Bullshit: the struggle goes on

When it comes to the way she speaks, a woman’s place is in the wrong. It’s a point I’ve made frequently on this blog, and last week brought a reminder of how true it continues to be. On the same day I published a post inspired by criticisms of Greta Thunberg’s ‘strident’ speech to the UN Climate Action summit (‘strident’ being a code-word for women who express their views in an ‘excessively and unpleasantly forceful way’), the Times Education Supplement published a piece complaining that women don’t speak forcefully enough. It started like this:

“I’m sorry, I’m no expert on this but could we possibly…”

Have you ever heard yourself say something like this to your team?

I certainly have and I cringe to think how I must have come across.

Such phrases can make women appear weak or ineffective to colleagues, which in turn may affect whether they are promoted or tapped on the shoulder for forthcoming opportunities. These phrases are also in stark contrast to the way some men big themselves up.

And such linguistic differences have a tangible impact: researchers have found that subjective views of men and women vary enormously, with women far more likely to be described as indecisive, inept and panicky.

This piece is an excellent example of the generic formula I laid out in a 2015 post called ‘How to write a bullshit article about women’s language‘. When I first saw it, I thought: ‘you’ve already dealt with that, let it go’. But then I thought: if the bullshit-merchants have no shame about repeating themselves, their critics shouldn’t either. It’s a bit like the battle to contain infectious diseases: even if you’ve been immunized in the past, you may still need an occasional booster shot. So, here’s a reminder of what makes this article, and others like it, such pernicious nonsense.

Problem 1: do women really talk like that?

‘Have you ever heard yourself say something like this to your team?’ is what my O level Latin teacher used to call ‘a question expecting the answer yes’. But is ‘yes’ the true answer? Do women-in-general really spend their working lives saying things like ‘sorry, I’m no expert, but could we possibly…’?

I ask because the example is clearly invented. It hasn’t been taken from a real-life workplace interaction, it’s been constructed by someone who’s trying to shoehorn the Big Three female deadly sins (apologising, self-deprecation, hedging) into a single utterance. It isn’t, in short, evidence of anything, except our willingness to believe a certain story about the way women habitually talk.

If you’re thinking the evidence will be presented later on, I’m sorry (sic) to have to disappoint you: all we get is an anecdote about a nameless woman-in-a-meeting who allegedly uttered the word ‘sorry’ 32 times in an hour. Even if this is a true story, a single individual is not a representative sample from which to draw general conclusions about half of humanity. And if you feel you recognise yourself or your co-workers in anecdotes like this, remember that your personal intuitions aren’t reliable evidence either—not least because they’ve probably been influenced by reading dozens of pieces just like this one.

Problem 2: do the ways of speaking the writer criticises mean what she claims? 

The people who write these articles take the meaning of the various expressions they castigate women for using to be both obvious and invariant. ‘Sorry’ means you’re putting yourself in the wrong; hedges like ‘could we possibly’ and ‘just’ mean you have no confidence in your own opinion; ‘I’m no expert’ means you’re disclaiming any authority to pronounce on the issue at hand. No wonder, taken together, they make women sound ‘weak and ineffective’!

But this is a massive oversimplification: in reality, the same linguistic form can have several different communicative functions. As I’ve pointed out in previous posts, ‘sorry’ and ‘just’ are good examples. ‘Just’ can be used for emphasis, as in Nike’s ‘Just Do It’; the existence of the phrase ‘sorry not sorry’ shows that ordinary speakers understand the potential for ‘sorry’ to be used in thoroughly unapologetic ways (like the snarky way I used it myself earlier on).

The prefatory formula ‘I’m not an expert’ is a similar case. A search for examples on Twitter confirmed that it’s not always used in a genuinely self-deprecating way, to mean ‘ignore me, I don’t know what I’m talking about’. On the contrary, its most common function seems to be (a) suggesting that some previously expressed view is mistaken, and (b) marking the speaker’s own view as obvious common sense. For instance:

I’m no expert in these things but I don’t think people who send death threats tend to write them by hand on monogrammed notepaper

I’m no legal expert but I don’t think Rudy Giuliani is helping his client by going on TV, sweating profusely and voice cracking, saying that he was working on behalf of the State Dept when he pressed Ukraine to dig up dirt on Joe Biden.

I’m no expert, but I don’t think removing UK nationality from UK citizens overseas was on the ballot

Used in this way, ‘I’m no expert, but…’ does not downgrade the speaker’s claim to knowledge; rather it implies that the speaker knows better than whoever came up with the claim she’s ‘politely’ (i.e., sarcastically) contradicting.

