The first time I was truly terrified by a book — in the visceral, sleep-with-the-lights-on kind of way — was between my sophomore and junior years of college, when I was 20 years old. The book was
House of Leaves by Mark Z. Danielewski, which I bought more for its formal unorthodoxy than because I thought it would turn out to be the most frightening book I'd ever read. Not that I didn't appreciate the macabre — in high school I'd read a few dozen
Stephen King books, and I had a soft spot for Rhode Island's favorite misanthropic mythicist,
H. P. Lovecraft. But this was a time in my life when I briefly found literary theory — especially the sweaty palmed convolutions of Roland Barthes and Walter Benjamin — to be at least as interesting as literature itself, if clearly less enjoyable. (But of course its difficulty was part of the point.) And when I
was reading contemporary fiction, it was primarily something to which the term "post-modern" could conceivably be applied, which meant anything from Donald Barthelme to
David Foster Wallace to the late-career freak-outs of
Philip K. Dick. So when I cracked open
House of Leaves, I was more excited about the alternating fonts, the architecturally deployed blocks of text, the quasi-scholarly footnotes, than the pulp-ish plot.
But by the time Will Navidson decides to explore the labyrinth that has inexplicably opened up in the middle of his house on Ash Tree Lane — a labyrinth first hinted at by the sudden appearance of an extra closet and by the discrepancy of one-quarter of an inch between the inside and outside dimensions of the house — all my preoccupations with theory and form went out the window, mostly because I was too frightened to think straight. Back at school the following winter, though — and back in the thickets of theory — produced a 35-page paper that demonstrated how Danielewski's novel satisfies the precepts of a "writerly" literature as outlined by Barthes in his maddening S/Z, while also embodying the future potential of hypertext (ah, 2002) and employing to great effect Freud's theory of the unheimlich. (Only the last of these has all that much to do with the actual novel, and you certainly don't need to have read any Freud to be scared witless by the book.) My conclusion was that we were most frightened by Danielewski's work because we didn't know how to read it: what was truly unheimlich was the act of reading such a formally disruptive book in the first place. While this might not have been wrong, exactly, it was almost certainly missing the point.
Yet far worse than this enthusiastic over-reading was the fiction I was writing around this time. In S/Z, Barthes asks:
[W]hich texts would I consent to write (to re-write), to desire, to put forth as a force in this world of mine? What evaluation finds is precisely this value: what can be written (rewritten) today: the writerly. Why is the writerly our value? Because the goal of literary work (of literature as work) is to make the reader no longer a consumer, but a producer of the text.
This may sound desirable in the abstract, but the only works of fiction that seem to entirely satisfy these dictates are at best willfully obtuse (André Breton's Nadja, anyone?) and at worst simply incomprehensible — literary Rorschach tests to which the reader must bring his arsenal of critical weapons to even begin to parse, let alone enjoy.
My own contribution to all of this was the first stack of pages I presented to the advisor of my undergraduate senior fiction thesis; in these pages, which were allegedly the first chapter of a novel, a gang of artists cruise around a vaguely post-apocalyptic city in pick-up trucks, writing political slogans with Sharpies on the leaves of oddly abundant (for a post-apocalyptic city, anyway) trees. My perplexed advisor gently suggested another tack, and what I eventually turned in five months later was a collection of Less Than Zero-lite stories about New York City teenagers behaving badly, which wasn't much better, but which was at least more representative of the kinds of fiction I was just then starting to rediscover. After graduation, I gave up writing for over a year while working in a London publishing house, a vantage point from which the intellectual contortions of Barthes, et al. appeared spectacularly irrelevant.
By the time I began to write seriously again in my mid 20s, I had set aside the formally experimental aspects of Danielewski's book, but its essentially Gothic core had stuck with me: the malevolent, shape-shifting house, our elemental fear of the dark, and, yes, a pervasive sense of the unheimlich — the uncanny, the familiar made strange. I began to place other, more conventionally structured books beside it in my personal pantheon of suspense and dread: Shirley Jackson's We Have Always Lived in the Castle, Donna Tartt's The Secret History, a few of Stephen King's slimmer, more ruthless early novels (Carrie, especially). These, I thought, were the kinds of book I wanted to write, novels that invade the reader's brain like a nasty virus, not only through their depiction of disturbing events or images, but by making you see things from the point of view of a twisted — or, at the very least, morally questionable — mind. The Secret History is essentially Richard Papen's lengthy justification of his participation in a murder, and by the end you would be hard-pressed to declare that you would have acted otherwise. Carrie, too, attacks the reader's moral certainty: Carrie's high-school tormentors don't deserve to die, but they certainly deserve some punishment for how they treated her, don't they? By forcing you to identify with such characters, these books implicate you, the reader, which, at its strongest, is a disturbing experience.
Fiction can also play with the trustworthiness and very identity of its narrators, and the literature of the macabre — dating back at least to Stevenson's The Strange Case of Dr. Jekyll and Mr. Hyde — has used this tactic creatively and with much success. In my own novel, In This Way I Was Saved, I attempted to push this narrative ambiguity — both moral and literal — as far as I could by making the identity of the narrator, Daniel, one of the central mysteries of the novel. Is he an aspect of the troubled mind of the protagonist, Luke Nightingale, first in the form of an imaginary friend and later as a manifestation of mental illness? Or is he something else entirely, a supernatural embodiment of despair and suicide? Every piece of information we might use to answer these questions is tainted by the fact that it came from Daniel's own mouth and is thus filtered through his (or, perhaps, Luke's) motives and prejudices. I was trying to evoke an atmosphere of uncertainty, of unease punctuated by violence. I wanted to create a fictional world devoid of solid narrative ground in the most basic sense: Who is telling this story? This tension over Daniel's ontological status is never fully resolved, and the resulting ambiguity is key to my goal of freaking readers out.
Looking at things this way, maybe Barthes wasn't being so absurd after all; maybe his error, as with so many theorists, was the extremity of his argument, rather than its general instincts. In my book, as well as in other literary thrillers like Patrick McGrath's Spider and James Lasdun's The Horned Man, the reader must do a fair amount of work sorting through the narrator's motives and misdirections, as well as the essentially subjective terms of the story itself. None of these books are what anybody would call experimental or difficult; yet neither do they assume their reader to be Barthes's passive "consumer" of texts. But all of this is again mostly beside the point. The techniques I used probably matter more to me than to anybody reading my novel. If I make — by any means — just a few readers feel the way I did eight years ago while reading House of Leaves, then I will have succeeded.