Friday, 20 January 2017

Creating a balanced reading list: speculative fiction by women and people of colour

One of the joys of creating a new course is putting together the reading list, finding a wonderful cross-section of authors, heaping up a pile of your favourite books, anticipating the pleasure of students recognising favourites or discovering fabulous books they might not yet know. Recommending books to people is always a delight. And then there's the more logical, intellectual pleasure of balancing out the list: in this case, for the Imaginary Worlds course, I'm looking for top-notch examples of each week's theme, but also thinking of a matrix of other factors. Have I included all the genres, do I have steampunk, urban paranoramal, alternative history, Young Adult fiction, dystopia, literary fiction, epic fantasy, space opera, hard sci-fi? Are any of the genres over-represented? Are these all the best examples I can think of? Is it a satisfying range of styles?

I sit back, study the completed list, and sigh with pleasure. This is a beautiful list of books. And then I do the gender count.

The gender count pisses me off. Guess what, my list is skewed towards men again. Well, so the hell what? This is a great list of books! And it's carefully balanced and I don't want to change it! Why should I try to change a fabulous list of recommended reading just to include more women?

Every time I do the gender count and strike out for gender balance, I get frustrated - not frustrated that even as a feminist, I've always chosen more books by men, but frustrated at doing it at all. I feel resentful and irritated. I grit my teeth and do it anyway, because it always ends up being worthwhile. I have always missed out brilliant, canonical work by women that absolutely should be on the list. When I wrote a blog post about unsympathetic characters, something female authors are far more criticised for than men, I easily thought of Kazuo Ishiguro's Mr Stevens and Ian McEwan's revolting Michael Beard in Solar. Only after I did the gender count did I think of Jane Austen's Emma and AS Byatt's Frederica. Emma is the definitive example of a character "whom no-one but myself will much like" and Byatt is my all-time favourite author, but both had slipped my mind. When I created the reading list for the Magical Realism workshop, I'd included Neil Gaiman's Neverwhere (which has a tangential claim to magical realism at best) and forgotten both Laura Esquivel and Isabel Allende, both Latin American authors at the absolute heart of the genre. For the Imaginary Worlds course, probably the most glaring error was including Stephen King's Green Mile as an example of theme (good theme, but it's barely speculative fiction) and forgetting The Handmaid's Tale by Margaret Atwood. I'm not even a huge fan of Stephen King - I don't read much horror and that's the only of his books that I've thoroughly enjoyed - whereas I've studied Margaret Atwood, and read, loved, and own everything she's written.

Why does this happen? Because male authors are just more visible. Male names count for more. To everyone. It's systemic sexism. Systemic means it's in the system: it's not nasty men oppressing noble women. That's worth repeating: it is not men against women. Both men and women are sexist, and that sexism isn't coming from some inner evil, it's coming from the system, ie, our culture. To really grasp this, it's worth looking at just how widespread it is throughout our culture.

One of the most famous anecdotes is about an orchestra trying against the odds to get a gender balance of musicians. In Geena Davis's telling, 'In the 80s, after a long effort to equal the gender composition of orchestras in the US, where they’d slowly increased the number of female musicians from 5% to 10%, they came upon the idea of “blind” auditions. If the panel, behind a curtain, couldn’t see who was playing, then they wouldn’t be able to discriminate. It worked! Sort of. The numbers of female musicians rose, but not significantly.' At that point, no doubt some people suggested that maybe some women were as good as men, but women clearly didn't have an equal talent, because even without gender, they still weren't equally represented. But the orchestra kept at it: 'Was some element revealing their gender and skewing the results? I’d love to have been in that room when they finally rasped: “Carpet the stage!” After the next round of auditions, Davis grinned, the orchestras were 50% women, because the panel had been able to hear their heels.'

It's hard to work out whether something really is sexism, rather than other confounding factors: maybe men are just better than women at some stuff, maybe women have different preferences, maybe the competing demands of childcare mean women can't pursue an area to the same level of excellence, and so on. The only way to be certain is to find a way to strip out gender completely, as the orchestra did, or test responses where the only change is gender.

One double-blind peer-reviewed study on academic hiring took the second option. They created application packs where the male and female 'candidates' were identical, except for gender, and got a range of participants to evaluate them. 'Faculty participants rated the male applicant as significantly more competent and hireable than the (identical) female applicant. These participants also selected a higher starting salary and offered more career mentoring to the male applicant.' The 'men' were rated more highly, purely for having male names. Importantly, the gender of the person making the decision didn't change anything: both women and men discriminated against women and in favour of men. Again, sexism is not something men do to women: it's an unconscious bias that both men and women have.

