Jeeves and the Sensitivity Reader
I had this published in Café Américain a little while back. It’s paywalled, so I thought I’d share the pre-publication version here—the one I was actually most happy with—with you, my Substack followers.
If you enjoy it and would like to support more writing like this, I have a Buy Me a Coffee account where readers can make small, one-off donations. Every contribution helps cover the hidden (and often frustrating) costs of independent writing—subscriptions, archives, firewalled journal articles, and so on.
Jeeves and the Sensitivity Reader
The recent rise of sensitivity readers in publishing raises two fundamental questions: first, what kind of ‘sensitivity’? And second, what kind of ‘reader’?
What is now clear is that the sensitivity in question isn’t to tone, rhythm, or narrative structure. It is not, in other words, literary sensitivity. Rather, it is sensitivity to feelings, and specifically to those associated with certain, pre-identified groups, whose day-to-day life is captured under the heading of that great incontestable of the woke age: “lived experience”. From there, everything is filtered through a script of systemic oppression.
But surely these sensitive “sensitivity readers” are at least trained in literary analysis? Not quite. In most cases, what qualifies them to advise authors on how that identity should be represented is simply their status as a purported representative of a particular group or experience. It’s not difficult to see how this identitarian solipsism might become a problem when editorial decisions demand a nuanced understanding of an author’s lexical, syntactic and rhetorical choices.
Typically, sensitivity readers are brought in after structural edits and copyedits, combing through the manuscript for any element of character, plot or narrative development that might cause offence. Sometimes this means advising the removal of individual words, sometimes entire scenes, character traits, or narrative arcs.
Underlying this nascent profession is a kind of naïve, pre-Wittgensteinian correspondence theory of style, in which literary representations are judged “true” or “adequate” only insofar as they match some presumed external reality. In other words: does the character “ring true” to the reader’s idea of reality? If, as Thomas Aquinas once wrote, “truth is the adequation of things and intellect”, then the sensitivity reader becomes the sovereign guarantor of that adequation.
And yet those of us living in the post-positivist era know things aren’t so simple. If literary meaning is a function of use, not mere correspondence, then style is embedded in genre, context, idiom, and can’t be verified by “lived experience”.
The confusion now runs so deep that the industry’s obsession with authenticity has begun producing editors who genuinely believe men shouldn’t write female characters, because they lack the experience required to capture the essential “truth” of that identity.
Writing in The Spectator, Zoë Dubno laments this “obsession with reading for verisimilitude in literature” and questions whether the experience of one individual can ever represent the collective experience of all:
I could just imagine a female sensitivity reader marking up the proofs of Anna Karenina and writing: “It’s not believable that a woman would jump in front of a train because she thinks her boyfriend cheated on her.” Would they then have to find a woman who had jumped out in front of a train?
But even if they could, what about the complexities of the text, the traces of authorial skill one finds therein? What about the layers of narration and voice—author, implied author, narrator—constructed in and through the narrative itself? Even if that hypothetical sensitivity reader were found—or rather, exhumed—and even if she judged that particular characterisation to be “wrong”, wouldn’t that be to confuse authorial intent with character construction?
What if the real author is deliberately throwing their voice, creating a character filtered through a particular kind of narrator (reliable, unreliable), from a certain point of view (internal or external focalization), and embedded within a specific narratorial position (extradiegetic, diegetic, story-level) such that it allows us to explore what is pernicious, or grimly inevitable, or tragically plausible about that version of “a woman jumping in front of a train”? What if that character portrayal, however flawed the sensitivity reader may believe it to be, is intended to provoke deeper reflection in the reader?
Publishers and practitioners like to insist that sensitivity reads are “optional”. But their own rhetoric makes clear that declining one marks you out as morally suspect.
Witness Swedish media company Bonnier, for instance, suggesting that while “authorial voice should be respected”, the company “believe sensitivity reads can play an important role in inclusive, forward-thinking publishing”, or Usborne insisting that all non-fiction titles be sent to “subject-matter experts” to avoid reinforcing “harmful stereotypes”.
This approach is troubling enough when applied to the living. But what of those authors who are no longer here to explain, to clarify… to resist?
Consider the case of author P. G. Wodehouse, who recently had his own run-in with the sensitivity police. Published in 1934, Thank you Jeeves is one of Wodehouse’s classic Jeeves and Wooster stories. The plot centres on Bertie’s atonal misadventures with a banjolele, a temporary estrangement from Jeeves, and a string of quintessentially Wodehouseian entanglements involving mistaken identity, impecunious aristocrats, and romantic confusion. In 2023, Penguin Random House added the following disclaimer to its new edition of the novel::
Please be aware that this book was published in the 1930s and contains language, themes and characterizations which you may find outdated. In the present edition we have sought to edit, minimally, words that we regard as unacceptable to present-day readers.
Reactions from the pro-free speech side were muted and largely missed the literary significance of what was happening.
One reviewer conceded that Wodehouse’s language “jars” in places, with “patronising racial undertones”, but was “inevitably of its time and class”.
Another critic played the character referee, reassuring us that “Bertie uses the [n-]word cheerfully… apparently oblivious of its oncoming capacity to offend”. It wasn’t intended to denigrate, he argued— “at least not to the extent that it would be today.” As to whether it matters, he concludes: “I doubt the n-word will be missed. The removal was conducted with the approval of Wodehouse’s estate, and I expect the stitches have been removed by now and the scar barely visible”.
But it does matter, and the scars aren’t healing. Consider the text—something, lest we forget, that almost no one in this debate has done.
[Bertie] “Of course, yes. With nigger minstrel entertainment to follow... they are coming all right, I take it?”
[Jeeves] “Yes, sir. The Negroes will be present.”
