In 1965, Kurt Vonnegut submitted a thesis to the University of Chicago outlining how classic stories such as Cinderella shared a basic “shape”. It was rejected, and he was furious, speculating that it had been turned down because “it was so simple and looked like so much fun”, and raging that “the apathy of the University of Chicago is repulsive to me. They can take a flying fuck at the mooooooooooooooon.”
The thesis formed the basis of a lecture Vonnegut would give years later – watch it here, it’s so worth it – expanding on his theory with graphs to illustrate the various forms of story, from man in a hole (“somebody gets into trouble and gets out of it again”) to boy gets girl. “There is no reason why the simple shapes of stories can’t be fed into computers; they are beautiful shapes,” said Vonnegut.
A new academic study has done exactly this, and gives us yet another reason to wish the great man were still with us to share his thoughts on it (and perhaps resubmit that thesis). Researchers from the University of Vermont’s Computational Story Lab fed 1,737 stories from Project Gutenberg – all English-language texts, all fiction – through a program that analysed their language for its emotional content.
Putting – maybe – an end to a debate that has been ongoing for millennia, the researchers found there are “six core trajectories which form the building blocks of complex narratives”. These are: “rags to riches” (a story that follows a rise in happiness), “tragedy”, or “riches to rags” (one that follows a fall in happiness), “man in a hole” (fall–rise), “Icarus” (rise–fall), “Cinderella” (rise–fall–rise), and “Oedipus” (fall–rise–fall). The most successful – here defined as the most downloaded – types of story, they find, are Cinderella, Oedipus, two sequential man in a hole arcs, and Cinderella with a tragic ending.
Their analysis (and the “simple shapes of stories” as theorised by Vonnegut) is provided online, and it’s fascinating to pick through. I liked the rise-fall-rise shape of Gulliver’s Travels, where words such as “destroy”, “enemy” and “ignorance” drag down the happiness rating, and the plunging “Icarus” graph of Romeo and Juliet, plagued by words such as “tears”, “die”, “weep” and “poison”.
The researchers acknowledge that “there have been various hand-coded attempts to enumerate and classify the core types of stories from their plots” in the past, whether it’s Foster-Harris’s 1959 assertion that there are three basic plot patterns (happy ending, unhappy ending and tragedy), or Christopher Booker’s The Seven Basic Plots. The latter cites seven possible narrative structures: overcoming the monster (as in Beowulf ), rags to riches (as in Cinderella), the quest (as in King Solomon’s Mines), voyage and return (as in The Time Machine), comedy (as in A Midsummer Night’s Dream), tragedy (as in Anna Karenina) and rebirth (as in Beauty and the Beast).
The author Ronald Tobias, the researchers point out, came up with 20 “master plots”, including “underdog”, “metamorphosis”, “ascension” and “descension”. Georges Polti topped Tobias with 36: ranging from “rivalry of kinsmen” to “falling prey to cruelty of misfortune”.
It’s Vonnegut, however, who inspired the University of Vermont’s researchers most, in particular the similarity he finds between Cinderella and the origin story of Christianity in the Old Testament. In his autobiography, Palm Sunday, Vonnegut wrote: “I confessed that I was daunted by the graph of Cinderella, and was tempted to leave it out of my thesis, since it seemed to prove that I was full of shit. It seemed too complicated and arbitrary to be a representative artefact … but then I said to myself: ‘Wait a minute – those steps at the beginning look like the creation myth of virtually every society on earth.’ And then I saw that the stroke of midnight looked exactly like the unique creation myth in the Old Testament. And then I saw that the rise to bliss at the end was identical with the expectation of redemption as expressed in primitive Christianity. The tales were identical.” Vonnegut’s journey led the researchers “to search for all such groupings”.
The academics’ analysis looks at 10,000-word windows, each of which has its emotional content rated. This means they capture “the major highs and lows” of a story, rather than any briefly discussed emotional moments. It also only considers English literature, and the fact it is all taken from Project Gutenberg means it is restricted to older material in the public domain. But it’s a start.
It also comes at a time when new plot arcs are possibly forming: Angus Phillips, head of the school of arts and director of the Oxford International Centre for Publishing Studies at Oxford Brookes University, speculates that “the increase in reading on digital devices is changing the behaviour of authors”, with “a divergence between plot-driven, genre fiction aimed at a digital audience; contrasted with the character-driven approach of literary fiction, often read in print”. He points to Fay Weldon’s comments on her blog, that “short, in this the day of the galloping e-reader, is best. Writers need to envisage readers not turning the page as the maid draws the curtains and brings a glass of wine, but on the train or bus on the way to work, eating a sandwich, or standing in the queue for coffee.” It would be fascinating to see the University of Vermont’s programme applied to contemporary fiction, and to see if anything has changed today.