The value of studies like these is that it completely removes any confounding factors, including any arguments about whether the women are really as good as the men, arguments which can (and are) thrown at any real-world study. If you read a real-world study that women are cited less in academic papers, you might argue that maybe their work isn’t as good or there's less of it. But in another blind study, 'students gave higher ratings to identical abstracts submitted with male author names, associating them with greater “scientific quality.”' Again, quality was rated more highly, just because of a male name. (If you want more on gender and racial bias in academia, there are a heap more studies collated here.)

It's not just orchestras and academia. The same thing repeats across sector after sector: having a male name is an advantage; having a female name is a disadvantage. In tech, hires of women increased from 5% to 54% when gender indicators were taken off all applications. That's from 1 in 20 hires being women to more than half.  In medicine, in another of those nifty double-blind just-change-the-gender studies, doctors looked at reports from 'patients' of their symptoms: 'Females were rated less seriously ill, less likely to require laboratory tests, and more likely to receive medication than males. Among depressed patients, counseling and reassurance were more likely for females, and a nonpsychiatric consult was more likely for males.' Remember: again, these were from identical reports. All that changed was the gender.

The disadvantage of being perceived as female runs across our culture. Even in maths tests – surely the most objective of exercises? – boys are marked more favourably than girls if the teacher knows their gender. Somehow maths can be more right if it has a boy's name attached. External examiners, who didn't know the children's genders, marked them equally.

Despite what all the stats and all the double-blind studies say, though, our perception can be very different. Crowd scenes in films famously have around 17% women, on average, and it doesn't strike anyone as odd, on screen. Even more startlingly, as Davis told NPR, from the studies of Gender in Media, 'If there's 17 percent women, the men in the group think it's 50-50, and if there's 33 percent women, the men perceive that as there being more women in the room than men.' The same skewed perception crops up in conversation: we think women are dominating a discussion when actually the men are. And when I say "we", I mean both men and women. (There’s a whole lot more on how we perceive men and women talking compared to what actual linguistic studies show here. Tl;dr: every cliche is wrong.) Even on Twitter, men are retweeted twice as much as women.

That's what systemic means: it's throughout the culture, replicating itself, as a bias that everyone has in favour of male names and against female names. And in today's No-Shit-Sherlock Awards, the world of books is not magically immune to this. Just as you'd expect, having a male or female name affects who gets reviewed, who gets recommended, who gets prizes, and who gets onto reading lists.

One of the longest-term studies into reviews was a 28-year study from Australia by Dr Julieanne Lamond of the Australian National University and Dr Melinda Harvey of Monash University, looking at 1985 to 2013 found that 'two-thirds of the books being reviewed are by men – a ratio that has remained largely the same for 30 years.' It hasn't changed since the 80s. And in case you're thinking maybe that's just Australia, the VIDA Counts yearly tot-up doesn't show anything like parity here in the UK: 'One of the worst culprits was found to be the London Review of Books which featured 527 male authors and critics on their pages in 2014, compared with just 151 women.' That's 7 men for every 2 women. 'It also saw a rare drop in reviews of books written by women from the year before, with 14 fewer than in 2013.'  And in the US, 'The New York Review of Books displayed a similar imbalance, featuring an overall 677 men to 242 women.' That's 3 men for every woman. 'The New York Times book review featured an overall 909 male contributors and authors, compared with 792 women; the Nation’s male-female split was 469 to 193; and at Harper’s fewer than half the authors reviewed were women.' That's a lot of visibility going to men and not to women.

Prizes show an even more striking discrepancy. In the sweet summer days of naivety, before I knew anything about how anything in publishing worked, I used to think that prizes simply went to the best books. I don't know how I thought that. Gradually, though, I learnt how prizes actually work and how much opportunity the process holds for that invisible bias. In the big prestigious prizes, it's usually the publishers choosing who to put forward for that prize - and of course they're not immune to the male-name effect, but they might also reasonably consider which of their authors have had the best reviews. (Let that sink in.) That's before the judges even see the books. In fandom prizes, it's a lot of nominations from a bunch of fans, and then people voting from a slate, which seems like a very level-playing-field crowd-sourced way to go about it, until you remember that the men have way more visibility, and male names, and we all take stuff by men more seriously and rate it more highly. (Which all the studies at the start of the article were saying.) To imagine that a prize process could happen without being touched by this widespread cultural bias is... nonsensical. Of course it's affected. And the stats bear that out.

