier movements they brought to the new one, their own.
The women’s movement first exploded taboos, speaking the unspeakable, thinking the unthinkable, occasionally even doing the undoable. While bras may never have been burned, old constraints were incinerated in those early years of “sisterhood.” It was recognized early that language was important, that there were consequences when grown women were “girls” and when the masculine pronoun was “normal” to refer to everybody.
Generative semantics and the women’s movement thus both arose as protests against the status quo, against the assumption of the unmarked and the “normal” as unquestionable. Both generative semantics and the linguistic arm of the women’s movement started on a small scale, looking at the smaller and more concrete aspects of language. Transformational theory (including generative semantics) had permitted the incorporation of syntax into linguistics (the antimentalism of structuralism largely kept levels above morphology off-limits for them), but the possibility of incorporating larger and more abstract units—conversational turns, paragraphs, discourse—into linguistic analysis was still well in the future. Linguists spoke on occasion of “structure above (or beyond) the sentence level,” but mostly about how it couldn’t be done. When we attempted it, we thought of larger units as concatenations of sentences: S + S + S …, rather than as structures with rules of their own, wholes different from the sum of their parts. Likewise, when linguists began to address the relationship between language and gender, we started with the smallest and most concrete units: sounds and words. Later, in the mid-1970s, when linguistics (and allied disciplines) provided the means for larger-scale analyses, those of us working in this area were quick to put them to use.
So generative semantics and the women’s movement had similarly revolutionary origins and a similar need to question and subvert established beliefs. For me at any rate they came together in another way as well. My interest in the intersection of language and gender arose on two fronts: my political involvement in the women’s movement and my academic engagement in the transformational dispute.
If generative semantics could demonstrate that it could distinguish those concepts that required underlying structure representation (because they had, in at least some languages, explicit linguistic encoding) from those that did not, we could demonstrate that our grammar was finite. But what kinds of concepts required inclusion? One potential answer: gender.
The inclusion of gender as an area of linguistic investigation was less obvious thirty years ago than it seems today. Scholars were willing to acknowledge that in a few “exotic” languages (Japanese, Dyirbal, Arawak, and Koasati) the genders of participants and subjects of conversations required linguistic encoding and hence (arguably) underlying structure representation. But this seemed not to be true of English and other languages with which investigators were likely to be familiar. And if a case could not be made for the inclusion of gender in English, then arguably, gender was not a part of universal grammar and hence not a part of underlying structure. It would remain beyond what was defined as “interesting” (like “structure beyond the sentence level”).
So my aim, in embarking on LWP, was threefold:
1. To demonstrate that at least one extralinguistic artifact (i.e., gender) required linguistic representation;
2. To demonstrate that gender required linguistic representation even in languages like English, where its presence was (perhaps) less clearly felt than in “exotic” languages (these were my concerns as a linguist);
3. To use linguistic discrepancies between women and men as a diagnostic of social and psychological inequities between the sexes. (This was my concern as a feminist.)
2003
We have come a long way (baby), to paraphrase an irritating advertising slogan from the 1970s. Like the sentiments expressed in the commercial, the changes in gender stereotypes may look encouraging but, when inspected more closely, are often depressing. Many of the expressions that were commonplace back then, serving to keep women in their place, have become marginal or nonexistent. The difference between the bad old ways of speech and the problematic language that swirls around us now illustrates how far we have come:
• Allusions to former first lady Hillary Rodham Clinton (even today, long after her time in the White House) in the media, as a strident feminist and obstreperous bitch (while the current first lady, Laura Bush, is often presented, crypto-contrastively, as sweetly submissive);
• The continued sniping at “feminism” and “feminists” as the cause of all things evil (the high crime rate, drug addiction, the decline of the educational system, and as in Sommers [2000], discussed below, the problems facing males); and the frequent denials of feminist sympathies by younger women, even as they espouse feminist values;
• The nostalgia for a past when men were men, women were women, there was no divorce, and children were seen and not heard as a halcyon golden age—although historical research shows that such a time either never existed or was much less pleasant for many of us than the myth suggests.
Linguistics has also come of age. While a generation ago “structure above the sentence level” had the status of the basilisk (mythical and toxic), now it is an accepted area of linguistics, and methods of discovery and analysis have been developed to understand it. By the mid-1970s conversation analysis had been brought into sociolinguistics; during the 1970s pragmatics focused attention on the function of sentence-level phenomena rather than their form, recasting them as “utterances” rather than “sentences.”
