Is the fuchsia female?

Here’s something that made me laugh recently when somebody shared it on Facebook. 83778263_3912267798786914_6847277020674523136_n

It’s from a page called ‘Men’s Humor’, which I initially found surprising, since on the face of things it’s a joke at men’s expense: it’s saying men don’t normally get to name eyeshadow colours because their colour vocabulary is so limited. Once they’ve used up basic colour terms like ‘orange’, ‘pink’ and ‘purple’, they’ll be forced to fall back on unappealing comparisons with meat, ‘spoiled milk’ and (ew) ‘diarrhoea’.

But on reflection that’s only part of what the joke is doing. It’s also indirectly poking fun at the names conventionally given to make-up colours, which are absurd in a different way. From that perspective the joke is on women, who are suckers for a flowery name–they wouldn’t buy an eyeshadow in ‘pork’, but they’d be happy to shell out for one in ‘tea rose’ or ‘autumn sunrise’. When it comes to colour-names, cluelessness is proof of manliness. Real men don’t know words like ‘taupe’; they have more important things to worry about than the difference between ‘lavender’ and  ‘mauve’.

Some readers may recognise that last bit an allusion to Robin Lakoff’s 1970s classic Language and Woman’s Place. Lakoff put non-basic colour terms like ‘taupe’, ‘lavender’  and ‘mauve’ into the category of ‘women’s language’, a register which in her view both derived from and reinforced women’s subordinate status in society:

Since women are not expected to make decisions on important matters, like what kind of job to hold, they are relegated the non-crucial decisions as a sop. Deciding whether to name a color ‘lavender’ or ‘mauve’ is one such sop.

These distinctions are regarded by men as trivial; and once they have been defined as the province of women, any man who does display an interest in them invites questions about his masculinity.

[A] woman may say ‘The wall is mauve’ with no one consequently forming any special impression of her as a result of the words alone; but if the man should say ‘the wall is mauve’, one might well conclude he was either imitating a woman sarcastically, or a homosexual, or an interior decorator.

So in addition to being something men ‘relegate’ to women because they consider it beneath their notice, colour-naming might be something men actively avoid, because they don’t want to be perceived as effeminate or gay.

As usual, though, what got into wider circulation was not Lakoff’s analysis of the cultural conditions that might produce gender differences in colour vocabulary, but just the ‘fun fact’ that women know more colour terms than men. The ‘if men got to name eyeshadow colors’ joke shows that nearly 50 years on, this is still part of our cultural common sense. Does research done since the 1970s bear it out, though? Many of these ‘fun facts’, after all, continue to be repeated long after they’ve been debunked by science. We’ve known for years that the ever-popular ‘Women talk more than men/use x times as many words per day as men’ is just straight-up BS. And other claims, including some of Lakoff’s, have turned out to be more complicated than they originally seemed. So, what’s the story about colour terms?

The short answer is that research done since the 1970s generally has supported the claim that women know more colour terms than men. There are also some finer-grained differences in the kinds of terms men and women produce when their colour vocabulary is tested. But what’s behind these differences is still a matter of debate.

By way of illustration, let’s take a closer look at one fairly recent piece of research in this tradition: an experimental study which was presented at a conference in 2012. It was conducted with 272 English-speaking subjects (159 women and 113 men), who were each asked to name a series of 20 colours selected randomly from a set of 600. The responses were ‘unconstrained’, i.e. participants could give each stimulus colour a name entirely of their own choosing. This procedure produced over 5000 responses, containing 1226 different colour terms. 29 per cent of these were ‘basic colour terms’ (red, blue, green, etc.), 23 per cent were single-word non-basic terms (mauve, scarlet), 42 per cent were two-word descriptions (a one-word term plus a modifier, like ‘bluish green’ or ‘pale yellow’) and 6 per cent contained three words or more (e.g. ‘pillar box red’).

As expected, there were differences between men and women–and there was more to this than the familiar finding that women produce more colour-names overall. Though the two sexes made similar use of basic colour terms, beyond the basics their choices diverged. Women chose more single-word terms like ‘mauve’, whereas men gravitated more to the two-word ‘bluish green’ type. The most frequently-used non-basic terms were also different for each sex. For instance, the top 20 female choices included ‘peach’ and ‘fuchsia’, neither of which featured in the men’s top 20; conversely, the male top 20 included ‘cyan’ and ‘magenta’, which did not make it onto the female list.

In their discussion of these findings the researchers remark that the colour men labelled ‘magenta’ is the same one for which women preferred the ‘fancy’ term ‘fuchsia’. I find this comment interesting, because to me it isn’t obvious that ‘fuchsia’ is any ‘fancier’ than ‘magenta’. On the contrary, I would argue that the two terms are not just equally fancy, they are fancy in exactly the same way. Both are derived from proper names: ‘fuchsia’, the name of a plant genus, honours the botanist Leonhard Fuchs, while ‘magenta’, the name given to an aniline dye invented in 1859, commemorates France’s victory at the battle of Magenta. The dye-colour was inspired by the fuchsia plant, and before its (French) inventor got carried away by patriotic pride he had intended to call it ‘fuchsine’.

We might suspect–in fact, I find it hard not to–that the researchers’ description of ‘fuchsia’, but not ‘magenta’, as ‘fancy’ reflects the pre-existing gender stereotype according to which women, but not men, use fancy colour terms. Similarly, they note that the part of the spectrum which women typically categorised as ‘turquoise’ was segmented by men into ‘turquoise, cyan and light blue’–but they make no comment on the ‘fanciness’ of ‘cyan’, nor on the fact that in this case it was men who made finer distinctions.

You may already have thought of an explanation for men’s more frequent use of ‘cyan’ and ‘magenta’. These terms have a tech connection: they’re two of the four ink colours used in colour printing. ‘Peach’ and ‘fuchsia’, by contrast, are terms you’d be more likely to encounter if you were shopping for lingerie. Such observations might suggest that women’s more extensive colour vocabulary is not a product of nature but an artefact of culture. Men don’t shy away from non-basic colour terms if they belong to a male domain; it’s just that there tends to be more use of elaborate terminology in domains which are coded as female.

Until recently, this cultural explanation was favoured not only by feminists, but also by most scientists. But in sex-difference science generally there’s been a resurgence of interest in biological explanations, and it’s been suggested that women’s more extensive colour vocabulary might reflect their naturally superior ability to discriminate colours visually. This isn’t an easy argument to settle: there’s an obvious chicken and egg problem. For instance, referring to a 1990 study of mail order catalogues which found ‘more variation in the terms describing women’s clothing than men’s’, the authors of the 2012 study suggest that the catalogue producers were ‘taking advantage’ of women’s greater facility for discriminating and naming colours. Yet it could equally be argued that women develop this facility precisely because the products they consume come in a wider range of colours and are described using a wider range of terms.

One way to investigate this further might be to consider variables other than sex/gender. Men and women, after all, are diverse rather than homogeneous, in ways that might be expected to make a difference to their use of colour-terms–especially if this is primarily a cultural thing. Pursuing that line of thought, I discovered a study published in 2007 which found a difference between older and younger men: while the older men, as usual, produced a limited range of non-basic terms, the younger men were far more similar to women. There is more than one possible explanation for this finding, but the most plausible one, arguably, is cultural change, both in gendered patterns of consumption and in ideas about masculinity. I found some indirect support for this in current clothing catalogues. When I looked at a (small) sample it was clear that the colour-terms used in the menswear and womenswear sections were less sharply differentiated than they reportedly had been in 1990. Today’s fashion retailers seem confident that their male customers, like their female ones, will buy shirts in ‘sage’, ‘forest’ or ‘acqua’ rather than just ‘green’.

But another variable that should probably be in the mix here is social class. The catalogues I looked at (which had come unsolicited through my letterbox and ended up in my recycling bin) were clearly addressed to a certain segment of the market, and that was reflected in their language as well as their prices. Words are among the semiotic resources companies use to construct an image that will appeal to, or flatter, their target customer; elaborate or unusual colour terms often seem to be associated with more ‘aspirational’ brands.

Consider, for instance, paint shade-cards. They’re a vast repository of ridiculous colour-names, but if you compare, say, the upscale Farrow & Ball range to B&Q’s more basic house-brand, you will find they are ridiculous in different ways. F&B sells shades called things like ‘Railings’; B&Q offers shades like ‘Pumpkin Pie’. Both might be described as whimsical, but one is–to quote a painter and decorator I once discussed this with–‘poncier’ than the other.

Of course, no one participating in a colour-naming experiment would be likely to produce something as obscure as ‘Railings’. But participants’ choices might still be influenced by the ‘ponciness’ factor. And this wouldn’t necessarily be just a direct reflection of their own class position. As I explained in an earlier post about swearing, there’s quite a strong tendency for middle- or upper-class linguistic norms to be symbolically associated with femininity while working-class norms are associated with masculinity. We might therefore expect men, including middle-class ones, to be more concerned than women about avoiding terms they consider ‘poncy’. This avoidance is central to the humour of the joke eyeshadow palette: in a context where poncy words are standard, the author of the joke has substituted aggressively plain, down-to-earth or crude alternatives.

