Girls, Uninterrupted

Long ago (in fact, seven years to the day), when the paint on the walls of the Language Lounge was still fresh, we talked about the ways in which sexism is reflected in the lexicon of English, using word maps in the Visual Thesaurus. The occasion was Women's History Month, and now, since that occasion has rolled around again, it's a suitable time to have another look at gender inequality in language.

New resources are available today for research into language use that were not around seven years ago. Two of them are  Google Ngrams, which enable us to look at trends of word usage in books over time, and the Corpus of Historical American English (COHA), which does a similar thing, showing us snapshots of changes in American English in all media through time. Corpora of English generally have also grown by leaps and bounds, along with the tools available to query them.

Starting in the 1960s with the influence of feminism, and in the 1970s with political correctness (the two have somewhat overlapping agendas), there has been an effort to replace the use of gender-biased terms and terms that inaccurately apply to one sex with more inclusive terms. This seems to have had an effect. In COHA, for example, the use of "mailman"  peaked in the 1980s:

However, the same corpus does not show a corresponding increase in the gender-neutral term letter carrier that is now favored by the Post Office. A look at stewardess, another term that was nominated for retirement in the early days of feminism, peaked in COHA in the 1970s. Today's preferred term, flight attendant, shows a steady increase to accompany the decline of stewardess.

The picture in Google Books shows a similar trend, though with stewardess still outnumbering flight attendant in books, and with a somewhat alarming (and possibly anomalous) uptick in the use of stewardess starting in 1999.

More encouraging news comes from other languages: a recent article on the BBC News website reports that a town in France is trying to rid itself of mademoiselle; the same article surveys the removal of Fraulein from German and the diminution of miss in English.

So far, so good. As we noted in the Lounge seven years ago, however, English — and perhaps all natural languages — contains a glut of words that cast females in a less than flattering light, and most of these words seem to be more synonym-rich than their male counterpart words, if such words exist. It's also the case that some words applying to females who engage in a particular behavior carry largely negative connotations, while the corresponding terms designating males doing the same thing do not carry a stigma. Look, for example, and the respective word maps and definitions for vamp and Lothario. While these entries and others like them in dictionaries might seem to perpetuate unfortunate stereotypes, dictionary publishers would do their users a disservice by removing these words from notice: the job of the dictionary is to define, and provide information about the use of words. Suppressing words or information about their usage only amounts to killing the messenger.

Here's an Ngram (1950–2000) of words in books that don't have a precise male equivalent, and that characterize or define a certain kind of female.

The usage of five of the words — coquette, jezebel, sex kitten, shiksa, and vamp — do not show any appreciable trend and are all low frequency in books. Are the reversed frequencies of concubine and  spinster from where they were 60 years ago significant? It's hard to say what accounts for the difference, but also hard to come up with an explanation that suggests progress for women.

Collocations can also tell us a lot about how people think about and use words. We looked recently at the contexts of some contrasting wordsin a 3.2-billion word corpus of English (including all dialects). Here, words that typically modify gender-specific nouns show a strong tendency toward disparate treatment. In all cases shown in the table below, the modifying word either did not appear with the opposite gender word, or occurred with much lower frequency:

In the same corpus, we looked at the nouns girl and boy for verbs that typically predicate them.

If you're wondering about shrink, it occurs in various phrasal verbs: shrink from, shrink back, shrink away.

The patterns above confirm what we already know: Anglophone males and females are socialized differently: they are not typically regarded or judged according to the same criteria, nor are their behaviors typified in the same way. There is also a strong suggestion that the beauty mandate — the idea that women are expected to maintain an appearance that is pleasing, especially to men — is still alive and well in English.

Is this sexism or simply a reflection of our reality? Of course, it can be both of these. What explains the persistent inequality in the sexes, as reflected in the printed word? A handy and facile explanation is "blame the patriarchy." Men have dominated so many aspects of culture for millennia, and that domination would certainly include control over the printed word, through writing more than women have written historically, and by being the main gatekeepers of publishing until very recent times. But this explanation further requires an assumption that men somehow have it out for women. Is there evidence of that? You might want to look at lexicographer Jonathon Green's answer to a question on Quora, "Why are expressions for having sex with women often synonymous with killing them?" It seems more likely that the patterns we observe in written language really are just representative of the way people — both men and women — think and behave.

Back when life was just primordial slime, some cells cleverly found a way to outsource the production of haploid gametes, and thus was sexual reproduction born. Along with it, males came into existence, and ipso facto, females. In our species and in every organism that uses sexual reproduction, the things we call females retain the sine qua non job of reproduction, the bearing of young, so it seems an irony that in our world today, and in our chief symbolic system — language — the gender that is essentially a spinoff is so dominant. The trope is repeated at a different scale everywhere you look: in the Biblical story of creation, in languages that use null morphemes for masculine gender and inflected forms for feminine, and in word pairs that make the feminine look derivative, such as man/woman, and male/female (although etymologically the picture is not so straightforward). It may be that the hand the rocks the cradle rules the world, but it happens in a way that flies under the radar of language analysis.