Nicola Griffith 'conducted an audit of the past fifteen years of books that won the Pulitzer Prize, Man Booker Prize, National Book Award, National Book Critics’ Circle Award, Hugo Award, and Newbery Medal. She found that the overwhelming majority of prize-winning books were by men. No surprise there, perhaps, but the rabbit hole went deeper: when Griffith looked at the gender of the protagonist, she found that the overwhelming majority were male. In fact, over the fifteen years of prize-giving, the Pulitzer was awarded to a book with a female protagonist a grand total of zero times. Even the best man for the job of protagonist is a man.' (You can read the full breakdown of the audit here.)

In all the double-blind studies above, when gender was removed, we got gender parity; when gender was included, and nothing else changed, we got a strong bias in favour of male names. And that strong bias in favour of male names bears out exactly what we see in the large-scale real-world data. So on average, men and women's work is of equal quality, but work with a female name is rated weaker.

In 2015, Catherine Nichols conducted her own “blind” study in book submissions. Frustrated by a stream of rejections, she tried sending out her manuscript under a male name, let's say 'George', using the same novel and the same submission letter. Only the name changed. She discovered, 'He is eight and a half times better than me at writing the same book.' Of 'his' 50 queries, he had 17 requests for the manuscript; under the author’s actual name, her figures never went above 1 in 25.

All this visibility, for male authors, adds up. The bias is definite and real and we can show it again and again with large-data reviews and blind-data studies, but we never want to think it applies to us. 'I just chose the best books!' I huff. Funny how I never end up with more women’s books when I 'just choose the best books'. Nobody wants to think they’re sexist. Everybody is sexist. Because it’s not a nasty thing that nasty people do, it’s a systemic unconscious bias. I'm sexist. Of course I am, I'm not immune to my culture. My only weapon, when I'm compiling lists of recommended reading, is the gender count. And still, even being committed to it, I’m intensely irritated by the process. Every time, I find myself thinking, 'Why should I include authors just because they’re women?'

But actually, I'm choosing men just because they're men – just because being men, having male names, makes them more visible. More reviewed, more recommended, given more prizes, taken more seriously, rated more highly. The stats show incontrovertibly that women are unfairly rejected, for their gender, in sector after sector. But what we often fail to consider is the flip-side: that men are being unfairly favoured, for their gender.

Every time I do the gender count, I find better examples when I force myself to think of books by women. The flipside? I'm instinctively choosing weaker examples because they're written by men. Recently, when my partner was choosing between two new books by authors he knew nothing about, one male and one female, I said, 'Pick the one by a woman. It'll be better.' Statistically, a book by a woman will be better than a book by a man. The bar is set higher. That seems such an inflammatory statement, and goes so against our unconscious bias, that it’s worth breaking it down statistically.

Firstly, we know the quality of work is evenly spread: when gender is taken out of the equation, men and women’s work is chosen equally. So the available quality looks like this:

So if you were taking 50% of each, judging purely on merit, this is what your selection would look like:

You've got all the A and B quality work from both, and a sprinkling of Cs from both. But that's assuming absolutely zero interference from our cultural bias. That's assuming 'George' isn't eight and a half times better at writing the same book as its actual female author, that men and women are getting equal reviews, prizes, and visibility, that they're getting equal levels of recommendations and retweets, and that there's no male-name effect suggesting that work with male names is better and more serious. But of course none of that's true. Our culture runs interference at every point, through our own unconscious bias.

So let's say our culture swings it just 20% in favour of men. Women’s share goes down to 30%, men’s goes up to70%. And this is our selection of books:

All the A-quality work still gets through. But overall, there is less B quality work and more C-quality work. And suddenly,  D-quality work starts appearing.

Now choose, at random, one from the red group and one from the blue group. Statistically, the one from the red group will be better. In other words, you have more chance of choosing a good ‘un from the red group than you do from the blue group. Statistically, once a book with a woman's name has made it through the various hurdles of the male-name effect, only the top stuff gets through.

But of course, that’s not how we choose, because we’re also choosing on visibility and with the same cultural bias that creates this situation in the first place:

 Choose one of the first ones you can see. It’s blue. The work by women is still there. It's just... not as visible, somehow. The Gaimans, Kings, and McEwans come easily to mind while you have to scratch your head and squint to spot the Atwoods, Byatts, Equivels, and Allendes. Again, and I cannot repeat this enough, this is not men against women. Both sexes show systemic gender bias. Men benefit more from it, but we all do it.

