Q: What constraints dictate the length of works of fiction?
A: Same as any other product: money and time …
The most familiar form of fiction in the English-language publishing world, today, is a stand-alone bound book containing a novel. (Perhaps the second most familiar form is the series novel, which recycles characters of a setting from earlier works, optionally continuing to unfold a multi-book story or hitting a reset button between novels, as with some TV serials.)
A typical modern novel is in the range 85,000-140,000 words. But there’s nothing inevitable about this. The shortest work of fiction I ever wrote and sold was seven words long; the longest was 196,000 words. I’ve written plenty of short stories, in the 3000-8000 word range, novelettes (8000-18,000 words), and novellas (20,000-45,000 words). (Anything longer than a novella is a “short novel” and deeply unfashionable these days, at least in adult genre fiction, which seems to be sold by the kilogram.)
One would think that it’s so much easier to write a 5000 word short story (it can sometimes be done in a day) than a novel (it can sometimes take years) that they should be commoner. But trade fiction authors who write for a living seem to focus exclusively on novels, to the point where some of us don’t write short fiction at all. Why is this? Stay with me below the cut and I’ll try to give you a [highly subjective, personal, biased] explanation.
Genre science fiction in the US literary tradition has its roots in the era of the pulp magazines, from roughly 1920 to roughly 1955. (The British SF/F field evolved similarly, so I’m going to use the US field as my reference point.) These were the main supply of mass-market fiction to the general public in the days before television, when reading a short story was a viable form of mass entertainment, and consequently there was a relatively fertile market for short fiction up to novella length. In addition, many of these magazines serialized novels: it was as serials that Isaac Asimov’s “Foundation” and E. E. “Doc” Smith’s “The Skylark of Space” were originally published, among others.
For a while, during this period, it was possible to earn a living (not a very good living) churning out pulp fiction in short formats. It’s how Robert Heinlein supplemented his navy pension in the 1930s; it’s how many of the later-great authors first gained their audiences. But it was never a good living, and in the 1950s the bottom fell out of the pulp market—the distribution channel itself largely dried up and blew away, a victim of structural inefficiencies and competition from other entertainment media. The number of SF titles on sale crashed, and the number of copies each sold also crashed. Luckily for the writers a new medium was emerging: the mass market paperback, distributed via the same wholesale channel as the pulp magazines and sold through supermarkets and drugstore wire-racks. These paperbacks were typically short by modern standards: in some cases they provided a market for novellas (25,000 words and up—Ace Doubles consisted of two novellas, printed and bound back-to-back and upside-down relative to one another, making a single book).
The market for short fiction gradually recovered somewhat. In addition to the surviving SF magazines (now repackaged as digest-format paperback monthlies) anthologies emerged as a market. But after 1955 it was never again truly possible to earn a living writing short stories (although this may be changing thanks to the e-publishing format shift—it’s increasingly possible to publish stand-alone shorter works, or to start up a curatorial e-periodical or “web magazine” as the hip young folks call them). And the readership profile of the remaining magazines slowly began to creep upwards, as new readers discovered SF via the paperback book rather than the pulp magazine. With this upward trending demographic profile, the SF magazines entered a protracted, generational spiral of dwindling sales: today they still exist, but nobody would call a US newsstand magazine with monthly sales of 10,000-15,000 copies a success story.
A side-effect of dwindling sales is that the fixed overheads of running a magazine (the editor’s pay check) remains the same but there’s less money to go around. Consequently, pay rates for short fiction stagnated from the late 1950s onwards. 2 cents/word was a decent wage in 1955—it was $20 for a thousand words, so $80-500 for a short story or novelette. But the monthly magazines were still paying 5 cents/word in the late 1990s! This was pin money. It was a symbolic reward. It would cover your postage and office supplies bill—if you were frugal.
There is some sign of a recovery in this area since the mid-00s. I can point to a couple of high-end web based markets whose peak rate (for “name” authors) is 50 cents/word; at $500/thousand words they’re actually competitive with newspaper op-ed writing. But in general, novels pay much better than short fiction. A 100,000 word mid-list novel that reaps a $10,000 advance has netted the author 10 cents/word, and because it’s a single articulated narrative the author hasn’t had to re-start from scratch with new characters, ideas, and settings every 5-10,000 words. 100,000 words of novel are much easier to write than 100,000 words of short stories.