Of course, there are cases where speakers do want to emphasise their lack of certainty. In the tweets I looked at, they tended to do that by using ‘I’m no expert’, or some variation on it, not as a preface to an assertion, but in a disclaimer following it. There’s an example in the tweet reproduced below, which was accompanied by a photo of an insect:

X found this beauty under our front porch table. My guide suggests Capnodis, a buprestid (jewel beetle). (But I’m no expert, that might be complete nonsense!)

So, what ‘I’m no expert’ communicates can vary, and one clue (though the pattern isn’t totally consistent) is its position in the utterance. As a preface, especially when followed by ‘but I don’t think…’, it’s unlikely to be heard as self-deprecating and ‘weak’.

Problem 3: if ‘research’ is cited, has it been represented accurately?

At this point you might be thinking: OK, but most people don’t subject everything anyone says to a deep linguistic analysis. They go by first impressions, and according to the TES writer there’s evidence that this works against women:

…linguistic differences have a tangible impact: researchers have found that subjective views of men and women vary enormously, with women far more likely to be described as indecisive, inept and panicky.

This sentence invites readers to infer that research has shown a causal relationship between the propositions on either side of the colon. It implies that women’s ‘linguistic differences’ from men are the basis on which they are judged to be ‘indecisive, inept and panicky’. But if you click on the link the writer provides and read the text it takes you to, you soon discover that the research she’s citing does not show, or claim to show, any such thing.

What you get if you click the link is a piece published in the Harvard Business Review, summarising a study in which researchers compared the words that were used about male and female leaders in a sample of more than 80,000 military performance evaluations. Although men and women scored similarly on various ‘objective’ measures of performance, these more subjective assessments of their strengths and weaknesses showed striking differences, as displayed in the graphic below:

W180511_SMITH_MANAGERSUSE-850x576

This study belongs to a growing body of work which seeks to uncover covert biases in organisations’ hiring, promotion and funding decisions by looking for sex-differentiated and/or gender-stereotyped patterns of word-choice in (for instance) job advertisements, performance evaluations and grant proposals. The existence and potential significance of this kind of bias is not in doubt; but what does it have to do with women’s alleged use of a ‘weak and ineffective’ speech-style?

The answer is, ‘nothing, so far as we can tell’. Apart from being about language, the claims made by these researchers are entirely unrelated to the contention that women’s style of speaking has a tangible impact on the way others perceive them. As is often the case in bullshit articles about women’s language, the allusion to what ‘researchers have found’ is just window-dressing: at best it’s irrelevant, and at worst it’s downright misleading.

Problem 4: is the writer proposing a workable solution to whatever she’s identified as a problem?  

Articles about women’s language are big on avoidance, and many of the things they tell women to avoid are conventional markers of politeness. The writers seem to think formulas like ‘sorry’, ‘just’ and ‘could we possibly’ are mere fluff: they add nothing of substance and can therefore be dispensed with. But while it’s true they contribute little substantive information, they are crucial to maintaining the interpersonal relationships on which successful communication also depends. Dispensing with them will not make you sound, in the TES writer’s words, ‘courageous and authentic’; more likely it will make you sound uncooperative or even hostile.

This problem becomes evident when you look at the way advice-givers tell women to reformulate utterances containing ‘forbidden’ expressions. In the TES piece, for instance, we get these suggestions:

Change “I would like to…” to “I want to…”
Change “If you get a moment, would you be able to…” to “Can you…”
Change “I know you’re really busy, but…” to “This is urgent.”

So, instead of asking a co-worker for a favour like this:

I know you’re really busy, but I would like to discuss your report with John before tomorrow’s meeting, so if you get a moment, would you be able to make a summary and send me a copy by the end of today?

you should do it like this:

This is urgent. Can you send me a summary of your report by the end of today, because I want to discuss it with John before tomorrow’s meeting.

But from a politeness perspective this is terrible advice. You might be irritated by the obsequious tone of the first email, but you’d surely be infuriated by the brusqueness of the second. Is the writer a human or a robot? If the former, who does s/he think s/he is? And that’s not just my own intuitive judgment. Research done during the heyday of assertiveness training found that if you follow the orthodox recommendations (be direct, don’t hedge, don’t apologise or explain yourself, stick to talking about your own needs and feelings), most English-speakers will think you’re weird, rude and obnoxious.