By the 1980s discourse analysts were examining many types of communication, often via conversation analysis: language in the courtroom, at dinner parties, between the sexes, in the workplace (to give a few examples). These analyses made it clear that discourse should be understood not as concatenations of S’s, but as language directed toward particular interactive and psychological purposes. These changes were not universally accepted. Some are unhappy about the extension of linguistics to this broad domain: they still think of language as a string of forms unrelated to function. Some feel that only thus can linguistics remain properly academic, unpoliticized, and “objective.” Some feel that only thus can linguistics be “scientific” and deserving of respect. But others feel that there is no such thing as language in the abstract, or in a vacuum; language can only be understood as the product of human need and desire. We note the enthusiastic attendance at language and gender conferences; the proliferation of journals on sociolinguistics, discourse, and pragmatics; and the fact that, when linguistics penetrates into the consciousness of the lay public, it is inevitably in one of its hyphenated variants (genderlect, “Ebonics,” language preservation). So it seems safe to say that “structure beyond the sentence level” is here to stay as an area of study within linguistics, just as it seems safe to say that women are here to stay as public speakers and makers of general meaning. The grumblings against both demonstrate their inevitability.
The changes in both fields since I wrote LWP mean that, if I were writing it today, I would go about it quite differently. We might still consider the occasional word or phrase in isolation and contemplate its significance: we might consider the continuing complaints about “male-bashing” (when historically very little was ever heard about “misogyny,” and much of that neutral or even positive); but we might note contrastively the acceptance of the notion of “spousal abuse” and “acquaintance rape” as negative, indeed criminally actionable, behaviors. We could comment more ambivalently on the frequent discussions of women’s desire for “having it all” (as a particularly female form of greed) or the rise of the “supermom” and the “soccer mom.” We might point out that such phrases indicate that women’s status is still not equal to men’s: men normally expect to “have it all,” and are not ridiculed for trying. There are no analogous “superdads” or “soccer dads,” or for that matter, “ballet-school dads.” A woman trying to juggle family and career is shunted to the “mommy track” or warned to stay home and tend to her family, lest the “Nannycam” prove that her child-care worker is up to no good. A man with both career and family is just a successful person. The “stay-at-home mom” (a term happily replacing “just a housewife”) is still normal—psychologically if not statistically. At the very least, even this sort of simple analysis indicates that language remains a window into the mind: the words we have constructed over the last couple of decades reflect our new uncertainties and our new possibilities.
But since we linguists have at our disposal today more powerful ways of connecting language to reality, a good way to rethink and recontextualize the claims of LWP would be to look not at words, or sentences, but at discourse on a larger scale. As a society, how do we talk about women, about feminists, about the genders and their respective roles?
We could look at discourse at a couple of levels: first, public discourse proper: what has been said, in the most public arenas, on the topics above. And, second, metadiscourse: what has been said about what has been said (reviews and think pieces). Contemplating either one, we should be at least as concerned with what is not said as with what is said.
We see that gender is on the front burner today as much as it was a generation ago. “Woman’s lib” is not a done deal, gender relations are still in ferment, evolution, and revolution, progress and regression. We are still determining what language is “normal,” what worthy of discussion or criticism; what novel attitudes, behaviors, or utterances mean, and how to categorize them. Our project is simply the reorganization of a primal aspect of our identities, our sexuality and gender—a reorganization facilitated and complicated by language.
A couple of articles that appeared in the New York Times Magazine while I was writing this essay illustrate our current dilemmas. While the topics they discuss existed in the 1970s, they would not have attracted the attention of such a prestigious venue. The content of the articles, the style in which they are written, and the fact that they have appeared when and where they have tell us something about our current ways of dealing with gender as individuals and as a culture.
First (chronologically) is an article by Margaret Talbot (2002) entitled “Supermom Fictions.” It discusses the conflicts today’s women encounter between family and career, the myth of “having it all.” Talbot is surprised at modern women’s surprise that life for working women tends to be “messy,” that there is always juggling to be done, and a ball or two is sometimes dropped. Talbot makes a reasonable point—life is that way and most of us learn to deal with it well enough—but asks why we believed it could be otherwise. The answer: we were misled by “feminism.” Her examples: a 1970s perfume ad (did “feminists” write it?) and Helen Reddy’s “I Am Woman,” which she calls “the anthem of ‘having it all’ “ (2002: 11).