The maleness that’s both parodied and celebrated in this joke is defined not only in opposition to femininity, but also by contrast with certain kinds of masculinity (for instance, anything too gay or too educated). Colour terms, it turns out, are a great resource for this kind of social symbolism. But precisely for that reason I think it’s hard to know what we can safely conclude from studies which claim to show that women know more colour terms. I’m not disputing the (very consistent) finding that women generally produce more terms; but is that simply because men’s vocabulary is smaller? If men write ‘pinkish-purple’ rather than ‘mauve’, does that mean ‘mauve’ isn’t in their repertoire, or does it mean they’re reluctant to present themselves as the kind of man who uses words like ‘mauve’?

There’s also a more fundamental question, one I always seem to end up asking when I post about any kind of sex-difference science: why does any of this matter in the first place? So what if you say magenta and I say fuchsia? Let’s call the whole thing off!

But while I don’t care about the question itself, I do care about what’s behind it. When a (real or alleged) sex-difference becomes the object of intense scientific and/or popular fascination, that often has less to do with its real-world importance than with the symbolic meaning we project onto it. We seize on certain generalisations because they fit with our beliefs about what men and women are or should be like. Sometimes the generalisations are completely false; sometimes, as in this case, they’re more robust; but either way, they contribute to the patriarchal project of keeping men and women in their prescribed, different-and-not-equal, places. And even when it’s funny, that’s no joke.

 

2019: (not) the end of an era

In a couple of days’ time we’ll be marking not just the passing of another year, but by most people’s reckoning the end of the current decade. All kinds of commentators will be looking back over the last ten years, and many will turn to language (or at least, vocabulary) as a source of insight about what mattered in the 2010s. They’ll remind us this was the decade that gave us ‘Brexit’, ‘fake news’, the ‘gig economy’ and ‘influencers’; it was when ‘climate change’ became the ‘climate emergency’, and when global protest movements formed around the hashtags #BlackLivesMatter and #MeToo.

This approach to documenting social trends—epitomised by the annual ritual in which dictionaries select a Word of the Year (WOTY)—has its limitations. It doesn’t capture the preoccupations of the speech community as a whole (if I quizzed a sample of my neighbours on the vocabulary items listed in the last paragraph, asking ‘have you come across this expression, can you define it, have you ever used it yourself?’, I suspect that only one item—‘Brexit’—would get affirmative answers across the board). It also imposes artificial temporal boundaries on a much messier reality: though some notable linguistic developments can be tied to specific events and dates, most don’t fit neatly into a single year or even a decade. In addition, the search for zeitgeist-defining terms encourages a focus on what’s new or what’s changed, though arguably it’s no less important (and may even be more revealing) to consider what has stayed the same.

That last point will be reflected in my own attempt to summarise the decade. When I look at this blog’s archive (over 100 posts going back to 2015) I see more continuity than change. The specifics differ from year to year, but the same general themes recur; and I’m sure they would have featured just as prominently if I’d started blogging in 2010. So, in this post I’m going to pick out (in no particular order) my top five recurring themes, using the way they presented themselves in 2019 as a starting point for some reflections on what has—or hasn’t—changed during the 2010s.

1. The return of crass sexism

In January this year, after belatedly learning that she had died, I wrote a post about the writer and editor Marie Shear, who will be remembered for her definition of feminism as ‘the radical notion that women are people’. She was also a sharp and uncompromising critic of sexist language, and the author of a widely-read piece which described what she called its ‘daily toll’: a continual insidious wearing down of women’s dignity and self-esteem whose cumulative effects she thought were too often underestimated.

Shear wrote this piece in 2010, at a time when sexist language had become an unfashionable topic. In the noughties some writers had argued that the overt sexism feminists had criticised in the 1970s was no longer a major issue: it survived only among ageing dinosaurs (like the surgeon in Shear’s opening anecdote) who would not walk the earth for much longer. Attention had turned to the subtler forms of sexism that were said to be more typical of the postmodern, ‘postfeminist’ era. But while postmodern sexism is still a thing (particularly in advertising and branding), the 2010s turned out to be the decade in which crassly sexist and misogynist language returned with a vengeance to the public sphere.

I say ‘with a vengeance’ because the crassness was more extreme this time around. In the past, the norms of mainstream public discourse discouraged the grossest expressions of contempt for women—they were reserved for taboo-busting radio shock jocks and men talking among themselves. But the 2010s saw the rise of public figures–most notably populist ‘strongman’ leaders like Donald Trump and Rodrigo Duterte–whose speech was not constrained by older notions of decorum (or gravitas, or honesty, or any other traditional public virtue). Crude misogyny is part of these men’s brand: I’ll leave aside Trump’s infamous reference to ‘grabbing [women] by the pussy’, since it wasn’t intended for public consumption; but think of his comment, made on CNN in 2015, that the journalist Megyn Kelly ‘had blood coming out of her wherever’ (her offence had been to question Trump about his earlier references to women as ‘dogs’ and ‘fat pigs’). In 2019 Britain got its own imitation strongman leader, Boris Johnson, who specialises in the crass sexism of the public school playground, denouncing his (male) opponents as ‘girly swots’ and ‘big girls’ blouses’.

But you didn’t have to be a political leader to broadcast male supremacist messages to a global audience. The internet gave ‘ordinary’ men with a grudge against women—incels, pick-up artists, MGTOWs et al—a megaphone for their misogyny (and for the violent fantasies which some of them, like Alek Minassian, would go on to enact in reality, making 2018 the year when mainstream, nonfeminist commentators started to talk about  ‘incels’, ‘misogyny’ and ‘toxic masculinity’). Not dissimilar messages also circulated under the banner of ‘harmless fun’. For instance, one of the items I reproduced in a post about greeting cards this year bore the message: ‘Women. Can’t live with them, can’t shoot them and bury them in the garden’. Which brings me to the second major theme of the decade…

2. The linguistic (mis)representation of sexual violence

Any feminist survey of the 2010s will be bound to treat the #MeToo movement as one of the most significant developments, if not the most significant, of the last ten years (the hashtag would be an obvious candidate for the feminist Word of the Decade.) But #MeToo also dramatized what for me was probably the most troubling linguistic trend of the decade: an increasingly marked reluctance on the part of institutions—educational establishments, the criminal justice system and above all the media—to name sexual violence and those who perpetrate it without equivocation, euphemism and overt or covert victim-blaming.

In 2017 and 2018, as #MeToo allegations multiplied, the media converged on a couple of phrases which were repeated ad infinitum: the whole spectrum of abuse, up to and including rape, was now covered (or covered up) by the bland euphemisms ‘inappropriate behaviour’ and ‘sexual misconduct’. This vague, affectless language was a boon to anyone who wanted to argue that the women making allegations were lying, exaggerating, reframing consensual exchanges of sexual for professional favours as abuse, or simply making a fuss about nothing (‘can’t men even flirt now without being accused of misconduct?’)

In 2019 we saw a similar pattern in reports on the case of Jeffrey Epstein, who was charged with child abuse and trafficking (though he killed himself in prison before he could stand trial). Oxymoronic terms like ‘underage women’ were used to describe girls who at the time were 14 or 15; and when attention turned, after Epstein’s death, to the actions of other men the victims had named, the words ‘rape’, ‘assault’ and ‘abuse’ were conspicuous by their absence.

Earlier in the year, most news outlets had even resisted using those words without qualification when reporting on the case of a severely disabled woman who unexpectedly gave birth in the care facility where she had spent most of her life. Though she could only have become pregnant as the result of a criminal assault—her vegetative state rendered her legally and medically incapable of consenting to sex (and also of lying about it)—reporters’ first impulse was still to hedge their statements with doubt-indicating words like ‘alleged’, ‘apparent’ and ‘possible’.

But in the last part of 2019 there were some memorable protests in which feminists harnessed the power of the R-word. In Spain, women who were disgusted by the verdict in a gang-rape case—the perpetrators were convicted only of ‘abuse’, because they had not used physical force against their barely-conscious victim—took to the streets to protest, shouting ‘no es abuso, es violación’ (‘it’s not abuse, it’s rape’). And in Chile on November 25, the International Day for the Elimination of Violence Against Women, women gathered outside the Supreme Court to perform a chant which has since been taken up around the world (its title in English is ‘The rapist in your path’), calling attention to the way individual men’s ability to rape and kill with impunity depends on a larger culture of complicity and victim-blaming.

In acknowledgment of the power of these protests, and because nothing has made me angrier this year than reading about men ‘having sex’ with 14-year olds or police investigating a ‘possible/alleged assault’ on a woman who gave birth while in a vegetative state, I choose ‘rape’ and ‘rapist’ as my words of the year for 2019.

3. Curious contradictions: the case of the authoritative woman speaker

Among the themes which have recurred in each of the four-and-a-half years of this blog’s existence are two that, taken together, embody a stark contradiction. On one hand, women are constantly castigated because their speech allegedly ‘lacks authority’: how can they expect to be taken seriously when they’re forever apologising and hedging every request with ‘just’? On the other hand, women who do speak with authority are constantly criticised for being ‘angry’, ‘abrasive’, ‘arrogant’, ‘bossy’, ‘immodest’, ‘shrill‘, ‘strident’ and generally ‘unlikable’.