[Bertie] “I wonder if there would be any chance of a word with the one who plays the banjo. There are certain points in his execution I would like to consult him about.”
[Jeeves] “No doubt it could be arranged, sir.”
In the revised edition, Bertie’s “nigger” is excised, and he and Jeeves end up speaking in the same register. But what seems like a minor substitution is, in fact, a collapse of the scene’s literary architecture.
When Thank You, Jeeves was published in 1933, Negro was the overtly respectable, formal term in British English—akin to using “whom” instead of “who”. The slur, by contrast, was common in certain upper-class colloquial registers. Bertie’s use of it indexes oblivious privilege. Speaking in a breezy, in-group register typical of the Drones Club, he’s more concerned with patter than propriety.
Jeeves’s “Negro”, by contrast, signals restraint, precision, social awareness. He’s the cultivated valet, unwilling to echo his employer’s coarseness. Wodehouse engineers this contrast, quite deliberately, for comic effect.
At this stage, a sensitivity reader might protest that none of this exonerates Wodehouse for including the slur in the first place. But the more technically interesting question is this: what does it mean to speak of “Wodehouse” in the context of his literature?
It’s an important question, because the scene unfolds across at least three distinct diegetic levels:
1. the extradiegetic level (Wodehouse himself, the real author, who exists outside the fiction);
2. the diegetic level (Bertie, the narrator, telling the story from within the fictional world);
3. the intradiegetic or story level (Bertie and Jeeves in conversation, as characters speaking within the scene).
Bertie the character says “nigger”, while Bertie the first-person narrator recounts it, entirely unabashed.
In other words, Wodehouse isn’t “there” in the text. By the time the slur is uttered, we’re already two layers removed from his authorial voice. And no sooner has Bertie spoken than Jeeves responds, pointedly and precisely, with the more formal, socially acceptable term of the time.
That’s the joke. Not just that Bertie was gently corrected, but that’s he so oblivious to the fact of the correction—so unaware he’s just been socially one-upped by a “bally valet” as he might put it—that he narrates it himself.
In that moment, the reader is drawn closer to Jeeves, watching from a slight distance, laughing knowingly at Bertie’s “wonderful innocence”. It’s a perfectly constructed example of what Wayne Booth once called the “secret communion” between author and reader behind the narrator’s back: irony not declared, but staged.
This is precisely what’s lost when the sensitivity reader shunts Wodehouse aside at the extradiegetic level. It isn’t just one word that changes. With both characters now speaking in the same voice and register, the delicate comic machinery collapses under the weight of blundering editorial do-goodery.
Some might argue that, contrary to Penguin’s protestations, this isn’t a “minimal” change, but a substantial infringement of an author’s posthumous right to freedom of expression.
Much of the debate around sensitivity readers has grown stagnant, caught in a weary loop of moral accusation and defensive outrage. Too often, denunciations of “woke madness” act as a proxy for critique, allowing us to avoid the deeper challenge of moving from knowing that something feels wrong to understanding how it is happening.
That’s a serious problem, given that much of what is dismissed as “woke madness” —morality clauses, literary sanitisation, trigger warnings—rests on well-developed philosophical frameworks now ensconced within the machinery of publishing, education, and cultural policy.
The progressive left has been doing its homework. It read poststructuralism, reworked postmodernism, absorbed deconstruction. It built theories of harm, trauma, and performativity. Then it operationalized them, proceduralized them, embedded them in HR policies.
Nor should we forget the army of low-paid yet dedicated foot soldiers it can call upon—thousands of them, all working on the granular details: line by line, page by page, amendment by amendment, clause by clause. Editors, consultants, policy officers, diversity auditors, cultural gatekeepers. It’s “only” about “reviving classics”, “modernizing texts”, “preserving readability”, they insist, our accusations of “woke madness” crashing impotently against these rhetorical fortifications, never to breach the inner sanctum where the real social engineering takes place.
There’s an irony in the fact that although the “woke” revolution has at its most fundamental level taken place in language, those who profess to care about freedom of speech, expression and belief rarely acknowledge its structural possibilities. Derrida is regularly lampooned on the right for his claim that “there is nothing outside the text.” Yet the aphorism, properly understood, contains an insight as destabilising to conservative certainties as to progressive ones: that while language may not be all there is in the world, it is nonetheless all there is in the world for representing reality to ourselves and others.
And what of the classical liberals, the small-c conservatives, the libertarians? Where is the counterweight to Butler’s historicised illocutionary model, Bourdieu’s symbolic violence, Deleuze’s “machinic” refusal of the “postulates of linguistics”? Where is the reckoning with Foucault’s claim that discourse doesn’t just reflect power, but is power? When have the defenders of free speech ever engaged seriously with Gramsci’s theory of the “war of position”, or his call to “read against the grain” —not politically, but interpretively?
It’s true that many contrarian and dissenting voices have been pushed out of the academy. And yes, there are exceptions: serious thinkers from a range of traditions who’ve challenged the orthodoxies of our time. But beyond the university—among the telegenic renegades, YouTubers, shockjocks, journalists, ex-academic pundits and maverick politicians who have done so much to monetise a certain theatrical apoplexy at “the absolute state of things!”—anything resembling a sustained, coherent intellectual response to the philosophical foundations of “woke” has been hard to find.
The counter-revolutionary “long march through the institutions”, if it ever materialises, will be futile unless armed with analytical tools sharp enough to cut through ideas that, like thin coils of narcissistic affront, have been creeping up from the intellectual basement of our culture for so long that they are in danger of taking root in our minds. Not the catchy slogans, Instagrammable rants and hyperbolic punditry so artfully curated for comfortable digestion by the cultural machine, but the quiet, fine-grained, poorly remunerated task of analyzing language-in-use.