Well, may be so. But the problem with Talbot’s analysis is that, as so often, women’s difficulties in the new world are ascribed either to women’s unreasonableness and greed, or to the impossible demands of “feminism.” But could there be a third answer? Perhaps women’s lives are messy because no one gives them much help. In the 1970s it seemed reasonable for women to compare their situation to men’s and ask: Why can’t—shouldn’t—female and male life trajectories be parallel? If men can have both families and careers without going crazy or being accused of greed, why can’t women? (And by the same token, if men want to spend many years of their lives staying at home with the children, why should that virtually always be seen as risible or unmanly?) Here as often the prevailing rhetoric prevents even reasonable and savvy people like Talbot from seeing that their arguments are deceptive in their failure to consider the need for deep social change.
Another article from the New York Times Magazine, “Fierce Encounters,” by Deborah Sontag (2002), questions the current treatment of spousal abusers by the legal system: automatic arrest and, typically, incarceration. It argues that many abused women would prefer for their abusers not to be arrested but to receive other forms of treatment. Many such women, says Sontag, feel that they bear at least some of the responsibility for the attack: they instigated it, they landed the first blow. Sontag (reasonably, I think) suggests that we reevaluate our absolutist stand: arrest as the only option. At the same time, she doesn’t discuss the opposite argument: that if you (essentially) leave it up to the woman to determine her partner’s punishment (or lack of it), you put her in a very vulnerable position, making her (should she opt for jail) even more in future danger. The unasked question is: Who decides who is responsible for things? Who gets the power to determine whose meaning?
This was a question that I couldn’t approach in LWP. Meaning, power, and responsibility can only be analyzed and discussed through the analysis of discourse. And, at the time, the women’s movement could not see things as we do today.
The pressing problem then was the fact that women had very little opportunity to speak for themselves in public so as to be heard and have an effect. If we got any response it tended to be belittling (as in, “You’re so cute when you’re mad”). To some, even today, this state of affairs is laden with comforting nostalgia: if you couldn’t take responsibility for anything, you couldn’t be blamed when it went wrong. But those nostalgic for the past forget that you also couldn’t get credit when something went right; and somehow, from the Book of Genesis to The Manchurian Candidate, women—powerless though in fact they might be—received disproportionate blame for all kinds of disasters throughout human history. At least if we had really had the power and the responsibility, that might have been more bearable.
Today we have achieved a significant amount of meaning-making power and responsibility. When women agitate to punish spousal abusers, they are listened to. When women try to ban prostitution and pornography, they are taken seriously. Both of these cases unfortunately pit women against other women: abuse experts against abuse victims; antiporn and antiprostitution advocates (often feminists) against pornographers, lovers of erotica, and prostitutes (many of them women). The questions raised by these paradoxes—who can speak for whom? Who should be listened to, and on what grounds?—turn out to resist simple answers.
In demanding that all abusers be punished, with no input from the abused spouse; and in arguing that all prostitution is the result of, and contributes to the creation of, women’s sexual slavery, powerlessness, and objectification, one (typically middle-class) group of women is—generally without meaning to—intruding on another’s (usually working-class) autonomy and self-esteem, saying in effect that no matter what these women say, they should not be listened to because they are the dupes of men, or they don’t know any better. Women, however, may find themselves in horrible positions as a result of powerlessness, prior abuse, dependency on men, and belief in misogynist stereotypes. That men once had the unquestioned right to make meaning for women, we can now clearly see as unjust. But how are feminists to deal with the temptation to make meaning for other women? Is it ever right? If so, when? If not, on what basis should policy decisions be made? These questions are about discourse rights, meaning-making, and the political function of language. They are difficult, but we have to find the tools to get answers to them.
One of the most valuable discovery procedures introduced into linguistics by transformational generative grammar is the use of the asterisk (*)—the recognition that nonoccurring cases define the limitations of the grammar and are therefore essential to a complete grammar. In sentence-level grammar, the asterisk could be used to mark a sentence (or a word, a combination of sounds, or a morphological ending) that could not occur in the language—one that would be recognized as aberrant by a fluent speaker. The rules of the grammar had to account for that gap in the paradigm. This was the point of entry of mentalism into linguistics: only the analyst’s knowledge of the language under investigation could predict whether or not a form could occur. Without this significant broadening and deepening of the tasks and goals of linguistics, the great achievements of TGG could not have been realized.