This familiar contradiction was on show again this year. We had more of the same old bullshit about ‘sorry’, ‘just’ and other female ‘verbal tics’, and more complaints about high-profile women leaders being ‘strident’ (teenage climate activist Greta Thunberg), bossy and ‘self-righteous’ (Liberal Democrat leader Jo Swinson), ‘angry’ (Democratic presidential hopefuls Elizabeth Warren and Kamala Harris) and ‘unlikable’ (every woman in the race for the Democratic nomination).

More unusually, two women—Marie Yovanovitch and Fiona Hill—attracted praise for their authoritative testimony during the proceedings that led to Donald Trump’s impeachment by the US House of Representatives. Unfortunately, I don’t think this is a sign of things to come. The positive reception Yovanovitch and Hill got was linked to their status as non-partisan public servants, and the same courtesy is unlikely to be extended to any of the female politicians who are still in the running for next year’s presidential election. It’s one thing for a woman to have authority thrust upon her, but actively seeking it is a different matter. Powerful and politically outspoken women will still, I predict, be ‘unlikable’ in 2020.

4. Studies show that women are rubbish

The training course where women executives at the accounting firm Ernst & Young learned that women’s brains are like pancakes and men’s are more like waffles (as reported in October by the Huffington Post) almost certainly wasn’t based on any actual science (or if it was, whoever designed the course should get the Allen and Barbara Pease Memorial Award for Neurobollocks). But while science can’t be held responsible for all the sexist drivel that gets talked in its name, it shouldn’t get a free pass either.

In the 90s and noughties we were endlessly told that women were naturally better communicators than men, but in the 2010s there’s been something of a shift: there are, it transpires, certain kinds of communication in which it’s men who are hard-wired to excel. This year, for instance, a widely-reported meta-analytic study put together the findings of 28 experiments investigating the proposition that men are better than women at using language to make others laugh. The conclusion was that men really do have more ‘humor ability’ than women, probably because this ability is ‘correlated with intelligence’, and as such is a useful diagnostic when females assess the fitness of potential mates. (Ah, evolutionary psychology: a 90s/noughties trend which sadly didn’t die in the 2010s.)

It isn’t hard to pick holes in these studies; but while it’s important to interrogate specific claims about why women are rubbish at [fill in the blank], we also need to ask more basic questions about why so much research of this kind gets done in the first place. What interests are served by this unceasing quest for evidence that sex-stereotypes and the judgments based upon them reflect innate differences in the abilities and aptitudes of men and women?

Another study published this year on the subject of gender and humour found that women who used humour in a professional context were perceived to be lacking in competence and commitment. This had nothing to do with their ‘humor ability’: in this study, subjects watched either a man or a woman (both actors) giving exactly the same scripted presentation, complete with identical jokes. But whereas the man’s humour was perceived as enhancing his professional effectiveness, the woman’s was perceived as detracting from it.

What this illustrates is the general principle that the same verbal behaviour will attract different judgments depending on the speaker’s sex. Judgments about women and humour are similar to judgments about authoritative female speakers, and they embody the same contradiction: women are widely disparaged for lacking humour, but those who don’t lack humour are disparaged as incompetent lightweights. What explains this effect–‘heads men win, tails women lose’–isn’t women’s behaviour or their natural abilities: it’s a consequence of sexism, which science too often reinforces.

5. The War of the W-word

In my round-up of 2018 I wrote at length about the increasingly contested status of the word ‘woman’, whose definition, use, avoidance and even spelling prompted heated arguments throughout the year. This isn’t totally unprecedented: as I’ve said before (beginning in my very first post), the W-word has a longer record of causing controversy than many people realise. But its current contentiousness is linked to something that is specific to the 2010s—the rise of a new politics of gender identity, which has also influenced language in other ways. It’s a development that divides feminists, and the kind of conflict we saw so much of in 2018 will undoubtedly continue in the 2020s.

In 2019, however, the most notable controversy about ‘woman’ was much more old-school. It started when the feminist Maria Beatrice Giovanardi was looking for a name for a women’s rights project she was working on. In search of inspiration she typed the word ‘woman’ into Google, and was shocked when her search returned a series of online dictionary entries full of offensive synonyms (‘bitch’, ‘chick’, ‘filly’) and insultingly sexist examples of usage (‘one of his sophisticated London women’; ‘male fisherfolk who take their catch home for the little woman to gut’). When Giovanardi started a petition calling on Oxford Dictionaries to change their entry, her intervention attracted extensive media interest, and by September the petition had over 30,000 signatures.

This is a good illustration of the point I made earlier—that the advent of new concerns does not mean the old ones become irrelevant. What Giovanardi drew attention to is one of many examples of the quiet survival of ‘banal sexism’, the kind of tediously familiar, low-level stuff whose ‘daily toll’ Marie Shear warned us not to underestimate. In the past five years I have come to agree with Shear. It’s striking to me that many of the most popular posts on this blog have been about things that would never register on any trend spotter’s radar: old chestnuts like ‘should women take their husband’s names?’, and ‘does swearing make women unattractive?’, which I could equally have written about at any time in the last 40 years, are still significant issues for many women. If feminism had started a linguistic to-do list in 1975, it would certainly be a lot longer now, but very few of the original items would actually have been crossed off.

So am I saying the next decade will look a lot like the last one? Yes: though change is a constant, in language and in life, what we mostly see is evolution, not revolution. That was true in the 2010s, and—barring some catastrophe that puts an end to civilisation as we know it—it will also be true in the 2020s. I know that’s not much of a prediction, and maybe not the happiest of thoughts when you look at the current state of the world, but there it is: we are where we are, and all we can do is keep going. I wish readers of this blog a happy new year/new decade (thanks as always to all the other feminists and/or linguists whose work I’ve drawn on in 2019), and I’ll see you on the other side.

Out of humour

I say I say I say, have you heard the one about a new study showing that men are funnier than women? Probably, because every media outlet in the universe has picked it up and rehashed it, from the BBC and Science Daily to Malaysia News and Reddit. Which is no surprise, because it’s a gift to the media, a perfect example of ‘soundbite science’. But if you read this blog regularly you will also have heard the one about why it pays to take a closer look at these sex-difference stories. And guess what, I’m going to tell that one again.

I’ll start by summarising the research itself, as described by one of the researchers, Gil Greengross, in Psychology Today. What he and his colleagues did was a meta-analysis: rather than carrying out their own experiment, they combined the statistical findings of 28 previous studies and calculated the overall size of the effect they were interested in, which was the effect of gender on funniness—or as they called it, ‘humor ability’.

In the studies they selected, one set of experimental subjects had been given some cartoons and asked to come up with amusing captions; another set of subjects then rated the funniness of each caption on a numerical scale. Crucially, the raters didn’t know if the caption they were reading had been written by a man or a woman. If they’d known, their judgments could have been influenced by the stereotype of women as less funny than men. The studies had been designed to investigate whether there was any basis for that stereotype. Would women still be judged less funny by people who had no way of knowing they were women?

The short answer turned out to be ‘yes’. As the Psychology Today piece explains,

In statistical technical terms, the effect size was 0.32, or roughly one-third of the standard deviation. In plain English, this means that 63 percent of men score above the mean humor ability of women. This is considered a small to medium difference.

Evidently aware that such findings often attract criticism, and are frequently misinterpreted, Greengross goes on to assure us that yes, they did look for possible confounding factors (e.g., if there were cultural differences which calculations based on aggregated data were masking, or if there were more male than female raters, which might skew the ratings in men’s favour because men prefer the humour of other men); and no, they weren’t saying that all men are funnier than all women. Some women are very funny, and the most successful female comedians are funnier than 99% of men. Their claim was only that men on average are funnier than women on average.

Greengross then turns to the question of why that might be the case, and tells a story I have discussed on this blog before, in fact more than once, because it features in some accounts of the evolution of human language. It’s about sexual selection, the concept developed by Darwin to explain evolved traits like the peacock’s tail, which can’t have been selected because it made the bird better adapted to its environment (it’s hard to fly away from predators trailing that monster); rather it was favoured because it was attractive to potential mates, ensuring that big-tailed peacocks passed their genes to more offspring.

Peacocks are animals that ‘lek’, display themselves at mating time so that peahens can judge their fitness. The idea is that being funny serves a similar purpose for humans: it’s a form of display put on by men so women can assess how good they’ll be as mates. ‘Humor’, Greengross notes, ‘is strongly correlated with intelligence’. He also reminds us that women advertising on dating sites often mention humour as an important quality in men, whereas men are less interested in whether a woman is funny.  The principle that ‘males do the courting and females do the choosing’ explains why ‘humor ability’ has been selected for in male humans, but not in females, whose reproductive success is not enhanced by it.

This explanation presupposes that ‘humor ability’ has a significant genetic component; but couldn’t being funny be a learned skill which is more common or better developed in men because they get more opportunity or encouragement to practise it? The researchers, it transpires, did consider that possibility, but they dismissed it on the grounds that

There is minimal evidence to support the view that our society suppresses women from producing and exhibiting humor.