If you’re a woman, following this kind of advice has an additional cost: you may well find that you’ve just exchanged one stereotypically negative judgment–that you’re ‘weak and ineffective’–for another–that you’re ‘aggressive, shrill and strident’. These two judgments may look like polar opposites, but at a deeper level they’re two sides of the same coin. The logic they follow is one in which (1) whatever men are said to do with language is axiomatically preferable to whatever women are said to do with it, but at the same time (2) any woman who doesn’t behave according to stereotype will be negatively judged for her ‘deviance’.

This is what I mean by ‘a woman’s place is in the wrong’. Whatever women do with language is always either too little or too much: too passive or too aggressive, too self-effacing or too domineering, too feminine or not feminine enough. The overall effect is to render women’s speech, whatever it may consist of, less legitimate than men’s. And it’s not just about delegitimising women’s speech: criticisms of women’s language function as post-hoc justifications for sexism more generally.  ‘I can’t vote for her because of her shrill voice’/ ‘we didn’t promote her because she’s always apologising’…these are just coded ways of saying that you don’t think women are up to the job. They’re used to rationalise prejudices the critic already holds, and defend decisions which were really made for other, even less defensible reasons.

Articles like the one in the TES help to justify this kind of sexism by suggesting that it’s a rational response to women’s own linguistic inadequacies. If you don’t talk like a boss, you can’t expect to be treated like one. But what that doesn’t explain is the treatment meted out to women who do sound like bosses. Those women, the Hillary Clintons and Greta Thunbergs, are no less harshly judged: they just get accused of a different set of female deadly sins.

The conclusion I draw is that the advice-givers are reasoning backwards. It isn’t women’s speech that causes prejudice against them, it’s prejudice that leads to the devaluation of their speech. Instead of fighting our own alleged bad habits, we should put our energies into fighting prejudice against women speakers. Because that’s the root of the problem: everything else is just bullshit.

Sorry, but it’s complicated

‘Sorry’ may have been the hardest word for Elton John, but to women it comes as naturally as breathing. Women, as everyone knows, apologise. They apologise constantly. They apologise unnecessarily. They apologise for things that aren’t their fault, and for things that require no apology from anyone. They’re like the proverbial cracked records with their ‘sorry, sorry, sorry’.

Recently we’ve become obsessed with the problem of the over-apologetic woman. She seems to pop up everywhere—not just where you’d expect to find her, in listicles with titles like ‘The Seven Ways Women Undermine Themselves At Work’ and products like the ‘Just Not Sorry’ plug-in which I wrote about last year, but also in shampoo adverts, comedy sketches and various other forms of humour (check out #7 in this set of cartoons).

The advice and coaching industry has an obvious interest in endlessly recirculating the idea that women apologise too much (it’s in the same category of female verbal misdemeanours as uptalk and ‘just‘ and ‘I feel like‘–all of them gifts that keep on giving if your business is creating problems that people will then pay you to solve). It’s not so obvious why it also gets uncritically recycled by feminists (there’s an example in this Woman’s Hour interview with the historian Amanda Foreman). It might seem that feminists have nothing to gain by repeating what is, after all, a negative stereotype; but some evidently see it as proving a point about the harmful effects of female socialisation. Excessive apologising, they say, is the behaviour of a person who’s been trained from early childhood to think of herself as a lesser being: to devalue and efface herself, put others’ needs before her own, take up as little space in the world as possible, and defer, in particular, to men.

As a feminist myself I don’t doubt that female socialisation has many damaging consequences for girls and women. But I still don’t buy the argument that women over-apologise because they’ve been socialised to be self-effacing and deferential. It may sound plausible, but here’s why I don’t think it’s true.

First, let’s just go back a step. So far I’ve been taking it for granted that women really are the more apologetic sex. But that shouldn’t be accepted without question: the evidence from research is mixed. Some studies have found women apologizing more than men, but others have found no difference. Which findings you give more weight to is a judgment call, and researchers have different views.

My own view, for what it’s worth, is that there probably is a tendency for women to apologise more frequently than men (though the usual caveats apply: this generalisation may not be valid for every group of women in every context, and it certainly won’t be valid for every individual woman). However, there’s a difference between description and interpretation: even if we accept that this tendency exists, it doesn’t follow that women apologise ‘too much’ (too much in relation to what?), nor that the reason must be their socialised unassertiveness and deference. In fact, if you dig a bit deeper into the research, you discover there are other patterns which don’t fit with that assumption.

