February 18, 2022

What did the sensitivity readers say? And did I care? Of all the aspects of the recent attempt to cancel my work, the one that seems to fascinate most people is the moment when my publishers sent my Orwell Prize-winning memoir, Some Kids I Taught and What They Taught Me, to be assessed by experts who would detect and reform its problematic racism and ableism.

Of course I cared. I’m horrified that people found prejudice and cruelty in my book. And I went into the process willingly: I’ve always enjoyed and benefited from editing and saw this as an extension. I did an initial rewrite — there were many things I was eager to change — in the autumn of 2021 and sent it off full of interest and optimism. I received the reports on it before Christmas. They were never formally used and I share the content here — anonymously, of course — because sensitivity reads are being used more and more widely, and mine gives a valuable insight into how they might work with non-fiction and memoir.


There are several reports — Picador did a thorough job — and they are varied. The novelty of the whole field is reflected in the fact that the Readers use different titles — sensitivity and authenticity — and different methods, too. Some write A4 reports, others use the comment button on Microsoft Word or an Excel sheet, still another presents a simple list of headings, done very possibly with a word search. More than one grades infractions, 1-3. They have of course special areas of expertise — Islam, blackness, disability — but these emerge through inference, not announcement.

Their scopes vary, too. One Reader fusspots around single words: I should not use “disfigure” of a landscape (infraction level 3, as presumably comparing bings — spoil heaps — to boils might be harmful to acne sufferers). Nor should I use “handicap” in its ordinary sense of “impede” (infraction level 2, serious); and I should prefer the acronym “SEN” to its origin phrase, special educational needs, because it is more inclusive (infraction level 2).

Others have grander ambitions: paragraphs, sub-sections and even entire chapters should be revised. Still others focus on issues around the presentation of the book. One suggests the authors of endorsements containing the words “love” and “humanity” might want to “rethink their stance”. To add to the cacophony, the Readers contradict each other freely, even praising and disparaging the same passages.

Given this diversity, it seems reasonable to start with areas of agreement. These mostly occur in the first part of my book, which is set in the Nineties. Perhaps this is because all of the Readers seem to be experts on sexuality and gender, and resisting homophobia is one of my themes. There is even a particular passage, the only one in the book, on which the whole Reader crowd comments and concurs.

The setting is London, 1992. After end-of-term drinks, a favourite student, Liam, comes out to me and then asks me to take him to G.A.Y — because, he says, no one else in his world would know where it was. I was very worried about doing this at the time; even though Liam had just left school, I still felt like his teacher, and I worry even more now, when teachers no longer take 18-year-olds to the pub and are much more aware of influence and consent.

None of these sensitive issues, though — raised at length in the book — worry the Readers. They are concerned, rather, that I might be boasting about helping a young gay person: “Straight white saviour trope”, suggests Wordsearch List, “could be problematic”. And they set up a chorus about what I feel and say after Liam hits the dance floor and I note:

… a new kind of pain, a physical, chesty anxiety that I was not to experience again until I watched my own children walk along ledges or cross a busy road. What would happen to Liam among all those strong bodies? What would happen to his body? He was too young to understand you only got one. Fortunately, it was only twenty minutes or so before he came back out of the crowd and grasped his beer.

‘Liam,’ I said, ‘I love you. You have to promise me to always use a condom and never get AIDS.’ 

I make, my Readers agree, a “reductive” and “rogue” remark. The preceding passage “comes across as homophobic” and is an LGBTQ infraction Level 2. But in 1992, people were still dying in large numbers from AIDS, and I would have urged all young people to use condoms. Excel Reader is kind enough to acknowledge this — “the author has chosen to reproduce contemporary dialogue which may not … reflect brilliantly on her” — but the other Readers seem to concur that the past should match an idealised present, in the same way that Anne of Green Gables, say, got a gay best friend when she went on Netflix.

There are similar injunctions throughout the text. I am enjoined not to quote from My Ántonia by Willa Cather, as it is “an old novel”; nor to state that homosexuality has historically been taboo in Nepal, as homophobia comes from colonialism; nor to mention that the Taliban were terrorists. Extending the principle of sunny improvement into the present, Wordsearch List breaks out of their list to make the helpful suggestion that I should remove references to terrorism from across the book, as it “over-sensationalises such a heavy topic, especially with minors involved”.

Nor should I say that more middle-class than working-class children go to university; nor that Foetal Alcohol Syndrome leaves children unable to progress; nor that a long tight dress restricts movement. All of these things are, for my Readers, “hurtful” notions of mine, not unfortunate facts. Writing, they imply, should represent the world as it ought to be, not as it is.

Perhaps this is a reflection of the sensitivity read’s origins in children’s and young adult fiction. There are good reasons for regulating children’s reading: it is foundational and formational and may be enforced by school choice or being read aloud to. It is genuinely important, there, to avoid oppressive stereotypes.

But Some Kids isn’t a novel, nor written for children. Adults are able to put books down if they upset them, so their books may safely contain difficult ideas. I don’t, for example, agree with my Readers that the references to looks, attraction and sexuality in my book should be removed in case readers are hurt by a metaphor as a child might plausibly be. I think adults can endure bings being compared to boils. I also believe that physical human beauty empirically exists, is enormously important for adolescents, and that I can observe its currency and often destructive power, especially for young women, in the classroom. I make an explicit argument about this, which readers may disagree with.