Mate, stop it, my sides are going to split—because this has to be a joke, right? You do realise that there’s a long tradition of discourse, both expert and popular, which ‘suppresses women from producing humor’ by telling them that they’re no good at it, it’s unfeminine, and men don’t find it attractive; and you do understand that YOUR OWN STUDY SITS SQUARELY IN THAT TRADITION? Oh, and while we’re on the subject, it presumably hasn’t escaped your notice that saying ‘men have more humor ability than women’ followed by ‘humor is highly correlated with intelligence’ will lead a lot of people to deduce the proposition ‘men are more intelligent than women’—which you wouldn’t be suggesting in all seriousness, would you?

So far I haven’t questioned the central claim made for this meta-analysis: that it confirms the robustness of the effect it investigated (i.e., sex affects funniness and men are funnier than women). Here are a few reasons why I think that claim needs to be at least qualified, and possibly rejected.

First, the studies used in the meta-analysis were chosen because they investigated the same question using a similar experimental design, which elicited evidence of people’s ‘humor ability’ by asking them to write funny captions for cartoon images. Presumably this task was used because it’s something you can get people to do in a lab, and because a written caption does not reveal the writer’s sex (whereas a recording of them telling a joke or a funny story would). But how good a proxy is captioning cartoons for the general ability to be funny?

I would say, not very good. Most everyday humour is produced spontaneously in the course of interaction, typically with people you know, like and have things in common with, and whose reactions guide your performance from moment to moment. The lab cartoon-captioning task reproduces none of these features. It requires subjects to be funny on command (not spontaneously), in isolation (not in a social group) and in writing (not speech), for an absent audience of strangers whose reactions they can only guess at. To me that makes it a dubious basis on which to generalise about ‘humor ability’. At best, the analysis shows that in lab conditions men write funnier cartoon captions than women—not that they’re funnier across the board.

Second, the finding that men are funnier than women is based on averaging the numerical scores given to subjects’ captions by the judges. But what do these numbers correspond to? Clearly, judges’ ratings must reflect their subjective response to a caption, but how, if at all, does that response relate to the caption-writer’s abilities? The problem is that there’s a lot of variation, individual as well as cultural, in what people find funny. If a judge rates one caption as a 2 and another as a 4, is that because of some objective quality of the two captions—the second is simply funnier than the first–or does it just reflect the judge’s preference for certain kinds of jokes over others? In short, can we have any confidence that these numbers are a useful measure of ‘humor ability’?

I’m sceptical, to be honest. I’m not convinced there’s such a thing as ‘humor ability’, and if there is, as I’ve already said, I have trouble seeing the very specific skill of writing funny cartoon captions as a fair test of it. It’s a test most of us would probably fail: back in 2011, in a Psychology Today piece criticising one of these studies, Ben Hayden noted that the commonest score awarded to captions, regardless of the writer’s sex, had been 1, which meant ‘not funny at all’. ‘That indicates’, he comments, that ‘the experimental subjects were not well-matched to the test—they were just swinging wildly’. They were like the proverbial roomful of typing monkeys, only in the field of cartoon-captioning rather than Shakespearean drama.

Third, even if we do trust the numbers, what the meta-analysis reports is only, as Greengross acknowledges, a ‘small to medium difference’. The effect-sizes reported in meta-analytic studies are generally described using a scale that goes from very small/slight to very large, so this effect isn’t even at the mid-point: to put it in context, it’s smaller than the reported effect of sex on spelling accuracy, and much smaller than the reported effect of sex on the distance someone can throw an object. As sex-differences go, it’s not a biggie. (Yet another ‘men are funnier’ study discussed in Psychology Today in 2011 found a difference so slight that even one of the authors described it as ‘just on the edge of detectability’–and he meant, by using statistical methods, not just observation.) Even liberally sprinkled with the words ‘on average’, the statement ‘men are funnier than women’ gives a misleading impression, because what ‘funnier’ means to most people is ‘noticeably funnier’: it implies a difference large enough to be significant in the ordinary rather than the purely statistical sense.

There are, then, a number of reasons to query the presentation of this meta-analysis as  clear confirmation that ‘men are funnier than women’. But my biggest problem with it isn’t the design of the studies analysed, the conclusions drawn by the researchers, or the exaggeration of their significance by the media: it’s the fact that this kind of research is seen as worth doing in the first place.

Think about it: in what other case would anyone design a study to investigate whether there’s a basis for some entrenched negative stereotype of a social group? ‘Is it true that poor people are lazier than rich ones?’ Nope. ‘Are Jews/Scots/Yorkshiremen really meaner than other people with money?’ Nope to the nth power. It would be utterly offensive: no one would fund it and the ethics committee would never approve it. Sex-differences, though, are always potential research fodder. And when the frame, as in this case, is evolutionary psychology, they will always turn out to be explained by the supposed—though in reality often misrepresented–activities of prehistoric hunter-gatherers, and beyond that, by the irreducible difference between the sperm and the egg.

Actually, let me go back a step: it’s not true that all sex-differences are potential research fodder. As Caroline Criado Perez shows in her book Invisible Women, in many areas where women are consequentially different from men, whether it’s the way their bodies respond to drugs or their patterns of travel on public transport, there’s a huge data gap because the differences have not been investigated systematically. Research that would have a positive effect on women’s lives does not get done; yet there’s always room for one more study explaining that women are rubbish at something—maths, map reading, being funny–because biology. (Or alternatively, that women are better than men at something trivial or menial, like nappy-changing or keeping track of people’s birthdays.)

I hope what I’ve just said explains why I think it’s important to challenge claims like ‘men are (on average) funnier than women’. In fact, as I’ve said before, I think the way we talk about who is or isn’t funny matters in its own right: the proposition that ‘Xs have no sense of humour’ has a long history of being used to dehumanise the group in question, and ‘Xs aren’t as funny as Ys’ can also, as noted earlier, be code for ‘they’re not as intelligent’. But even when the difference under discussion really is trivial (for instance, a study once found that female shoppers spend more time browsing than their male counterparts—like their prehistoric ancestors they are natural foragers, whereas men are mighty hunters), it must be seen as part of a larger enterprise, in which the point of cataloguing sex-differences is ultimately to affirm that the sexes are not just different, they are different in ways that make equality a pipe-dream. The meta-message being communicated is ‘women, know your limits’.

Whenever we hear this message we should remember that the whole history of feminism is a story of women refusing to accept the limits imposed by their societies—limits which have always been represented as ‘natural’ rather than man-made. Once, scientists maintained that if women were admitted to universities, the intellectual stress would cause their ovaries to shrivel. Now we recognise that as just self-serving male chauvinist nonsense. The claims made by many scientists today are destined to end up in the same dustbin. Let’s use our intelligence and our ‘humor ability’ to speed them on their way.

Broad men and narrow women: the perils of soundbite science

Last week a few people asked me what I made of a new study that was generating some interest on social media. At the time I hadn’t read it: I only knew Nature had reported it under a headline–‘Male researchers’ “vague” language more likely to win grants’–that made it sound both baffling (why would scientists get points for being vague?) and infuriating (as usual, it seemed to be men who were benefiting and women who were losing out). So I decided to investigate further, and then share my conclusions in this post.

The study was conducted by researchers at the US National Bureau of Economic Research (NBER), and their write-up is available as an NBER Working Paper. The data they analysed consisted of 6794 grant applications submitted to the Bill and Melinda Gates Foundation, which operates a policy of anonymous reviewing. Because reviewers weren’t told whether applicants were men or women, the researchers assumed that any gender differences in success rates could not be the result of direct discrimination. Whatever was leading reviewers to favour men must be contained in the application itself. And since most of a grant application consists of words, they decided to look for gender-differentiated patterns of word-use.

What their analysis revealed was a tendency for reviewers to give higher scores to applications that contained ‘broad’ words and lower scores to those that used ‘narrow’ words. Since broad words were used more frequently in men’s proposals, while narrow words appeared more often in women’s, this preference for broad over narrow words was also a preference for male- over female-authored applications. The researchers found no reason to think that broad words were associated with better proposals. When they looked at what applicants had gone on to achieve, the words used in their proposals appeared to be a poor predictor of research quality. Overall, then, the study’s conclusion was as infuriating as the Nature headline suggested: men whose research was objectively no better than women’s were receiving more funding from the Gates Foundation because reviewers preferred a particular style of grant writing.

One question the researchers didn’t attempt to answer was why men and women writing grant proposals might favour, respectively, ‘broad’ and ‘narrow’ words. But many people who commented on their findings thought the answer was obvious: simply and bluntly put, men–or at least a higher proportion of men–are bullshitters. Whereas women offer specific, realistic accounts of what they think their research can deliver, men have fewer inhibitions about making sweeping, grandiose claims.

This take is an example of a common interpretive strategy. If you present people with a generalization about language and gender—especially one whose significance isn’t immediately obvious—they will often try to make sense of it by invoking some other, more generic gender stereotype. In this case what they did was map the alleged linguistic difference (‘men use broad words, women use narrow words’) onto a higher-level, more familiar male-female opposition: ‘men are over-confident, women are over-cautious’.

You might ask: what’s wrong with that? Stereotypes aren’t always false: there’s plenty of other research you could cite in support of the thesis that men are over-confident (for instance, experimental studies showing that male test-takers consistently overestimate how well they’ve done, or the fact that men are more likely than women to apply for jobs when they don’t meet the advertised criteria). I don’t dispute any of that: in fact, I agree that ‘men are over-confident and women are over-cautious’ captures a real and significant cultural tendency. But there are, nevertheless, some problems with using it to explain the findings of this study.