One of the most detailed investigations of gender differences in apologising was done in the 1990s by the New Zealand linguist Janet Holmes. Holmes recorded naturally-occurring talk in a range of social situations, and extracted all the sequences that included apologies (183 in total). As well as counting the frequency with which men and women apologised, she looked at who they apologised to, what they apologised for, how they formulated their apologies and how those apologies were received.

Holmes’s is one of the studies I mentioned earlier which did find women apologising more than men. A lot more, in fact: women produced nearly three quarters of all apologies. But there was another, equally striking pattern: the majority of female apologies, accounting for 55% of all the apologies in the data, were cases of women apologising to other women. Apologies from women to men accounted for about 18% of cases, apologies from men to women for about 17%, and apologies from men to other men for just 8.5%.

These figures are hard to reconcile with the belief that women’s apologies are displays of deference and subordinate status. If that were the issue, you’d expect women to apologise more frequently to men than to other women; you might also expect women, as the subordinate sex, to receive fewer apologies from men than they give to men. In this study, however, cross-sex apologies occurred with more or less equal frequency in both directions. Almost the whole of the overall gender difference was the result of the very large difference between women’s behaviour towards other women and men’s towards other men.

What Holmes actually found, then, was not a general tendency for men and women to behave differently, but a more specific tendency for them to behave very differently with members of their own sex. The standard story about unassertive and deferential women doesn’t explain that. Why would women become less deferential when addressing their social superiors, men, and conversely why would men defer to their subordinates, women, twice as often as they defer to their male peers?

But if it isn’t about deference, what is the gender difference about? Holmes offers two suggestions.

The first is that men and women may understand the act of apologising differently, with women seeing it primarily as a way of maintaining good relationships by displaying concern for others, whereas men see it primarily as an admission of inadequacy or guilt. This second interpretation is more negative, and that may explain why men avoid apologising (or dare I say, ‘under-apologise’?) with other men who share their understanding of it as demeaning. (Though it doesn’t really explain why men are more willing to apologise to women: perhaps they think women won’t judge them in the same way, or perhaps they don’t care so much what women think.) Among women, on the other hand, apologising is interpreted more positively: it need not diminish your status, and it may even earn you credit for being a nice person.

Holmes’s second suggestion is that men and women may have different perceptions of what’s offensive enough to require an apology. In her data, a higher proportion of men’s apologies were for relatively serious offences (like forgetting a date with someone or damaging one of their possessions). This led her to wonder if women apologised more because they felt the need to apologise for things men would dismiss as trivial. (Interestingly, she found women were more likely than men to apologise for what she calls ‘space and talk offences’—things like inadvertently touching or interrupting another person. She suggests that women may be more sensitive to these transgressions because they’re so often committed against women–most typically, of course, by men.)

The idea that men have a higher threshold for perceiving behaviour as offensive has received some support from other research. The psychologists Karina Schumann and Michael Ross  got research subjects to keep a diary of all their offensive acts, and to say, in each case, whether they had apologised. They found that women and men reported apologising for a similar proportion of the offences they’d committed, but the raw numbers—of both offensive acts and apologies—were higher for the women than the men. This suggested that women had considered more acts offensive enough to need redress. To test this hypothesis, the researchers designed an experimental task where subjects had to rate the seriousness of various offences. They found, as they expected, that their female subjects gave higher ratings.

Readers who know something about the study of language and gender may think that Holmes’s arguments are somewhat reminiscent of what’s been labelled the ‘difference’ or ‘two cultures’ approach, because it treats differing patterns of language-use among men and women as the product of quasi-cultural differences. The idea is that the sexes behave differently because they understand things differently, and they understand things differently because their experiences and values are different. But you can’t say that one sex’s way of doing things is better or worse than the other’s, any more than you would say Japanese culture is better or worse than Greek culture. They’re just different: ‘different but equal’. The classic statement of this position is Deborah Tannen’s popular bestseller You Just Don’t Understand, which has been criticised by other scholars (full disclosure: I’m one of them) for glossing over the issue of structural sexual inequality. Does Holmes’s explanation of gender differences in apologising invite the same criticism? Is she saying that women’s tendency to apologise more than men has nothing to do with power and inequality?

The short answer is no: Holmes’s position is not the same as Tannen’s. She believes that ultimately, ‘power is the issue’:

Women’s ways of talking differ from men’s because each group has developed interactional strategies which reflect their societal position. The different patterns of interaction into which girls and boys are socialised are not randomly different. Their features are attuned to the requirements of the society.