I did my gender count. I gritted my teeth, grumped and harrumphed, scratched my head, and resented it, and did it, and I ended up with a much better selection of work, as always. I felt the same delight I feel every time, at the much better reading list I've compiled, and the same embarrassment at the work and authors I'd forgotten, and I leant back, sighed happily, and scanned my gender-equal list.

Then something in my head twisted. I don’t know what removed my usual spectacles. I think I was questioning whether I had a really good example of urban paranormal, and thought of Ben Aaronovitch - then I thought, 'Hold on, wasn't his protagonist black? Is Ben Aaronvitch black?' Googled him - nope. Then I suddenly looked at my list through different eyes, and scanned down the author names. Everyone on the list was white. Not one person of colour.  At least I'd had a few women, to start with, in the original list. And if the bar is set higher for women, how much higher is it set for people of colour? In other words: how much incredible stuff am I missing out on? How much A-list work am I not including?

I have years of curated studies and stats on the systemic bias against female names. When it comes to race, though, I have the clear gulf of my own blindness. The 2015 VIDA count was the first to count race as well as gender, and just as my whitewashed reading list indicated, it only gets worse. For instance, in 2015, the New York Review of Books featured 4 men for every 1 woman. But women of colour were only 10% of the women. I hone in on data about bias against female names, because that affects me directly, but had stayed oblivious to bias against non-white names, even when the same articles were flagging that up, as in the VIDA count.

The same picture of name-bias emerges as something deep, systemic, and running across fields. A just-change-the-colour study into job applications in the US found people with black-sounding names had to send out 50% more CVs to get a call back - ie white names were 50% more likely to get a positive response. A similar study in the UK found that people with white names were 75% more likely to get a positive response. The same pattern shows in academia. Researchers sent out almost 7000 emails from 'potential students' enquiring about a course, changing only the students names between male / female and white / other groups, and found it's not just a male name that helps: it's a white male name. And again, in the No-Shit-Sherlock awards, the same sytemic bias pops up in publishing. The Bookseller found more people called David reach the bestseller list than people of colour: in the top 100, you have 11 Davids to 1 person of colour. Nine months after it launched, the new Jhalek Prize for books by BAME (Black and Ethnic Minority) authors had received a paltry 51 submissions from publishers.

Nonetheless, my first reaction to my all-white reading list was just the same as the gender count: 'Well, so the hell what? This is a great list of books! And it's carefully balanced and I don't want to change it!' The amount of resistance I feel trying to rebalance my own bias is extraordinary. In this case, I had to combat not only my own resistance but my own blind ignorance. My bookshelves weren't a total whitewash, but they were when it came to speculative fiction. (I'm discounting magical realism, because that uses world-building completely differently so it isn't included on the Imaginary Worlds course.) The temptation to brush it aside was so strong, but I didn't, for the most selfish of reasons. If work by women that makes it through is statistically better, and the bar is set even higher for people of colour, how much incredible stuff am I missing out on?

Cue lots more research. I started Googling and finding long list after long list after authors I’d never heard of. I drew up complicated spreadsheets to see which names were recurring on multiple lists and trying to work out just how much I could read. When I was totally tangled in my very long lists, I asked for help on Twitter - I really was trying to do my own research, I said, but I could do with some hand-recommendations as well. (It's always tricky, asking the people who're on the receiving end of your unconscious bias to help you with your homework.) Lea Fletcher, Joanne Hall, Aliette de Bodard, Cheryl Morgan, and 'Old Sareemaa Witch' all came to the rescue, listing further suggestions, advising which of the names on my long list were canonical authors, and flagging up names that were missing that really should be there.

Having thought I'd completed my reading list, I now had a list of 37 other authors to consider and a ton of reading to do. That, at least, was the very very good news! I started downloading Kindle samples, buying books at Blackwells, devouring a book a day, and hungrily happily immersing myself in new worlds. I'm always looking for new speculative-fiction authors and suddenly discovering a whole range of new authors was Christmas! I also had a new logical puzzle with a fresh complicated spreadsheet, to include at least 50% authors of colour, once more make sure the list was gender balanced, that fantasy and sci-fi were equally represented, that all the subgenres were represented, that each week's theme had top-quality examples of that particular aspect - and that the books I wanted to include were actually available for my students, in physical bookshops in Oxford.