This is by way of explaining that, from the perspective of an ordinary working writer (who is trying to earn a living), short stories are good for three things:
Learning the trade
Advertising your wares
Fun and experimentation
First, learning the trade. If it’s your ambition to write novels, why would you start with short stories? Many people don’t; it’s a peculiarity of the SF/F field that we have this tradition of starting with shorter works (because, unlike the mainstream genre categories, we had some surviving markets for them). But I think if you want to write, diving in and trying to start with a novel is asking for trouble. Novels are complex beasts and if you write one you have to be able to keep track of a whole bunch of different narrative structures that overlap, on different levels: the plot arc, character development, thematic elements, and so on. (For a whistle-stop tour of these items I’d strongly recommend one book: On Writing: A Memoir of the Craft by some guy called Stephen King, who obviously thinks he knows something about putting together novels.)
Short stories are short. Consequently, you can’t cram everything in. As Isaac Asimov observed, short stories are about what you leave out—and the shorter the format, the more things you can safely ignore. This makes them an ideal learning vehicle. You can write a 2000 word character study in an hour or an afternoon. Someone else can read it in 3-10 minutes. It’s not demanding. You can show it to other readers and other writers and they can dip their toe in the water and tell you what they think without the sinking feeling that comes from receiving a 500 page doorstep, or a megabyte of prose.
Short stories therefore lend themselves to workshopping and learning and training, so they may not be very commercial, but they’re bloody useful all the same (and it’s no accident that while I sold my first short stories in the mid-1980s it took another decade before I could produce a readable novel, and another few years after that before I was firing on all cylinders).
Secondly, Advertising your wares. This probably doesn’t require any explanation, but: because short stories are short and easy to read, there’s a low threshold to uptake. It’s much easier to try out a new author by reading a short story than by diving into a trilogy. Or to try out the world on an on-going series by reading a single story. (Quick: how many of you first began reading my Laundry Files series after you stumbled across a short story in that setting?)
Importantly, among the folks who read short stories to discover new authors this way are numerous editors. If you’re an editor, one way to build a reputation in your field is to discover the next big thing. And because the next big thing probably starts out writing short fiction, that’s where a lot of the smart novel editors keep a weather eye pointed. If someone appears out of nowhere and begins grabbing Hugo nominations for short fiction, then the manuscript of their first novel lands in your inbox, you pay attention.
Thirdly, Fun and experimentation. When you’re writing novels for a living, your income stream for months or years in advance depends on getting the current project right. You are writing to a deadline, and novels are big and cumbersome: wrestling one into a new shape can take weeks or months. So there’s a natural tendency to be extremely conservative with your writing style, to avoid big risks. (Examples of “big risk”: writing a police procedural in multi-viewpoint second person present tense. Or using a highly unreliable narrative viewpoint who is basically lying to the reader, with the intention that the reader will eventually smell a rat and begin to interpret the real story hidden in the background. And so on.)
Short fiction is short. If you’ve got a few spare days you can push out a novelette that does something so radical and experimental with language that it would have an editor reaching for the smelling salts if you did it at novel length. (Unless it’s the late 1960s and your name is Brian Aldiss). A handful of readers will appreciate it, you’ll get your stationary and printer consumables paid for, and you might win a shiny award from those who care. Meanwhile, you’ll have learned whether or not something works, and you can use it later in your bread and butter novels. (Personal example: Bit Rot. I had an invitation to do a far-future space opera story for an anthology. I had a universe lying around, under-used, from “Saturn’s Children”. I wanted to see if there was any life left in it, so I wrote this short story. Yes, it was still viable: I could still work in it. So then I knew I could use it as the setting for “Neptune’s Brood”.)
Finally, I have mentioned the existence of format shifts, when an entire wholesale/retail distribution chain goes away and a new one emerges, causing a shift in the type of work that authors are paid to write. A huge side-effect of the ebook shift is the sudden resurgence of the novella, a format that was previously in eclipse. A novella is a month’s work: a gamble, but if you take a month off to play with one between books, you’re probably not going to starve—you’ll just have to work harder to make up time. Novellas are too short to bind and sell as novels in their own right (with a few exceptions—notably signed limited edition numbered hardcover runs that don’t get discounted down to peanut husks by Amazon), but they thrive as ebooks at a price point between $0.99 and $2.99. And they’re taking off these days, in numbers and in readership. So I find that, interestingly, the novella seems to be a natural overlap point between a format I can get experimental with and one that can pay its own way.
But in general, if you were wondering why I don’t write short fiction much (if ever) these days, the reason is simple: I’m not learning the basics of the trade, and I don’t need to work for chickenfeed to advertise my wares. Fun and experimentation is still valid, but I’m in the happy position of having been given so much scope to experiment in my novel-length work that I seldom need to go there. And so, this is why many (but not all) working novelists start out as short story writers but seem to go off the boil after a few books.