If we are to extend the domain of the grammar into extrasentential structure, how would the asterisk, or an equivalent, function? Just as feminist argument must be made more supple and subtle, more open to ambiguity and indeterminacy, so linguistic argumentation needs to develop in order to be useful in exploring this new terrain. The analysis of discourse has been devised within many fields. Some, such as conversation analysis, deriving from small-group sociology via ethnomethodology, are aggressively empirical and antimentalistic. Strictly empirical analysis has proved a valuable tool; but it has made it impossible to explore many of the most interesting questions about ourselves that linguistic investigation is ideally designed to answer. Yet—and this is the same paradox faced by the early-twentieth-century developers of the social sciences—if we permit the attitudes and mindsets of analysts to enter their analyses, we may be inviting irresponsible and unfalsifiable conclusions, like the first analysts of “exotic” languages, the missionaries and travelers whose dubious claims led to the requirement of antimentalism in the first place.
Some aspects of Chomskyan TGG may reasonably be brought into discourse analysis. When I wrote LWP, I made this assumption without understanding that it was controversial. I had been trained as a transformational grammarian. At the time, I was not aware that, to people not trained as theoretical syntacticians, the extension of TGG to interaction and discourse (even as it existed inchoately then) would be seen as controversial, even outrageous. (And syntactic theorists saw the extension of the assumptions of TGG to interaction and discourse as equally outrageous—for opposite reasons.) I wish I had appreciated the controversial nature of what I was trying to do, in bringing asterisks and paradigms into sociolinguistic examination. But I thought that language was language: if a method of analysis yielded interesting results, it was justified. If a method worked in one area, perhaps it would work in others.
Without the application of mentalistic methods and intuitionist discovery procedures, the work I wanted to do could not be done. For all its great usefulness, conversation analysis is of no help in dealing with paradigm gaps: it cannot talk about what does not occur in the transcript. Nor can it talk about why something is absent from the data: Is the omission due to an accidental gap or to rule-governed necessity? So intuition and introspection must play a role in at least some kinds of analyses. Criticism of my method that stops (as much of it does) at the outraged discovery that it is not entirely “empirical” is, therefore, not to the point. Linguistics must have both mentalistic and empirical methods at its disposal.
If we are to be able to make predictions about what arguments or contributions occur in contiguity, we must talk about items that are absent from the text although logically we might expect to find them. There might be any number of reasons for their absence. They might be like true asterisked sentences, logically or rationally uninterpretable: *Sincerity admires John. Or their absence might be due to culturally “normal” ways of thought: the absent argument might fail to occur to speakers because the stereotypes of their culture impose blinders on them, making them unable to see alternatives to the world they are assuming in their arguments.
Consider, as examples of this method of looking at discourse, two texts, or, better, a text and its metatext: Christina Hoff Sommers’s (2000) The War against Boys: How Misguided Feminism Is Harming Our Young Men and reviews of it in the New York Times (daily and Sunday Book Review), and another review in the neoconservative journal Commentary. What I find surprising about the entire set is less what it says than what it does not say. Sommers’s text is peppered with illogical conclusions, omitted considerations, and the like. Even purportedly negative reviews of the book do not confront these problems. The logical gaps in the basic text are especially curious: Sommers was trained as a philosopher and taught in a philosophy department before joining the conservative American Enterprise Institute. While philosophers are not known for statistical acumen or for painstaking collection of empirical data, they are expected to have a grasp of logical argumentation.
Sommers’s argument starts from observations (not her own) of reality. In recent years, in the U.S. educational system (especially in primary and secondary schools), boys appear by several measures to be falling behind girls. They drop out at a higher rate; their grades in most subjects are lower; they are more prone to get involved in drugs, crime, and violence; and they do more poorly on Scholastic Aptitude Tests and other objective tests of achievement and aptitude.
