The Danger — and Necessity — of Paranoid Reading
I first came across the term “paranoid readings” on Twitter, discussed by trans women writers about how their works are often received by other trans readers. To understand this particular example, we have to go back to a sci-fi story that took over Twitter (at least, a certain subset of Twitter) for a brief time: “I Sexually Identify as an Attack Helicopter” by Isabel Fall.
The story was a response to the transphobic meme of the same name, but whether it was a reclamation or further mockery was up for debate: the author was an unknown, and there was no definitive proof how she identified. The story was nominated for a Hugo award, but despite the critical acclaim it received, it was also dragged through the mud on Twitter by readers who assumed the worst intentions.
Criticism that had started out as constructive had transformed into accusing Fall of being a neo-Nazi, a right-wing troll, and more. Fall read the bad faith criticism compulsively: “I sought out and read everything written about the story. I couldn’t stop. It was like that old nightmare-fantasy. What if someone gave you a ledger of everything anyone’s ever said about you, anywhere? Who wouldn’t read it? I would read it; I would go straight to the worst things.”
Within the same month the story first was published in Clarkesworld, after unending criticism and harassment on Twitter, Fall asked for the magazine to unpublish the story and checked herself into a psychiatric ward for suicidal thoughts.
Clarkesworld replaced the story with an editorial note clarifying things, including, “Isabel was born in 1988. That does not make one a neo-Nazi. I’m honestly surprised and disappointed that I have to say that.” They explained the deletion of the story with: “The recent barrage of attacks on Isabel have taken a toll and I ask that even if you disagree with the decision, that you respect it. This is not censorship. She needed this to be done for her own personal safety and health.”
When she wrote the story, Isabel Fall was putting one foot out of the closet as a trans woman, not yet ready to be completely out in her life. Since the backlash, she has stopped using that name and withdrawn her stories on similar themes that had been in progress. She saw publishing this story as an “important test for myself, sort of a peer review of my own womanness. I think I tried to open a door and it was closed from the other side because I did not look the right shape to pass through it.”
Her first steps out of the closet were met with such intense, personal criticism, that she saw the endeavor as a failure: “Isabel was somebody I often wanted to be, but not someone I succeeded at being. I think the reaction to the story proves that I can’t be her, or shouldn’t be her, or at least won’t ever be her. Everyone knew I was a fraud, right away.”
The story of Isabel Fall is heartbreaking. It’s also illustrative of a larger phenomena: paranoid reading. This is a term coined by Eve Kosofsky Sedgwick in an excellently named 1997 essay, “Paranoid Reading and Reparative Reading, Or, You’re So Paranoid, You Probably Think This Introduction is About You” in Touching Feeling. Fittingly, Sedgwick is also discussing queer theory and queer criticism.
Essentially, paranoid reading approaches a work — whether it’s a story, tweet, TV show, photograph, or anything else that can be critiqued — from a defensive position. It anticipates bad actors and maliciousness, seeking out clues for them. Reparative reading, on the other hand, searches for the positive in even a deeply flawed work. It seeks pleasure instead of avoiding pain. It recognizes where the text might benefit some viewers/readers, even if it’s not personally pleasurable.
If you’re on Twitter, paranoid reading will sound immediately familiar. It’s the go-to strategy for approaching art on the platform, if only because outrage gets clicks. A declaration that a beloved TV show is, in fact, everything wrong with the world will get a lot more attention than a tweet arguing that while flawed, this movie brings up some interesting ideas.
Isabel Fall’s story fell into this cycle. Nuanced criticism, as well as feedback from trans readers who found value in the story, fell by the wayside, while paranoid readings were amplified. Their reach grew by being boosted by cis readers who felt they were supporting trans people by spreading the word. This Twitter algorithm leaves little room for nuance or legitimate disagreements — the same story can be positive representation to one person and damaging to another reader.
It’s tempting to be dismissive and angry at these paranoid readings: “They’re looking for things to be angry about!” But that sentiment seems remarkably similar to right-wing complaints about leftist “cancel culture”: that queer and trans people, people of colour, disabled people are creating problems where there are none. It’s a strawman feminist stereotype.