One general problem is that you can use the same interpretive strategy to explain pretty much any set of findings, including made up ones. Suppose I told you the study had found that men use narrow words and women use broad words (i.e., the opposite of what it actually found). You’d be able to come up with an equally plausible explanation for that (non) finding just by switching to a different gender stereotype. Instead of ‘men use broad words because they’re overconfident bullshitters’ you might suggest that ‘women use broad words because they’re more attuned to their readers’ needs’; or ‘men use narrow words to show off their expert knowledge’. Since the supply of gender stereotypes is inexhaustible, there’s no statement of the form ‘men do x and women do y’ that can’t be slotted into this explanatory frame.

In the case of the NBER study, though, there’s a more specific problem with explaining men’s use of broad words as a linguistic manifestation of their over-confidence. When the researchers use the terms ‘broad’ and ‘narrow’, they don’t mean what people have assumed they mean (i.e., what the words would mean in ordinary English).

By way of illustration, here’s a list of six words taken from the study: three of them were classified as ‘broad’ and the other three as ‘narrow’. Which do you think are which?

  1. bacteria
  2. brain
  3. community
  4. detection
  5. health
  6. therapy

My guess is that you defined words as ‘broad’ if they were just basic, everyday vocabulary, and ‘narrow’ if they were a bit more abstract and technical. On that basis you probably categorised ‘health, ‘brain’ and ‘community’ as broad and ‘bacteria’, ‘detection’ and ‘therapy’ as narrow. That wasn’t, however, what the researchers did. Their definition wasn’t based on the characteristics of the words themselves, but on their frequency and distribution in the sample. Broad words were those that occurred in proposals on a wide range of different research topics; narrow words were restricted to proposals on a particular topic. By those criteria, ‘bacteria’, ‘detection’ and ‘therapy’ were broad, whereas ‘brain, community’ and ‘health’ were narrow.

If you think these definitions are confusing, I agree: the researchers might have done better to choose a different pair of terms (like, say, ‘core words’ and ‘peripheral words’). But once you’ve understood how they made their broad/narrow distinction and looked at the words in each category, it becomes difficult to argue that what’s behind the gender difference is men’s propensity for writing grandiose bullshit and women’s dogged attention to detail. (Is ‘health’ more precise than ‘bacteria’? Is ‘therapy’ vaguer than ‘brain’, or more grandiose than ‘community’?)

The fact that so much discussion revolved around the question of explanation suggests that most people had simply accepted the findings themselves at face value. This always bothers me: in my view, any claim that men use language in one way and women use it in another should be approached with a degree of scepticism. And that’s especially true if what you’re basing your assessment on is a report in the media. For obvious reasons, the media pay most attention to studies whose findings will make an eye-catching headline or a killer soundbite; this means they have a bias towards research which makes bold rather than cautious claims (stories like ‘men and women fairly similar, study shows’, or ‘we looked, but we didn’t find anything’, are not exactly clickbait). But for feminist sceptics it’s always worth asking whether the finding everyone’s talking about is supported by any other evidence. Have other researchers found the same thing? Or have they asked similar questions and come up with different answers?

There is, in fact, other research investigating the influence of writing style on grant decisions. Earlier this year, the Journal of Language and Social Psychology published an analysis of the language used in a sample of nearly 20,000 abstracts taken from research proposals submitted to the US National Science Foundation. This study considered only successful applications, taking the amount of funding applicants had been awarded as a measure of how positively their proposals had been assessed. It found there was a relationship between the funding researchers received and the language used in their proposal abstracts, but the linguistic features which made a difference were not the same ones the NBER study identified. The NSF gave more money to applicants whose abstracts were longer than average, contained fewer common words, and were written with ‘more verbal certainty’.

But I’m not just lamenting the uncritical reception of the NBER findings on general scientific principles. It also bothers me because I know how easy it is to propagate myths about the way men and women use language. ‘Men use broad words and women use narrow words’ is exactly the sort of thing that gets mythologized–detached from its original context (a study in which, as I’ve already pointed out, it meant something completely different from what most people thought) and repeated without elaboration in dozens of other sources, until eventually it turns into one of those zombie facts–like ‘Eskimos have a lot of words for snow’, or ‘women utter three times as many words per day as men’–that refuse to die no matter how many times they’re debunked.

If it does become part of our collective folk-wisdom on this subject, there’s every chance that ‘men use broad words, women use narrow words’ will also be filtered through the kind of deficit thinking which sees whatever women do with language as a problem in need of remedial intervention. Using ‘narrow’ words could join over-apologizing, hedging and tilting your head on the list of bad habits which are said to hold women back, and which it then becomes women’s responsibility to fix. (I can already imagine the TED talks exhorting women to ‘think broad’, and the workshops for female grant applicants on ‘choosing the right words’.)

To be fair to the authors of the NBER study, that isn’t what they think should happen. As they see it, it’s the reviewers who need training: their bias towards certain ways of writing elevates style over substance and leads to less than optimal funding decisions. But it’s hard for researchers to control what people make of, or what they do with,  findings that have entered the public domain. Even a study that was intended to be part of the solution can end up becoming part of the problem.

This is a dilemma for everyone who researches or writes about language and gender, myself included. Whenever I criticise some questionable claim or mistaken belief, I’m aware that I could be amplifying it just by giving it airtime. Though I’m only repeating it to explain the arguments against it, those arguments won’t necessarily be what people take away. But as you’ll have noticed, that hasn’t caused me to retreat into silence. I do believe that knowledge can set us free–but only if we’re willing to interrogate it critically.

 

Joking aside

Do you use humour at work? Have you ever cracked a joke to liven up a boring meeting, or kicked off a presentation with an amusing anecdote? Would you agree that the ability to make people laugh is a useful professional skill?

If your answer to these questions is ‘yes’—and if you also happen to be a woman—then I’m afraid I’ve got some bad news for you. According to a recently-published article in the Journal of Applied Psychology, using humour at work enhances men’s status, but for women it has the opposite effect. Whereas men’s humour is seen as ‘functional’, a tool for producing all kinds of positive outcomes (defusing tension, reframing problems, bonding team-members), women’s is more likely to be seen as ‘disruptive’, a sign that they’re lightweights who lack focus and dedication.

How, I hear you ask, did the authors of the article reach that conclusion? The answer is that they conducted an experiment: they recruited a sample of judges and asked them to evaluate a presentation made by a store manager named Sam (in fact the presenter was an actor and the presentation was scripted). Some judges watched a male Sam, others a female one; in each case half of them saw a presentation in which no use was made of humour, while the other half saw a version of the same presentation that included five humorous statements. Their article doesn’t specify what these were, but in a write-up for the Harvard Business Review they do reproduce the first one:

So, last night, my husband/wife gave me some good advice about this presentation. He/she said, whatever you do don’t try to be too charming, witty, or intellectual… just be yourself!

It’s not exactly side-splitting stuff, but subjects did judge both versions of Sam funnier when their presentations included it. However, those who had watched the female Sam rather than the male one were more likely to agree with statements like ‘the humor distracted from the purpose of the presentation’. And when they were asked about Sam’s career prospects (‘in your opinion, how likely is it that Sam will advance in the organization?’), the judges gave higher scores to the funny male Sam than either the non-funny male Sam or the funny female Sam. Female Sam did better on these questions when she was not funny (though she still did less well than her unfunny male counterpart). When she was funny, the judges accused her of, as one put it, trying ‘to cover up her lack of real business acumen by making little jokes’.

The authors explain their findings as the product of gender bias: their study shows, for the n millionth time, that even if the behaviour of men and women is identical, it is liable to be interpreted in different ways and judged by different standards. He is ‘direct’ and she is ‘abrasive’; he uses humour to get things done and she uses it to ‘cover up her lack of real business acumen’. That’s why, as I have pointed out in other posts about language in the workplace, advising women to imitate men’s behaviour is unlikely to solve their problems. These researchers agree, warning that

The potential for women to advance in the workplace may be harmed by the use of humor. Thus, recommending the use of humor to women leaders may in fact reduce their perceived effectiveness and opportunities for career advancement.

But this is not very helpful either, because avoiding humour also has costs. The humourless, po-faced boss or co-worker is not, generally speaking, a popular figure; if she’s female, her refusal to lighten up is likely to prompt the judgment that she is arrogant, or—that cardinal female sin—’unapproachable’. It seems women are damned whatever they do: if they’re funny they’re seen as disruptive, but if they aren’t they’re seen as unlikeable.

The authors say they’re not suggesting women should stop being funny at work, they’re just drawing attention to the problem in the hope  that ‘increased awareness of prejudice can help to reduce its occurrence’. I can’t say I share their optimism: many people have raised doubts about the effectiveness of interventions based on this principle, like unconscious bias training.  As with all discussions which start by asking how women’s behaviour might be holding them back at work, I think the main effect of ‘increased awareness’ will probably be to make women even more anxious and self-conscious than they are already. It’s predictable, depressing and infuriating—but before we throw up our hands in despair and look for new careers as self-employed spoon-whittlers, we should pause to ask if this study tells the whole story about gender and humour.