Let us grant that these claims are correct (although many are open to alternative analyses). Why, according to Sommers, have these things happened since the 1960s? She has a simple, snappy answer: the “feminization” of education at the hands of “feminists.” “Misguided” feminists, that is. “Misguided feminism” in her subtitle is syntactically ambiguous: Is the noun phrase derived from a restrictive (“[that kind of] feminism that is misguided”) or nonrestrictive (“feminism, which is [necessarily] misguided”) relative clause? From the evidence I suspect the latter.
Sommers claims first that primary and secondary school teachers, who tend to be female, have imposed female standards of behavior on their charges. (It is harder for little boys to sit still than for little girls, for instance, and school demands that children sit still for long periods.) Here Sommers makes use of obvious evidence: of course it’s true that women dominate elementary and secondary school teaching. (A feminist plot? Hardly!) And it is likewise true that the kind of conduct demanded in school comes easier to girls than boys. But this has always been true—schools demanded the same kind of behavior even when teachers were male and women were not educated. Some might call it “middle-class training,” rather than “feminization.” Second, Sommers asserts that feminists and sympathetic educational and psychological theorists have “pathologized” normal rambunctious males. The book begins with a dazzling riff on this theme: “It’s a bad time to be a boy in America. As the new millennium begins, the triumphant victory of our women’s soccer team has come to symbolize the spirit of American girls. The defining event for boys is the shooting at Columbine High” (2002: 13).
A very grabby beginning, to be sure, and a scary one if you are male or a parent of young males. But an opening also full of hyperbole likely to encourage illegitimate conclusions. First, the events at Columbine cannot be—and never were—equated with “normal” male behavior. Next, Sommers makes a kind of implicit equation, soon to become an explicit part of her argument: The “pathologization” of boys (as demonstrated by Columbine’s being their “defining” event) is a direct outcome of the “triumph” of girls. Post hoc, ergo propter hoc. Even if we take her two statements to be true at face value, is the second the necessary result of the first? Is it even related to the first? Is gender equity (as Sommers’s argument implies) a zero-sum game: females win only if males lose?
The questions go deeper. Is it really true that these are the defining events of femaleness and maleness at the turn of the millennium? Certainly both were important, and both received a great deal of media attention—perhaps too great in the case of Columbine. Certainly most Americans cheered Mia Hamm and her team—but their victory defined one set of options for women. It did not exclusively symbolize the spirit of American girls. It—along with other events—did symbolize the fact that women have made great strides and are now in a position to do things they had not formerly been able to do. But there are other positive defining events for women: the increasing participation of high school girls in science contests, the electoral victory of Hillary Rodham Clinton and other women, the appointment of Madeleine Albright as Bill Clinton’s secretary of state. And there are negatives, like the great success of the reality show The Bachelor.
But if those are evidence of women’s success, we can easily find their male equivalents: many Olympic victories by all-male teams; the election victory of George W. Bush and his appointment of Colin Powell as secretary of state; and others too numerous to mention. If females and males are roughly equal in the population, then the achievements of men are still, proportionally, greater than those of women. So it is hard to claim statistically that men have lost by women’s gains. Perhaps they get a sliver less of the pie. But their share is certainly no less than 50 percent of the total. What has begun to shift is society’s definition of the normal or unmarked for women. Options previously unmarked for males (e.g., success in sports and politics) are increasingly less marked than they used to be (but still marked) for women. It’s hard to see how this shift leads inexorably to Columbine, but for Sommers it does just that.
Sommers makes the claim that boys are now defined by Columbine. Columbine, as a horrible event, received an inordinate amount of attention. Some of that attention focused on certain kinds of boys: on the one hand, bullies; on the other, the targets of the bullies, who are (very occasionally, though you’d never guess this from the coverage) driven to avenge themselves in hideous ways. But there was, to my recollection (and Sommers provides no counterexamples) no media discourse to the effect that Columbine and similar events represent a “pathology” common to all or even the majority of young males—any more than there was a claim that all or most women could have won Olympic medals. At most, we might say that as the Olympic victory stood for the best women could achieve (in one area at any rate), Columbine illustrated the worst males could descend to. But that is far from a “defining event.”
The rest of Sommers’s case depends on just such flimsy argumentation. Its bases often look sturdy: tables of figures and authoritative quotations. But the juxtapositions that make the arguments superficially persuasive and the claims that Sommers derives from her data are all questionable. Yet they receive no serious questioning, even by the most purportedly “negative” reviewers.