What’s missing, too, from this dismissal, is where this impulse is coming from. There is a very good reason that trans readers are approaching stories from a skeptical perspective, and that’s because historically, the vast majority of trans representation has been malicious. Trans or gender-nonconforming characters have been the villains or the butt of the joke, whether in centuries old plays or modern TV.
Kai Cheng Thom, author of I Hope We Choose Love: A Trans Girl’s Notes from the End of the World, both acknowledges the pain, trauma, and fear of trans audiences while also questioning whether this impulse to punish and banish is really useful. She has received similar paranoid reading of her work, and she shares that most trans writers she knows have. She asks, “What would have happened if these angry folks had simply asked me genuine questions about what I was doing with the piece, rather than assuming some malicious intent?”
It is completely understandable that a group that is often faced with violence, harassment, and harmful representation would assume bad intent and be on the look out for damaging elements of a work. It’s a survival strategy.
At the same time, this impulse becomes weaponized against other trans creators. It’s also limiting: paranoid reading is deletive. It can only take away. It can easily lead to restrictive standards that only allow a certain kind of art to be acceptable, one that checks a series of prescribed boxes and doesn’t include any messiness, grey areas, or uncomfortable complexities. An example Sedgwick uses is camp, which is can be a source of queer joy and creativity, but is misrecognized by paranoid reading. Reclamation, drag, and many other forms of queer expression can be misread or limited by a paranoid reading being applied.
Is All Criticism Paranoid?
Now that we’ve seen paranoid reading in action, I want to return to Sedgwick’s original essay for context. First of all, she is not dismissing critique in general. We’re allowed to disagree and discuss texts. Instead, Sedgwick argues that criticism has become synonymous with paranoid reading — that we can’t imagine other strategies of criticism: “In a world where no one need be delusional to find evidence of systemic oppression, to theorize out of anything but a paranoid critical stance has come to seem naive, pious, or complaisant.”
Sedgwick isn’t as much condemning paranoid reading as she is offering an alternative, because the practice has subsumed all other strategies. She also isn’t saying that paranoid reading reaches incorrect conclusions: “I am saying that the main reasons for questioning paranoid practices are other than the possibility that their suspicions can be delusional or simply wrong…They represent a way, among other ways, of seeking, finding, and organizing knowledge. Paranoia knows some things well and others poorly.” She compares a lack of diversity of critical approaches to a “shallow gene pool” that can’t react effectively to change.
A guiding principle of paranoid reading is that it is anticipatory: “The first imperative of paranoia is There must be no bad surprises…” We can see that in play in the reaction to Isabel Fall: as long as there is a chance that the story could later be revealed to be a cruel joke, then it should be recognized as one now. In a paranoid reading, if malicious intent or prejudice cannot be ruled out, then it must be assumed.
Because criticism as a field is rooted in paranoid reading, Sedgwick argues, it is dismissive of reparative reading, seeing it as “merely aesthetic” or “merely reformist,” as if amplifying pain is the only way to effect change. Reparative reading doesn’t mean denying systems of oppression. It seeks pleasure because it recognizes that the environment is hostile and will not provide that pleasure itself. It’s also empathetic, recognizing others as worthy of love and care — which is why reparative reading can see where a text can be useful to other people even if it’s not personally. For example, there are many queer books I’ve read that I found stereotypical or simplified, but I can recognize that other readers may find the same story affirming.
Sedgwick is most interested in what knowledge does. The essay begins with a discussion of HIV and whether it was possibly government manufactured, a debate at the time. Would the answer really change anything? She quotes activist Cindy Patton, on the conspiracy theory that HIV was created by the U.S. military:
Sedgwick argues that believing this knowledge or proof of this would change anything is naive: “What marks the paranoid impulse…[is] the seeming faith in exposure…as though to make something visible as a problem were, if not a mere hop, skip, and jump away from getting it solved, at least self-evidently a step in that direction.” Despite being suspicious of all motives, the position of paranoia is trusting that a “fully initiated listener” would be moved, that all they need is to get the story out there. It seeks to reveal hidden violence, but it doesn’t offer a lot to handle explicit violence — the kind of violence that is bragged about in political speeches or regularly broadcast on the news. When the violence has already been “unveiled,” there doesn’t appear to be a next step.