Argument, though, is not a word the Readers use, or even something they seem to think I should be engaging in. “Please avoid generalisations about anyone”, snaps Comment Button Reader from the margins, while Excel Reader trails a noisy stream of alerts, like a lorry reversing, down their squares: “chosen to make the stance/vent the opinion”, “Author gives her view”, “Author stance”.

I find Excel Reader irritating, but at least they acknowledge I am constructing a point of view. Comment Button, on the other hand, thinks I am accidentally misrepresenting the world because I am not aware of my “benevolent racism”. It’s exoticising, they say, when in my section on “Faith” I show a group of young people from diverse cultures and religions discussing the terrorist attacks of Paris 2015. “It is as if these children and their friendships are especially praiseworthy because of their backgrounds, rather than just treating them as people, as “normal” children”. But the young people were behaving in a special way that day: that’s why I wrote about it. I sincerely (Author taking a stance) found hope for humanity in the capacity they demonstrated for forgiveness and communication. I also believe (Author giving her view) that I am entitled to say so.

Similarly, in my chapter about art and poverty the Readers ignore several pages about history and economics to home in on a conflict between me and a character I named Cheyenne. The chapter is read as pure autobiography: “It’s clear the author did not like this pupil”. It’s “hard on Cheyenne personally”, sighs Excel Reader, “to the point [she] becomes an allegory of poverty”. But Cheyenne is an allegory of intergenerational poverty, a composite carefully constructed from details observed in different people over many years. She is a literary figure, not a person, doing what literary figures do: make abstract ideas visceral and moving.

My Readers though, have not been hired as literary people. They are there to help create a book that would play better on Twitter, not one that is better written. For example, they like my liberal conclusions to chapters so they recommend I put them as introductions. I should eliminate journeys of thought across chapters, ambiguity from paragraphs, and nuance from sentences. Love, they assure me, is never expressed with shadows.

I struggle with all this. I baulk, besides, perhaps snobbishly, at their language: the imprecision of phrases such as “feels like the kind of saying that could be deemed insensitive these days”, or “white knight tone/verve” (verve?). I snarl when Excel helpfully suggests I have made a typo with e. e. cummings, and lost his capital letters. It upsets me in particular, when so many of their criticisms depend on it, that none of the Readers deploy the word “irony”, but use “sarcasm”, “jocular aside” and “subtlety” instead, always as negatives. Comment Button condemns my entire chapter on Prizes as “it shows none of the adults involved in a good light”. Indeed it doesn’t. They are being satirised, even though one of them was me.

Irony, for me, is more than satire, and far more than sarcasm. It is a way of saying that two things can be true at once, of holding comedy and tragedy, romanticism and practicality, grand emotion and self-deprecation, together in a paragraph. Because it contains contradictory elements, irony doesn’t settle. It can keep the tension and life in any sentence: except of course on Twitter, where elements of a sentence can be taken out and caricatured; except in the hands of sensitivity readers, where it dies from granulation. As A4 Report comments: “I sometimes found it difficult to tell when the author was poking fun, and when she was not … if sarcasm is too subtle, all that is left is the bad implications”. They recommend stronger signposting.

Times change. In 1997, a second edition of my first poetry collection, Slattern, was published with Picador. I was thrilled, and immediately took a copy to show off at a festival in St Andrews. Two poems into reading from it, though, I was startled to discover the words had been changed.

A copy editor had decided my irony, commas, and word-choices were too eccentric, and had taken it upon himself to alter them in several poems. I emailed Picador, and the next morning found myself speaking (on a pay phone) to the legendary Peter Straus, Picador’s Publisher. He noted down each tiny adjustment himself and explained he would be ordering the pulping of the books, the carpeting of the copy editor and the printing of a new edition. He wouldn’t be thanked. He seemed to think this was Picador’s mission: to preserve, within its profit-making superstructure, a space for the miniature nuances of a poet.

A quarter of a century later, publishing had moved on. Philip Gwyn Jones, the new holder of Peter Straus’s position, tweeted: “I now understand I must use my privileged position as a white middle-class gatekeeper with more awareness to promote diversity, equity, inclusivity, as all UK publishing strives to put right decades of structural inequality”. At the same time, one of the sensitivity reads Picador had commissioned was advising that Some Kids “exemplifies why publishing is so unwelcoming to marginalised people”. The Readers were explicit that this was not to do with any actual words on the page, but because they could tell that I personally had not done “the self-reflection and self-education that is necessary to understand the underlying reason that so many people felt harmed by [my] work”. They recommended the book should not be republished.

Before we could discuss this, Picador and I agreed to split. After the announcement was made, Swift Press offered to take on the book. I thought carefully about all the notes I had been given and, in the end, adopted none of the suggestions proffered by the Readers. The Kindle edition of Some Kids I Taught and What They Taught Me now on sale with Swift is the same one I originally sent to the sensitivity readers for report, and if you want to know more about what caused such deep offence, and seems to exemplify so much about publishing, I suggest you read it.