As a general observation this is not, of course, incompatible with the popular story about over-apologetic women (which says that women’s societal position requires them to be unassertive and deferential). The trouble with that story is that it leaves some patterns unexplained (in particular, the difference Holmes observed between cross-sex and same-sex behaviour). The relationship it posits between apologising and subordinate status is too simple and general to account for the reality of women’s (or men’s) behaviour. But there’s another way to relate apologising to the requirements of an unequal, male dominated society. Rather than treat it as a general expression of female powerlessness, we could connect it to a more specific feature of social structure which is directly linked to women’s subordinate status: the division of labour which makes caring for others a female responsibility.

The aim of an apology is to maintain or restore harmony. Apologising says: ‘I accept that I have given (or might give) offence, I regret that and I ask for your forgiveness’.  By implication it also says: ‘your feelings matter to me: I understand that you may feel bad and I want you to feel better’. In that sense apologising can be seen as a form of emotional labour, part of the work of managing your own and others’ feelings. And willingness to perform emotional labour is one of the most basic things that’s expected of women just because they are women. Even when they’re not explicitly cast in a caring role (like ‘mother’), women are routinely expected to pay attention to others’ feelings, and pour oil on troubled waters when harmony is threatened. They’re expected both to apologise when others are or could be offended, and to forgive when others have offended them (Holmes found that women were less likely than men to reject an apology and more likely to accept one).

I would argue that women’s apologising behaviour has more to do with these gendered expectations than it does with lack of confidence and self-esteem. In fact, I suspect that for some women, this behaviour may actually be a source of self-esteem. Not all women embrace their prescribed role with joy, but even if we don’t much like our jobs, most of us derive some satisfaction from being good at them. And while selflessness and nurturance may be poorly rewarded in material terms, they do come with moral and social benefits.

In this case one key benefit may be the approval of your female peers. Although you often hear women castigating themselves publicly for the sin of ‘over-apologising’ (a term  which covertly treats what men do as the norm), the high frequency of apologies between women suggests that privately they may not regard men’s behaviour as the gold standard. And indeed, why should they? What’s wrong with showing concern for others? From a feminist perspective, nothing: the problem isn’t that caring is inherently worthless or degrading, it’s that the responsibility for caring is unfairly distributed, obliging most women to do too much while most men do far too little.

This already unfair distribution of rights and obligations can give rise to a further injustice, with men getting more credit than women when they do show consideration for others. In a study conducted by a group of researchers in Israel, subjects were asked to imagine that a work colleague had scheduled a meeting and then failed to turn up for it. They were then given a letter which the colleague had supposedly written, apologising for the inconvenience and promising it would not happen again. The letter was always the same, but some subjects were told that the writer was their boss, others that the writer was their subordinate. In addition, the writer was presented to some subjects as male (‘David’) and to others as female (‘Rachel’). After reading the letter, the subjects were asked to judge the sincerity of the apology and to say how willing they were to forgive the writer.

The researchers found that people’s judgments were influenced by both status and gender:

the most effective apology was by a male manager, followed by a female manager, a male subordinate and finally a female subordinate.

They explain:

The less expected an apology, the more effective it is.  …When an apology is not socially obligatory, [it] will be more accepted and more highly esteemed.

A subordinate who apologises to the boss is merely doing what’s expected, whereas a boss who apologises to a subordinate is perceived as displaying unusual magnanimity. Gender has a similar (though slightly weaker) effect. A woman who apologises is meeting others’ expectations, whereas a man is exceeding them, and the response is therefore more positive. The researchers conclude by suggesting that managers shouldn’t treat apologising as a ‘powerless’ and therefore risky move: done strategically, it can actually enhance your reputation. Especially, it seems, if you’re a man.

Where does this leave us? Again, I’d say it’s complicated. On one hand, I think we should reject the popular stereotype of the over-apologetic woman, along with the language-policing it’s used to justify. It isn’t true that women ‘over-apologise’ because they are timid, deferential creatures: the fact (if it is one) that they apologise more than men seems to have more to do with showing concern for others’ feelings. Do feminists really want to disparage that as a weakness?

On the other hand, as Janet Holmes points out, women and men don’t make linguistic choices in a vacuum: they are expected to behave in certain ways, and if they choose not to they will pay a price. Maintaining harmonious relationships is an obligation imposed on women rather than something that just comes naturally; it is one of the many unjust burdens women carry in this world, and as such there is every reason for feminists to criticise it. The question is whether we can find a way to do that without ridiculing, shaming or bullying women for the way they do or don’t speak.