Another approach is to concentrate on the impact of this knowledge and use your resources accordingly to effect change. “To know that the origin of HIV realistically might have resulted from a state-assisted conspiracy — such knowledge is, it turns out, separable from the questions of whether the energies of a given AIDS activist intellectual or group might be best used in the tracing and exposure of such a possible plot. They might, but then again, they might not.”
While paranoid reading is a “strong theory” that can be broadly applied, reparative reading is a “weak theory”: it is only narrowly relevant. While paranoid reading is concerned with the inevitable, reparative reading can “can attune it exquisitely to a heartbeat of contingency.” My reading of a text may not be applicable to all texts or to all readers. Criticism in this strategy can be specific to a particular circumstance without having to be a rule for all contexts.
(I have to admit at this point that I am not claiming a complete understanding of Sedgwick’s ideas. Queer theory is a notoriously complex and slippery field, and I’m content to just grasp at the ideas that stand out and let sentences like “Tomkins does not suggest that there is no metalevel of reflection in Freud’s theory, but that affect itself, ordinary affect, while irreducible corporeal, is also centrally shaped, through the feedback process, by its access to just such theoretical metalevels” drift past, incomprehensible to me.)
Sedgwick argues that this reparative reading has been a strategy of queer and other marginalized groups for millennia. I am reminded of lesbians in the 1950s and ’60s hiding lesbian pulp fiction under their mattresses, skipping over the homophobia to marvel at the idea that women like them existed somewhere out there (Greenwich Village, presumably) and that lesbian relationships were a possibility.
“What we can best learn from such practices are, perhaps, the many ways selves and communities succeed in extracting sustenance from the objects of a culture — even of a culture whose avowed desire has often been not to sustain them.”
It also brings to mind Amal El-Mohtar’s brilliant essay in Queers Dig Time Lords, which has a passage that has stuck with me in the years since reading it:
The Necessity of Paranoid Reading
In his essays “In Defense of Paranoia” (Part 1 and Part 2), Stephen Ira points out how Sedgwick’s concept has been flattened into a criticism lobbed at “purity teens”: young queer and trans people looking for the one true, unproblematic path. These queer kids today and their paranoid readings are ruining everything, policing queerness and gender into a new restrictive taxonomy!
But Sedgwick was never arguing that paranoid readings are bad and should be eliminated, while reparative reading is the only right way. Instead, Sedgwick describes “mutual inscription” between paranoid and nonparanoid reading: “For example, Patton’s calm response…drew on a lot of research, her own and other people’s, much of which required being paranoiacally structured.” Paranoid reading can provide a base to build on.
To slip into cliche, it’s also worth reiterating that paranoia is a natural response to a hostile environment: It’s not paranoia if they really are out to get you. Ira explains that, “as a transsexual living under neoliberal homonationalism, I need my paranoid reading skills at hand when I encounter cops, doctors, and other types of imperial myrmidon.” There are many situations where a paranoid lens is more useful than a reparative one, especially when your immediate safety is at stake. Ira continues, “No, it is as Sedgwick said: we must keep all of our tools to hand, both the paranoid and the reparative.”
Ira also points out how reparative reading can be damaging. It can bolster support for a harmful narrative, smoothing over necessary critique. We also are much more likely to apply a reparative reading to our own actions, as is powerfully explored in Ira’s example of a non-apology following the Isabel Fall story. Similar to the fundamental attribution error, it’s very easy to apply a paranoid reading to others and a reparative reading to ourselves, excusing our own harmful actions while combing through others’ tweets for anything that could be construed as problematic.
The takeaway, though, is not that either reparative or paranoid reading are the right kind of criticism. Instead, they are most effective when used in concert. Reparative reading may depend on a foundation of paranoid reading, while also exploring possibilities that would otherwise be unseen. Sedgwick acknowledges that most people will use both in different contexts, seeing the two as “practices, not as theoretical ideologies (and certainly not as stable personality types of critics), but as changing and heterogeneous relational stances.”
While paranoid reading is a specific term coined by Sedgwick, I find that this concept intersects with a lot of conversations we’ve been having recently online, which I think is why it has come up. The idea of having to fit an ideologically pure image to exist online has become more and more entrenched, and any infraction can cause a disproportionate backlash — the more marginalized you are, the most dangerous and angry this reaction seems to become.