As the authors themselves acknowledge, their methodology had some obvious limitations. If you ask subjects to judge a scripted presentation delivered by a person they have never seen before, you are maximizing the probability that their judgments will rely on stereotypes: what else, after all, have they got to work with? In real life we usually have information about people that goes beyond obvious characteristics like sex, race and age. Also, in our real working lives our judgments aren’t abstract and decontextualized: rather we assess behaviour in relation to the whole situation—one which we are not just observing at a distance, but are actively involved in ourselves. The question arises, then, of whether the reactions of the judges in the experiment tell us anything very useful about real workplace situations.

As it happens, the use of humour was one of the issues examined in a large qualitative study of gender and workplace talk that Janet Holmes and her colleagues carried out in New Zealand. This study found that although the amount and type of humour people used varied in different workplaces, humour itself was a ubiquitous feature of working life, and its uses were similar for employees of both sexes. In Holmes’s words, ‘Both women and men crack jokes, exchange jocular abuse and tell funny stories at work’. Her account did not suggest that engaging in these behaviours reduced women’s perceived effectiveness. In fact, it suggested that women could use humour as a means of asserting or maintaining their status.

One function of humour is to soften criticism (and other acts that might cause hurt or offence) and reduce the risk of provoking conflict. Making a joke of something renders it both less overtly threatening and more difficult to take issue with (since if you object you risk coming across as humourless). This is what makes humour such a useful resource for sexists: when women protest about jokes or comments they find offensive, they can be met with the time-honoured ‘just banter’ defence (‘we weren’t being serious–can’t you feminists/PC-types take a joke?’) In the New Zealand data, however, there were cases where women used humour as a resource for either contesting sexism or turning the tables on men. For instance, at one project team meeting a woman initiated a humorous exchange that traded on a well-known stereotype of male incompetence:

Clara: he wants to get through month’s end first. He’s –  he can’t multi-task
[Other women laughing]
Peg: It’s a bloke thing
[General laughter]
Clara: [laughs] yeah yeah

The ‘softening’ effect of humour can also make a woman’s authority more palatable. Clara is noted for her direct, decisive and not especially collaborative management style; but one way in which she maintains good relationships with colleagues is by taking it in good part when they jokingly refer to her as ‘Queen Clara’. This nickname, which likens her to a monarch issuing commands to her subjects, is itself evidence of the way women are judged by a sexist double standard. I did once know a man whose workplace nickname was ‘King X’, but he wasn’t just direct and decisive, he was a tyrannical megalomaniac whose subordinates lived in fear of him. But Clara’s willingness to go along with the joke serves a pragmatic purpose: she gets what she wants from her team, while also deflecting the criticism to which all powerful women are vulnerable, that she’s an overbearing stuck-up b****.

The New Zealand study presents evidence that workplace humour is a complex phenomenon which serves a range of different purposes, and that in real-life work situations gender is only one of many factors that shape its use and interpretation. Other contextual variables, such as the culture of the organization, the roles of individuals and their relationships with colleagues, are more significant influences than gender in and of itself. By stripping out all that other stuff, the experimental study almost certainly amplified the gender difference it was investigating, potentially leading women to overestimate the risk that using humour in the workplace would harm their careers.

Methodological limitations aside, studies like this one also prompt the more basic question of why a certain issue is being investigated in the first place. The researchers didn’t pluck their hypothesis from thin air: there’s a long tradition of scientific (or ‘scientific’) discourse on gender and humour, and its starting point has always been that there’s something anomalous about women being funny.

When feminists took up the subject in the 1970s, one of their goals was to challenge the sexism of previous accounts, both scientific and popular, which essentially argued that being funny was a guy thing and women were just no good at it. They were either seen as innately humourless (an accusation commonly levelled at feminists, and even more frequently at lesbians), or else as too dim and ditzy to do humour well. If they tried to tell a joke they’d get confused and forget the punchline; if they embarked on a funny story they’d keep going off at tangents until their listeners lost interest. This thesis came in various theoretical flavours: Freud was popular in some quarters, Darwin in others (the Darwinian argument—that men use humour to attract mates, whereas women don’t need to be funny, they just need to be physically attractive—survives to this day).

One possible response to this argument was to call it out as sexist bullshit. Another, however, which was popular among some feminists, was to say that men didn’t find women funny because they defined ‘being funny’ in a way that excluded women’s distinctively female style of humour.

Descriptions of this style will remind anyone who knows the work of Deborah Tannen of her ‘difference’ or ‘two cultures’ approach, which posits a fundamental contrast between status-oriented and competitive men and rapport-building, collaborative women. Well before Tannen popularized it, this contrast had been invoked to make generalizations about the kinds of humour that were typically favoured by women or men. For instance: whereas men compete to top each other’s contributions, women collaborate to produce intimacy through shared laughter. Whereas men like jokes that climax with a punchline, women prefer less structured personal anecdotes. (If this one reminds you of another much-discussed sex-difference, I can only say you’re not alone.) And whereas men tend to make others the butt of their humour, women are more likely to poke fun at themselves.

A number of linguists I respect have used this framework, and I don’t dispute their observations about the way humour was used by the women and men they studied. But it’s a mistake to generalize about half of humanity from such a limited body of evidence–one that’s heavily skewed towards a particular subset of women, often talking in contexts where you’d expect to see collaborative, rapport-building behaviour (e.g. long-established female friendship groups, feminist ‘rap groups’, and support groups for mothers with young children). Even in the 1970s there were cases that didn’t fit this template—such as Rayna Green’s 1977 study of women’s bawdy talk in the US South, which included one woman’s riposte to a comment from her granddaughter on the sparseness of her pubic hair: ‘grass don’t grow on a racetrack’.* Some later research noted that both women and men used different kinds of humour in single-sex and mixed-sex groups. The New Zealand workplace study documented both ‘feminine’ and ‘masculine’ styles of humour, but it didn’t find that either style was used by one sex exclusively.

The more examples of humour we look at from different communities and settings, the more difficult it becomes to argue that there are clear-cut gender differences. As usual, there’s a gap between the actual behaviour of men and women, which shows more overall similarity than difference (along with a lot of variation inside each of the two gender groups), and our cultural beliefs about their behaviour, which are much more consistent—largely because they’re not derived from observations of what men and women do, they are expressions of our deeply-held convictions about what men and women are or should be like.

The authors of the study I began with suggest that what’s behind the prejudice against women being funny at work is our belief that men are more agentive, rational and goal-oriented than women. That’s why men’s workplace humour is interpreted as functional, deployed by rational agents as a way of achieving their goals, while women’s is seen as disruptive, signalling a lack of dedication to the business at hand. But this doesn’t really account for the fact that the prejudice isn’t confined to situations like the workplace where humour can be seen as ‘functional’ or ‘goal-oriented’. I can’t help thinking it skirts around some much more general points about humour, gender, sex and power.

Being funny is, in a number of key respects, incompatible with conventional femininity. For one thing, it involves putting yourself centre-stage: when you embark on a joke or a funny story you’re saying ‘pay attention to me’, and when you finish you’re expecting some sort of acknowledgment, like laughter or applause. That kind of attention (and the feeling of power you get from it) is still widely seen as a male prerogative: women who usurp it are not only displaying a lack of feminine modesty, they are also failing to play their prescribed role as supporters and cheerleaders for men. (Some studies have reported that women laugh more at men’s jokes than vice versa; and anecdotally it’s been suggested that when men advertise for a female partner with a good sense of humour, what they’re looking for isn’t a funny woman, but a woman who will tell them they’re funny.)

For another thing, it’s fairly difficult to make people laugh while also projecting the kinds of feminine qualities our culture defines as sexually alluring—like elegance and glamour, or innocence and grace. Funny women and sexy women are frequently presented as different ‘types’. That’s why so many films and TV shows pair a sexually attractive female protagonist with a less attractive best friend/sister/roommate: the sexy woman gets the guy, while the plain, fat or dowdy one gets the laughs. Behind this division of labour is the old idea that humour unsexes or de-feminizes women, and that those who make a speciality of it are trying to compensate for being ugly and unattractive.

Nevertheless, women persist in being funny—and so they should, whatever studies show. What studies mostly show is that women can be criticized however they behave, particularly in the workplace. And if the critics are never going to like what you do, you might as well just do what you like.

 

*I haven’t linked to Green’s study, ‘Magnolias grow in dirt’, because the source isn’t available if you’re searching from a location in the EU–but it’s discussed in this generally useful review of 20th century gender and humour research. 

Lekkers and losers

It will not be news to readers of this blog that I take a keen interest in popular literature on the subject of gender and communication. In my house there’s a whole shelf of old books I can’t keep in my office because they’re too embarrassing: they include Men are from Mars, Women are from Venus and several examples of vintage neurobollocks. Fortunately, the internet now allows me to sample this kind of material more discreetly (and usually for free). Really, I’ve no excuse not to stay current.

Yet until two days ago I’d never heard of Dan Bacon, a self-described ‘dating and relationship expert’ and the proprietor of a website called The Modern Man (‘if you have a problem with women, I have the perfect solution for you’). Dan has recently become infamous for writing a piece entitled ‘How to talk to a woman who is wearing headphones’. Or, put another way, ‘how to make a woman take her headphones off and pay attention while you try to pick her up’.

