Why Epistolary Novels Are a Computer-Science Problem
An epistolary novel, for those who don’t know, is a novel told through documents written by the characters, usually letters, but sometimes also journal entries, newspaper articles, emails, postcards, notes passed in class, and whatever else the author might think up. It’s an unusual and immersive storytelling format. It’s also a challenging format with its own unique difficulties that don’t apply to traditionally narrated books. Not just the obvious difficulty of creating a plausible scenario for why the characters would be sending each other book-length accounts of their activities, but the subtler difficulty of creating something that functions both as an engaging novel and a plausible letter.
It turns out that novels and letters are fundamentally different forms of writing. Letters don’t sound like novels. How are they different? That’s where computer science comes into it.
There’s a concept in computing called endianness. Jonathan Swift fans will recognize the name but not the concept. Endianness refers to the order in which data is stored or transmitted: Either biggest value first (big-endian) or smallest value first (little-endian). For instance, if a computer were transmitting the number 123, a big-endian system would transmit the digits in the order 1-2-3, while a little-endian system would transmit them in the order 3-2-1.
This concept can be readily extrapolated from computer communication to human communication. For instance, American dates (month-day) are big-endian, while European dates (day-month) are little-endian. (American full dates, month-day-year, are middle-endian, which I think we can all agree makes no sense at all.) Languages can also have endianness: in Russian, for instance, you tend to put the most important information at the end of the sentence, making it a little-endian language. Is everyone lost yet? Great, back to epistolaries.
Now we see the difference between novels and letters: Endianness. Novels tend to be little-endian, gradually revealing information over the course of the story and saving the biggest reveals for last. Mysteries are the most obvious example: You begin the story with few facts about the murder and gradually learn more and more, culminating in the big reveal of the murderer’s identity. This makes for an engaging read, keeping the reader curious to find out the next bit of information.
Letters, on the other hand, are big-endian. Since they’re primarily meant to convey information, rather than to entertain, after the initial pleasantries they usually present the key information first and any elaborating details afterwards. An invitation might read “You’re invited to my birthday party. It’s at my place on June 7 starting at 5 PM. Please don’t bring a present.” If the sentences were in the other order, the invitation would be incomprehensible.
This presents a difficulty when combining the two formats. Nearly any narrator would logically follow the big-endian convention when writing a letter, but it wouldn’t make for a very interesting story if the hero began the account of zir final showdown with the villain by saying “I fought so-and-so and I won” and only then relating the details. (Writers of non-epistolary novels aren’t totally free of this problem: Spoken dialogue is also big-endian. Many novels feature the kind of circuitous dialogue that would have the listener smacking the speaker upside the head and telling zir to get to the point in real life.)
There aren’t many solutions to this problem. Elizabeth Wein found a clever answer in Code Name Verity: A narrator who doesn’t actually want to communicate with the person she’s writing for, and who thus has a good reason to put off mentioning the important information as long as possible. This strategy works beautifully, but unfortunately isn’t generally applicable; it doesn’t even apply to the second half of Code Name Verity. I sidestepped the issue in the laziest way possible by writing a half-epistolary and using traditional narration for the climax. But the most common solution is to simply dispense with realism and let your narrator adopt a more novel-like voice as the story progresses. Practical, but not entirely satisfying.
The lack of well-established strategies for narrating epistolary novels is probably simply a factor of the lack of modern epistolaries, lost in a tide of changing communications. Still, epistolaries are an immensely enjoyable format, and I hope there will continue to be authors who rise to the challenge.