The Helicopter Story discourse also brings up questions of the intracommunity discussions and what happens when allies or outsiders begin to join in. Cis people boosted the criticism of Fall’s story because they believes themselves to be allies helping the cause. It’s incredibly difficult to have nuanced intracommunity discussions online, where ideas are flattened and widely disseminated — because simple, outrage-inducing tweets and headlines spread the farthest and the fastest.
Another idea related to paranoid reading is hypervigilance. Hypervigilance is a response to trauma, often a symptom of PTSD, that causes a heightened alertness to possible threats. This sounds very similar to Sedgwick describing paranoid reading as a “terrible alertness” to danger. In marginalized communities where trauma is common, it’s not surprising to see these reactions play out.
While writing this post, I watched Princess Weekes’s video about criticism from within Black and queer communities. Although she does not use the term “paranoid reading,” she does talk about “creators of color feeling stifled by the hyper criticism they can receive from their communities,” which is very relevant to this discussion, and I highly recommend watching it.
Bonus: My Own Paranoid and Reparative Reading
While reading Sedgwick’s article, I was reminded of the two sides of a debate that’s been chasing itself around my brain for years. It’s about my favourite fictional character of all time, Tara Maclay from Buffy the Vampire Slayer. (I know, I know, this is Book Riot and I’m admitting my favourite character is from TV. I’m sorry.) All critiques of the show and its creator aside (and there are many legitimate ones), I watched it at a formative time in my life, and Tara was the first lesbian character on TV I could really relate to. And then came season 6. (Spoilers for season 6 of Buffy.)
I already knew Tara would die. I came to the series late, and that was one of the few spoilers I had stumbled on, but I was glad for the warning. As the sapphic fanbase all agreed, it was an egregious example of the bury your gays trope. Here was a character so beloved by queer fans that the actor had been sent countless letters about how the character had saved their life, had given them the courage to come out to their parents, had made them accept their sexuality. And she had been unceremoniously killed by a stray bullet, immediately followed by turning her girlfriend into a villain — the “Psycho Lesbian” trope, which was also well trod even then. It was a terrible, thoughtless bit of writing that was a slap in the face to queer fans, and I hated it. Except…
The thing is: some part of me also appreciates Dark Willow, and not just for her campy ridiculousness. I had fallen in love with Tara. As an 18-year-old, watching Buffy in bed on my laptop with my comforter pulled around my shoulders, I appreciated the character so much. I had just started university. Tara was in university! Tara was my role model: this kind, funny, fiercely loyal lesbian who was accepted into the Scoobies and had (originally) the most healthy romantic relationship in the show. When she died, I could relate to Willow. There was something cathartic and moving about Willow’s love for her girlfriend being so powerful that she wanted to burn the whole world down. I could relate to that. (Little did I know that my new queer college relationship would soon burn my own world down, and I would stand back and watch the flames.)
So those are my paranoid and reparative readings of Tara’s death, caught in a loop, and I hold both of them to be true. I think it was a damaging depiction — and I can find something affirming in it, a confirmation of the power of queer love and the value of this lesbian character. I don’t think they negate each other.
There’s no easy answer here of how to approach criticism, because there are too many factors involved, too many specific circumstances to weigh. That’s exactly why Sedgwick advocated for having a variety of tools at are disposable, including paranoid reading and reparative reading.
When you find yourself joining a Twitter pile-on against someone, consider extending a reparative lens, at least to try it. But also keep in mind that dismissing paranoid reading out of hand is also unproductive. We need a balance of approaches, and room for conversations to include both playing off of each other. Otherwise, we risk missing valid critique as well as chasing creators off the internet for exploring nuance.
It’s easy to forget that the social internet is still very new. We haven’t figured it out yet, and we’re bound to make mistakes as we experiment. But we need to find ways to have productive discussions that move collective understanding forward, or we’ll continue to be stuck in this cycle of outrage, punishment, and banishment. I’m not sure what the best strategy will be, though it will likely involve both paranoid and reparative reading. One thing I do know is that we can do better than this.