I’m reading Mark Z. Danielewski’s House of Leaves, which is on one level a horror book. I don’t read many horror stories: while I do, like most humans, enjoy the occasional self-induced frisson, and I have a fondness for ghost stories, I can imagine all manner of phantasms into the darkness without professional help. In short, they tend to stick with me for too long (estimated time from viewing The Ring before could stare into darkness without eyes resolving nonsense lights into you-know-who: 10 months.). This book came highly recommended by Ryan, and threatened, via said Ryan, to transmute my love for Poe’s Haunted into fear. Of course I wanted to prove him wrong on that point. On top of these inducements, Ryan told me a recent xkcd strip would be hilarious when and only when I had read House of Leaves.

I’m a third of the way through the physical volume, a “remastered full-color edition” snagged from Ryan’s shelves, which means I’m almost halfway through the body of the novel (there are copious appendices). It is legitimately creepifying, and its typographical oddities do work, for me, to physicalize the wandering in the story and depict an erosion of the rational and solid. These contrivances are part of why the book’s considered experimental and postmodern. But one device it uses is not in itself postmodern, and has been used to great effect in the horror genre over at least a century.

I speak, of course, of framing. The frame story is a very old device indeed (at least some versions of the Ramayana are framed, and that’s around 2,500 years old), but I’m thinking especially of its use in Henry James’s The Turn of the Screw. In that classic novella, there are two frames. The ostensible narrator that begins the book (Frame 1) has a friend who introduces and reads a manuscript to him (Frame 2) that has been written by a woman once governess to the friend’s younger sister (the story itself). Much has been made of these nested frames, which put layers of hearsay and distance between the reader and the governess. If The Turn of the Screw were presented without these introductions, the narrator might not be seen as unreliable (though in the 19th century a governess, an unmarried woman in an interstitial social class, was fairly unreliable to begin with) and the reader might take at face value the ghost story she relates. The frames remove some of our intimacy with this primary narrator and help to create the piece’s legendary ambiguity.

Other uses of frames in horror include the very effective stage play of The Woman in Black by Stephen Mallatratt, where the audience is invited to identify the artifice of the performance with the frame and thereby believe the central story to be real, and found-story narratives like The Blair Witch Project, where the story itself (a manuscript or video) is found as a mysterious artifact after the characters in it have disappeared or died. As I said, I’m not terribly well-versed in horror, so I may be overlooking other striking examples.

House of Leaves has a number of frames I have not yet counted. Let’s count them together! (Ah-ha-ha!) The book begins with a Foreword from “the Editors” who make “every effort” to translate and attribute, et cetera, and make references to previous editions, in what appears to be Bookman Old Style. Given that this is a novel, not a found manuscript as the text maintains, it’s safe to assume this is a narrow little frame, Frame 1. Then we have the Courier stylings of manuscript-finder, anecdote-teller and annotater Johnny Truant, Frame 2. Then we have the main body of the manuscript, a discursive and highly footnoted piece of eccentric criticism by Zampanò (Frame 3) on a documentary. This documentary, The Navidson Record, does not seem to exist as far as Truant can discover, nor do the endless articles of criticism on the work Zampanò references. The documentary is the main story. Although arguments could probably be made that it’s not, they would, themselves, get pretty damn academic pretty fast, which would probably please the spirit of the book. The documentary film, since this is a book, is unseen. There’s an empty center at the middle of the frames. It’s interesting, since Ryan told me that this book actually shows you what drove a character mad (unusual) — but the central story remains unseen, or half-seen, translated, transcribed, half-captured.

These frames within frames do remind me of Turn of the Screw (which is mentioned, and its connection to the Navidson Record rejected along with all the classic ghost stories of Western Literature in footnote 167), but there’s more at work here. The book is, despite the nebulosity of the term, postmodern. Here’s the definition of postmodern from the American Heritage Dictionary (as I said, it’s a nebulous term, and this is only one opinion:)

Of or relating to art, architecture, or literature that reacts against earlier modernist principles, as by reintroducing traditional or classical elements of style or by carrying modernist styles or practices to extremes

“Extreme” is a good word for this book. The use of three frames (James’s two is already pretty unusual), the way the frames do not just stand aside once the central story is introduced but continue to interfere and interact with it - which may make the term ‘frame’ less literally applicable° - the multiple appendices, and the crazy typography are all pretty out there. In addition, the classic horror trope I mentioned, the found-story, recurs. Within The Navidson Record, some footage is found. The Navidson Record itself may not qualify, since it’s un-found: Truant can’t find it, even though Zampanò and all his perhaps fictive academics have access to it. Zampanò‘s manuscript is, again, found. And while Truant is very much present in the earthy footnotes to it, in footnote 197 if not before SPOILER: it becomes clear Johnny has disappeared. As I’ve said, framing is older than Modernism, but he’s certainly applying it to extremes here.

Without a doubt, the frames do allow Danielewski to cast doubt on his narrator(s). But it seems like the interplay of doubt and faith in a narrator is one of the things he’s playing with: the way Zampanò‘s academic pretensions seem to be a way of bolstering his authority, a plea to convince the reader (and, I think, perhaps the universe) to believe him; the way you find yourself attentively reading Zampanò’s arguments for the authenticity of The Navidson Record even when you ‘know’ that in the universe of the book – assuming the universe of Truant and Zampanò to be the same – the film doesn’t exist, hoax or not! Who is reliable? What’s truth anyway?

Tolkien condemned dream-frames, in his essay “On Fairy Stories”, as deprecating the fantastic nature of a framed story. A realistic frame can trick a skeptical reader into accepting a non-realistic story. In horror, there’s another reason for a realistic frame — it’s a reality the reader/watcher accepts, which makes it easier to suspend disbelief and enter upon the fantasy of the inner story. That suspension of disbelief is necessary in order to give the audience the fear they came for. In House of Leaves, the nesting of frames lets the House exist several removes from the world of the reader, the world the reader accepts natively as real and true. But at the same time, the frames jostle and shift against each other, making the reader militate for the existence of the horror, making the reader want to believe the film exists, the House exists, because it’s interesting, and because it’s scary. A found-story frame allows the audience to see the story-record itself as its own evidence, its own artifact, even though they know the frame is fiction as well. In House of Leaves, the reader is even more complicit in scaring herself. Reading this book is tricky, it’s work. I am not only scaring myself, I’m working hard to do so. How much more complicit can I get?

° Update, 10/4/08: in footnote 308, Zampanò calls them ‘generations’:

Here in particular, he mockingly emphasizes the fallen nature of any history by purposefully concocting an absurd number of generations. Consider: 1. [event] —> 2. Navidson’s perception of [event] —> 3. Navidson’s description of [event] to Reston —> 4. Reston’s re-telling of Navidson’s description based on Navidson’s recollection and perception of [event].