From what I could see on Twitter, the general consensus was that the premise wasn’t just sexist, it was creepy and borderline rapey. I don’t disagree. According to Fiona Vera-Grey, who has done research on women’s experiences of being harassed by men in public places, wearing headphones is one of several tactics women use specifically to protect themselves from unwanted male attention. Like looking down at your phone to avoid eye contact and sitting near the door on the tube, it’s a form of everyday ‘safety work’.

But to men like Dan it is a truism that a woman’s apparent lack of interest in you is not to be taken at face value. It’s a challenge. Reading his piece reminded me of when I was in my early 20s (before anyone had a mobile phone or routinely wore headphones when out and about), backpacking around Spain with my then-girlfriend. After days of being hassled by men, we went into a bookshop and bought large badges adorned with lesbian symbols and the sentence (in Spanish) ‘we want to walk in peace’. Which we soon stopped wearing, because they proved to be the opposite of a deterrent. The more clearly we broadcast the message WE’RE NOT INTERESTED IN YOU, the more of a challenge we were, and the more persistent the men became.

In those days we had no idea how men rationalised this kind of behaviour to themselves. Today we have the benefit of websites like Dan’s, where practical tips on picking up women are often padded out with more philosophical reflections on masculinity and gender difference.

Dan’s views are perhaps most clearly summarised in his concluding paragraph.

As you may have noticed, women don’t usually go around approaching men. Women know that it’s the man’s role to be confident enough to walk over and talk to women he finds attractive. Women, on the other hand, have to look their best and try to attract the attention of the confident alpha males who approach.

From this we learn that in Dan’s imagination, nothing is what it seems. Deliberately ignoring a woman’s ‘leave me alone’ signals isn’t proof that you’re a jerk, it’s proof that you’re an ‘alpha male’. The woman’s headphones aren’t saying ‘I’m not available to talk’, they’re a test she’s set for the men who cross her path, and the losers who take her literally will fail. The man she’s actually attracted to is the one who disregards what she appears to want–who knows that her resistance is token, only there to be overcome. In this Mills & Boon theory of male-female relationships, no woman wants her man to be a wimp, or indeed an equal: he needs to make his dominance felt.

In an earlier post about lists of ‘things not to say’ to the opposite sex, I pointed out that women are told not to say things that men might perceive as criticisms or demands, whereas men are told not to say things that women might perceive as concessions or indications of weakness.  Dan Bacon observes this convention: his piece includes a list of mistakes men make which includes ‘allowing her to take control of the interaction’.

No matter how confident or challenging a woman might behave, she still dreams of meeting a guy who is more confident than her. A woman doesn’t want to be forced to control an interaction with a guy.

He goes on to point out that just getting the woman to take her headphones off is only the first step: now you’ve got her attention, you need to deliver on the promise of having something to say.

Engaging conversation skills are essential in keeping a woman’s attention at the best of times and even more so when she can switch herself off with a click of the “play” button.

Obvious? Banal? Well, yes, but it’s also an illustration of something that’s puzzled me for quite a while. Advice on communicating with the opposite sex presents men in two distinct and apparently contradictory ways. When the subject is dating, men—or at least, the alpha males the reader is encouraged to emulate—are depicted as articulate, smooth-talking charmers who ‘take control’ of interactions and use their ‘engaging conversation skills’ to keep women hanging on their every word.  But when the subject is marital relationships, men are most often presented as verbally-challenged idiots who can barely string a sentence together. They are said to prefer action to words, to be incapable of expressing their feelings, and to have great difficulty understanding what women say to them.

What unites these two accounts at a more abstract level is that each contains an inbuilt justification for problematic male behaviour. In the first case it’s ‘women want men to dominate them’, and in the second it’s ‘men care about women really, they’re just Martians with poor communication skills’. But it still seems odd that they can exist side by side. Maybe it’s a case of telling the target audience what it wants to hear: the smooth-talking charmer appears mainly in dating advice addressed to men, whereas the idiot appears in advice for heterosexual couples which is probably read more often by women.

The same two characters also turn up in scientific discussions of the evolution of language. Here there’s more explicit awareness of the contradiction: it’s a topic of debate among evolutionary scientists with an interest in how and why the human language faculty evolved. The question that’s relevant here is the ‘why’ one: our linguistic abilities are costly (because they require such a big brain), and other primates have done fine without them, so what survival advantage did they confer on us?  One early theory about this was that language enabled humans to co-ordinate joint activities that contributed to survival, like hunting. But more recently attention has turned to two other stories.

I’ll call the first one the ‘social networking’ theory. It says that the essential advantage conferred on humans by their ability to speak was to do with forging social bonds and transmitting social knowledge: talking helps individuals keep track of what’s going on in the group and maintain social relationships with other members. This produces a more cohesive (and therefore more successful) group. One prominent advocate of this theory, Robin Dunbar, argues that gossip—by which he means everyday talk about what’s happening and who’s doing what with whom—has a similar function in human groups to physical grooming among non-human primates. Dunbar thinks early human females would have been central in the development of language. Females play a key role in the organisation of primate groups, and females caring for infants would have had a particular interest in building networks of mutual co-operation.

Supporters of this theory often claim that it’s in line with contemporary observations about sex-differences in linguistic behaviour. Everyone agrees that women are the ‘more verbal’ sex, more talkative and more verbally skilled than the action-oriented male. What’s happening here is that the verbally-challenged idiot who appears in books like Men are from Mars, Women are from Venus (and the unspeakable neurobollocks classic Why Men Don’t Listen and Women Can’t Read Maps) is being projected back into pre-history.

But as you’ll know if you read another of my early posts, it isn’t actually true that women talk more than men: most evidence points in the opposite direction. And while women do, on average, do slightly better on certain tests of verbal ability, the abilities being tested are not things that would have helped our preliterate ancestors as they foraged on the African savannah. These points have been seized on by supporters of the second story about what language did for humans, which I’ll call the ‘lekking theory’.

A ‘lek’ is what students of animal behaviour call a type of courtship ritual seen among, for instance, peacocks, where males display themselves in groups to an audience of females, and the females make judgments on their reproductive fitness. The peacock’s tail is the example famously given by Darwin to illustrate the concept of ‘sexual selection’, where a trait that may confer no practical advantage is selected because it makes its carrier more attractive as a mate. Some evolutionary scientists suggest that the human capacity for language was selected in the same way and for much the same reason: speaking offered an excellent way for early human males to show off to females, and for females to judge the males’ fitness, given that verbal behaviour is a clue to both intellectual abilities and social skills.

Supporters of this theory, like the behavioural scientist John Locke, also claim that it receives support from contemporary observations about sex-differences in linguistic behaviour. Women may have the edge in lab tests of verbal skill, but who dominates—in virtually all the cultures where the question has been studied—the oral performance genres that display most skill and creativity? Who are the most accomplished orators, debaters, poets, rappers, preachers, stand-up comedians? Who fights verbal duels to assert their superiority over their rivals?

In this account it’s the smooth-tongued charmer who gets projected back into prehistory—conveniently ignoring the more recent evidence (by which I mean evidence from the past thousand years or so) that women’s non-participation in various kinds of oral performance reflected neither lack of skill nor lack of interest, but had more to do with men’s strenuous efforts to exclude them. (Which have not entirely ceased, but they’re becoming steadily less effective.)

I’ve got nothing good to say about either of these competing accounts of gendered linguistic behaviour: both are thoroughly sexist, as well as being unconvincing for other reasons. They also illustrate that sex-difference science and folk-wisdom aren’t always as far apart as we might think. Sometimes one is just a scienced-up version of the other. It’s possible Dan Bacon got his ideas from a book about evolutionary psychology, but it’s more likely he’s just channelling the wisdom of the ages. Either way, the effect is to reinforce the belief that women have no right to withhold their favours from men. Sorry, Dan, but the headphones stay on. If you don’t like it, lek off.

Misogyny by numbers

Last week saw the launch of Reclaim the Internet, a campaign against online misogyny. Both the campaign and the (copious) media reports of it leaned heavily on research conducted by the think-tank Demos,  which investigated the use of the words ‘slut’ and ‘whore’ in tweets sent from UK-based accounts over a period of about three weeks earlier this year. The study identified 10,000 ‘explicitly aggressive and misogynistic tweets’ containing the target words, sent to 6500 different Twitter-users. It also found that only half these tweets were sent by men—or, as numerous media sources put it, that ‘half of all online abusers were women’.

So frequently and insistently was this statistic repeated, the message of the day almost became, ‘look, women are just as bad as men!’ Women like the journalist and feminist campaigner Caroline Criado-Perez, who were sought out for comment because of their experience of online abuse, got drawn into lengthy discussions about the misogyny of other women.

Of course, it isn’t news that some women call other women ‘sluts’ and ‘whores’ (or that women may be involved in the most serious forms of online abuse: one of the people prosecuted for sending death-threats to Criado-Perez was a woman). But ‘who sends abusive messages?’ is only one of the questions that need to be addressed in a discussion of online abuse. It’s also important to ask who the messages are typically addressed to and what effect they have, not just on their immediate recipient but on other members of the group that’s being targeted. But those questions weren’t addressed in this particular piece of research, and it was difficult to raise them when all the interviewers wanted to talk about was that ‘half of all abusers are women’ statistic.