The final reading list is, frankly, dazzling. It's a full 16 books (on average, we'll look at two extracts in each week of the eight-week course), so I don't expect my students to read all the books before the course, but I am thrilled at the calibre and range of books I'll be introducing them to. It's not meant to be “the 16 best 16 SFF books” – there are far too many fabulous authors not on the list – but it is a fantastic cross-section:

     Audrey Niffenegger, The Time Traveller's Wife     Cixin Liu, Three Body Problem     George RR Martin, A Song of Ice and Fire     Kazuo Ishiguro, Never Let Me Go - NB: Read this one in advance, to prevent SPOILERS     Ken Liu, The Grace of Kings     Margaret Atwood, The Handmaid's Tale     Malinda Lo, Ash     Nalo Hopkinson, Midnight Robber     Nisi Shawl, Everfair *     NK Jemsin, Fifth Season     Nnedi Okorafor, The Book of Phoenix     Patrick Rothfuss, The Name of the Wind     Peter F Hamilton, The Neutronium Alchemist     Robin Hobb, The Farseer Trilogy     Suzanne Collins, Hunger Games     Terry Pratchett, Jingo

Working against one’s own systemic bias is frustrating and irritating at first – until you start discovering the amazing stuff you’ve been blind to, because it’s just not as visible. Like creating an imaginary world, you have to throw out some of the assumptions you didn’t even realise you were making, to discover a new world full of exciting new possibilities.

The Imaginary Worlds course starts in Feburary 2017 - email me to book or read more about it below.

Develop your world-building to improve or invent your own imaginary worlds: the Imaginary Worlds course is an eight-week evening course on writing the genres of imaginary worlds, starting February 2017, and covering... • the many genres • how to constrain magic • making unlikely stuff convincing • your world's physical detail • why your world matters • ripple-through effects • names and language • your characters' political and economic realities • techniques for exposition. Read more about it and book here.

Particular thanks, again, to  Lea Fletcher, Joanne Hall, Aliette de Bodard, Cheryl Morgan, and 'Old Sareemaa Witch' for helping me discover so many fabulous new authors. The eight works of speculative fiction by people of colour that are on the final reading list are Ash by Malinda Lo, Everfair by Nisi Shawl (pending final confirmation of its availability; since it won a prize recently, the publishers seem to be struggling to keep up with demand), Fifth Season by NK Jemsin, Midnight Robber by Nalo Hopkinson, The Book of Phoenix by Nnedi Okorafor, The Grace of Kings by Ken Liu, Three Body Problem by Cixin Liu, and Never Let Me Go by Kazuo Ishiguro (the one author I already knew!)

Some fabulous authors that I wanted to include, or would include on a longer reading list, I had to leave out - most notably, Octavia Butler, Samuel Delaney, Angela Carter, Scott Lynch, Neil Gaiman, Aliette de Bodard, Ben Aaronovitch, and Juliet McKenna, for starters. These were for various reasons: whether they were available in hard-copy in Oxford, whether they wrote novels rather than short stories, whether a particular sub-genre was already well-represented, avoiding overlap with my reading lists for other courses and workshops, not covering magical realism in this course, and simply limited space on the course.

The other speculative-fiction authors of colour I found through recommendations and assorted blogs are below. So if you're also looking for new worlds and incredible stuff you've been missing out on, fill yer boots!
  • Aliette de Bodard
  • Charles Yu
  • Cindy Pon
  • Daniel Heath Justice
  • David Anthony Durham
  • Derrick Bell
  • Gerald Vizenor
  • Jewell Parker Rhoades
  • Jy Yang
  • Kalpa Imperial
  • Karen Lord
  • L. A. Banks
  • Lee Perry
  • Octavia Butler
  • RSA Garcia
  • Saladin Ahmed
  • Samuel Delany
  • Sarah Kuhn
  • Silvia Morena Garcia
  • Stephanie Saulter
  • Steven Barnes
  • Tade Thompson
  • Tananarive Due
  • Terry Bison
  • Tobias Buckell
  • Vandana Singh
  • Wesley Chu
  • Yoon Ha Lee
  • Zen Cho

Get occasional emails about interesting things

* indicates required

I won't share your email with anyone else. You'll get emails from me only, about upcoming courses, writing competitions, publishing opportunities, interesting articles about writing, new blog posts, and creative events in Oxford. All emails are sent via MailChimp and you can unsubscribe at any time. Add to your address book if you want to keep the emails from vanishing into spam.