Let us, for the sake of argument, take her statement (A) at face value: boys are failing in school and in society. Let us also accept her other premise (B): that, in part driven by feminism, women are becoming more successful in both areas. From those truths, can we safely draw the conclusion that A is the result of B?
Of course not: the relationship between A and B could take any of several forms. Further research is needed to determine which, if any, is the case:
1. Sommers’s claim: B caused A.
2. A caused B.
3. Neither caused the other. The occurrence of both, at around the same time, is coincidence.
4. Neither caused the other. Both were caused by a third factor, C, not yet identified.
And there are probably other possibilities. No reputable scientist would stake her career on (1) being correct without examining the other possibilities and eliminating them. No reputable philosopher would support an argument based on logical errors. I’m not a philosopher. (I don’t even play one on TV.) But I can detect a logical flaw when it is big enough to drive an SUV through.
But the flaws in Sommers’s arguments are not the strangest thing in this set of texts. What is particularly odd is that few if any of the commentators on the book have noted these flaws. The negative reviews take issue with the interpretation of particular cases, or with Sommers’s taste for hyperbole, but they never look at Sommers’s overall logic. This is odd since (as any author knows) a reviewer likes nothing better than picking holes in authorial arguments: the bigger the hole, the smarter the reviewer looks. A neoconservative (Finn 2000) writing a strongly positive review in the neocon journal Commentary describes Robert Coles’s review (2000) in the New York Times Book Review as “thrashing” Sommers. On this evidence, we might assume that Coles’s review was particularly savage. But it isn’t, by a long shot. He criticizes Sommers’s “high, hectoring pitch” in her attack on Carol Gilligan. But his criticism is chiefly that we should be concerned about the future of girls and boys alike (as indeed we should). There is no analysis of Sommers’s arguments themselves and nothing that qualifies, in my lexicon, as “thrashing.”
How can we explain the absence from the public discourse of careful analysis of Sommers’s arguments? I can only suppose that those flaws have gone unnoticed by critics and remain inaccessible to most readers. Here is a gap in the paradigm, a missing piece, analogous to an asterisked sentence in the syntactic inventory of English. The absence needs to be explained, in order for the discourse (basic and meta-) of The War against Boys to become “grammatical”—to make sense.
Here’s one explanation: the culture still perceives male achievement towering over female achievement as “normal,” that is, not requiring explanation, and therefore misses the first point, that males aren’t really falling behind—females are simply beginning to catch up. They don’t see that the true norm is for both women and men to succeed in equal proportions, at the Olympics or anywhere else. So when the percentage of success edges a little closer to that mythic 50 percent apiece, that (rather than rampant inequality) is construed as abnormal, eliciting fear and requiring “explanation.”
If we turn to the problem of male pathologization, neither Sommers nor her critics notice that, over history, females have been “pathologized” to a far greater degree than males are now. Think of Freud; think of the Malleus Maleficarum; think of millennia of misogyny. What is shocking in current discourse, to Sommers and similar traditionalists, is that males are now receiving some of the pathologization, if indeed they are. Women are still the recipients of at least their share. We are blind to female pathologization, because it’s “normal.” But serious social critiques, and critiques of those critiques, need to be able to perceive the abnormality in normality, and comment on the roles and expectations of both females and males from a fully objective position. We need to remove the blinders that millennia of misogyny have normalized. While none of us can achieve this yet, we must at least be aware of our unawareness and its causes.
Just as talking about widow versus widower, master versus mistress, and differences in tag-question usage were ways that the linguistic sophistication of thirty years ago provided to enable me to demonstrate inequality between female and male roles in society, today’s linguists have additional options, permitting us to examine, through language, the possibilities open to women and men in different and deeper ways. Our work is still before us.
Coles, Robert (2000). Boys to men. New York Times Book Review, June 25:20.
Finn, Chester E. Jr. (2000). The war against boys. Commentary 110(2) (September):68.
Friedan, Betty. (1963). The feminine mystique. New York: Dell.
Sommers, Christina Hoff (2000). The war against boys: How misguided feminism is harming our young men. New York: Simon & Schuster.
Sontag, Deborah (2002). Fierce entanglements. New York Times Magazine, November 17:52.
Talbot, Margaret (2002). Supermom fictions. New York Times Magazine, October 27:11.