These discussions reminded me of the way anti-feminists derail discussions of domestic violence with statistics supposedly showing that women are as likely to assault men as vice-versa. Feminists have challenged this claim by looking at the finer details of the data the figures are based on. They’ve pointed out, for instance, that female perpetrators are most commonly implicated in single incidents, whereas men are more likely to commit repeated assaults, and to do so as part of a larger pattern of coercive control. It’s also men who are overwhelmingly responsible for the most serious physical assaults, and for the great majority of so-called ‘intimate partner killings’.

Once you focus on the detail, it’s clear domestic violence isn’t an equal opportunity activity. Online misogyny probably isn’t either (especially if you focus on the kind that really does deserve to be called ‘abuse’—stalking, repeated threats to rape and kill, etc). But the Demos study didn’t capture any of the detail that would allow us to see what’s behind the numbers.

In this it is fairly typical of the kind of research which funders, policymakers and the media increasingly treat as the ‘gold standard’, involving hi-tech statistical analysis of very large amounts of information—what is often referred to as ‘big data’, though that term has come to be used rather loosely. Strictly speaking, the ‘big data’ label wouldn’t apply to the Demos study, whose sample of 1.5 million tweets is very small beer by big data standards. At the same time, it’s too much data to be analysed in detail by humans: the researchers employed NLP (natural language processing), using algorithms to make sense of text, and their findings are essentially statistical—figures for the frequency of certain kinds of messages, along with the gender distribution of their senders.

You may be thinking: but doesn’t it make sense to assume that ‘bigger is better’—that the more data you crunch through, the more reliable and useful your results will be? I would say, it depends. I’m certainly not against quantitative analysis or large samples: if the aim of a study is to provide information about the overall prevalence of something (e.g., online misogyny on Twitter), then I agree it makes sense to go large. Actually, you could argue that Demos didn’t go large enough: not only was their sample restricted to tweets which contained the words ‘slut’ and ‘whore’, the time-period sampled was short enough to raise suspicions that the findings were disproportionately affected by a single event (the surprisingly high number of woman-on-woman ‘slut/whore’ tweets may reflect the massive volume of abuse directed at Azealia Banks by fans of Zayn Malik after she attacked him publicly).

What I am against, though, is the idea that the combination of huge samples and quantitative methods must always produce better (more objective, more reliable, more revealing) results than any other kind of analysis. Different methods are good for different things, and all of them have limitations.

The forensic corpus linguist Claire Hardaker knows a lot about what can and can’t be done with the tools currently available to researchers, and she has explained on her blog why she’s sceptical about the Demos study. Her very detailed comments confirm something a lot of people immediately suspected when they first encountered the claim about men and women producing equal numbers of abusive tweets. That claim presupposes a degree of certainty about the offline gender of Twitter-users which is not, in reality, achievable. (This isn’t just because people disguise their identities online, though obviously a proportion of them do; Hardaker explains why it’s a problem even when they don’t.)

Another thing Hardaker is sceptical about is the researchers’ claim to have trained a classifier (a machine learning tool that sorts things into categories) to distinguish between different uses of ‘slut’ and ‘whore’, so that genuine expressions of misogyny wouldn’t get mixed up with ironic self-descriptions or mock insults directed at friends. Her observations on that point deserve to be quoted at some length:

We can guarantee that the classifier will be unable to take into account important factors like the relationships between the people using [the] words, their intentions, sarcasm, mock rudeness, in-jokes, and so on. A computer doesn’t know that being tweeted with “I’m going to kill you!” is one thing when it comes from an anonymous stranger, and quite another when it comes from the sibling who has just realised that you ate their last Rolo. Grasping these distinctions requires humans and their clever, fickle, complicated brains.

When you depend on machines to make sense of linguistic data, you have to focus on things a machine can detect without the assistance of a complicated human brain. A computer can’t intuit whether the sender of a message harbours particular attitudes or feelings or intentions; what it can do, though, is identify (faster and often more accurately than a human) every instance of a specific word. So, what happens in quite a lot of studies is that the researchers designate selected words as proxies for the attitudes, feelings or intentions they’re interested in. In the Demos study, these proxy words were ‘slut’ and ‘whore’, and the presence of either in a tweet was treated as a potential indicator of misogyny.

One obvious problem with this is that it excludes any expression of misogyny that doesn’t happen to contain those particular words. The researchers themselves were well aware that tweets containing ‘slut’ and ‘whore’ would only make up a fraction of all misogynist tweets (one of them told the New Statesman they were ‘only scratching the surface’).  But that point got completely lost once their research became a media story. The media need short-cuts: they’ve got no time for the endless qualifications that litter academics’ prose. Consequently, the figures given in the report for the frequency of ‘slut’ and ‘whore’ soon began to be presented as if they were a definitive measure of the prevalence of online misogyny in general.

The researchers were also aware that in context, ‘slut’ and ‘whore’ aren’t always expressions of misogyny. They may be being used in an ironic or humorous way; they may turn up in feminist complaints about ‘slut-shaming’ or ‘whorephobia’. So after searching for every instance of each word, the researchers used a classifier to filter out irrelevant examples and sort the rest into various categories.

Since the full write-up of the 2016 Demos study doesn’t seem to be available yet, I’ll illustrate how this works in practice using the report of a study which the same research group carried out in 2014, apparently using much the same methodology. In this earlier research they investigated three words, ‘slut’, ‘whore’ and ‘rape’. When they analysed the ‘rape’ tweets, they started by getting rid of irrelevant references to, for instance, ‘rapeseed oil’. Then they used a classifier to distinguish among tweets which were discussing an actual rape case or a media report about rape (these made up 40% of the total), tweets which were jokes or casual references to rape (29%), tweets which were abusive and/or threats (12%), and tweets which didn’t fit any of those categories and so were classified as ‘other’ (it’s possibly not a great sign that 27% of the sample ended up in the ‘other’ category).

This looks like the kind of classification task that computers aren’t very good at for the reasons explained by Claire Hardaker (distinguishing abuse from humour, for instance, calls for human-like judgments of tone). But the limitations of current technology may not be the only problem. As a check on the classifier’s reliability, a few hundred tweets from the sample were classified by human analysts. Some of these manually-categorized examples are reproduced in the report to illustrate the different categories. To me, what these examples show is that once messages have been extracted from their context, there’s often enough ambiguity about their meaning to cause problems for a human, let alone a machine.

Here’s a straightforward case—a tweet the humans categorised as a joke.

@^^^^ that was my famous rape face 😉 LOL Joke.

This is unproblematic because the tweeter has taken a lot of trouble to make its status as a joke explicit (adding a winky face, LOL, and then the actual word ‘joke’). But how about this tweet, which comes from the 12% of rape references which the humans categorised as ‘abusive/ threats’?

@^^^^ can I rape you please, you’ll like it

If you think that’s a repugnant thing to tweet at someone, you’ll get no argument from me. But I don’t think it’s self-evident that this strangely polite request is intended as a serious threat rather than a ‘mock’ or ‘joke’ threat. The original recipient will have decided how to take it by using contextual information (e.g., whether the tweeter was a friend or a random stranger, what if anything the tweet was responding to, whether there was any history of similar messages, etc.) Without any of that context, the significance of a message like this one for its original sender and recipient is something an analyst can only guess at.

The example I’ve used here is a ‘threat’ that might conceivably have been intended and taken as a ‘joke’, but it’s likely there are also cases of the opposite, tweets the researchers classified as ‘jokes’ which were intended or taken as serious threats. So I’m not suggesting that the proportion of actual rape threats was lower than the reported 12%; I’m suggesting that the classification—even when done by humans—is not sufficiently reliable to base that kind of claim on. And that the main reason for this unreliability is the way large-scale quantitative studies of human communication detach individual communicative acts from the context which is needed to interpret their meaning fully.

Whether we’re academics, journalists or campaigners, we all like to fling numbers around. There’s nothing like a good statistic to draw attention to the scale of a problem, and so bolster the argument that something needs to be done about it. And I’m not denying that we need the kind of (large-scale and quantitative) research which gives us that statistical ammunition. But two caveats are in order.

First, large scale quantitative research is not the only kind we need. We also need research that illuminates the finer details of something like online misogyny by examining it on a smaller scale, but holistically, with full attention to the contextual details. There’s a lot we could learn about how online abuse works—and what strategies of resistance to it work—by using a microscope rather than a telescope.

Second, if we’re going to rely on numbers, those numbers need to be credible. In that connection, the Demos study hasn’t done us any favours: I’ve yet to come across any informed commentator who isn’t at least somewhat sceptical about its findings. While some of the problems people have commented on reflect the way the media reported the research—pouncing on the ‘women send half the tweets containing “slut” and “whore”’ claim and then reformulating that as ‘women are half of all online abusers’ (an assertion whose implications go well beyond what the evidence actually shows)—there are also problems with the researchers’ own claims.

‘My issue’, says Claire Hardaker, ‘is that serious research requires serious rigour’.  When research is done on something that’s a matter of concern to feminists, its quality and credibility should be an issue for us too.