Mathematically Generated Folk Tales

Work in progress for my elective unit about Fairy-tales. This is a bit of an ongoing process but I am essentially writing my own fairy-tales mathematically, using the constraints they give me as a jumping off point to set out the plot. I basically got interested in this type of work from working at the college library. In an art library, the Dewey Decimal system is split down into increasingly lengthy decimals to account for all the similar art-related books there are. You would not believe how many authors thought "Perspectives on Perspective" was a witty title and naturally there has to be some way to distinguish them. This leads to most decimals being about 6 digits long, indicating the subcategories; Art->Artists->Monographs->Painter->European->British->Abstract etc.
I got to thinking that with a designation in the Dewey system gives you a pretty good idea exactly what the book is about, and that you could theoretically write a book to fill a vacant slot as it were. I liked the idea of the system designating content rather than cataloging it.

Fairytales are a pretty good place to start off because they've been elaborately analysed by generations of Western Literature Theorists, as have the conventions of storytelling themselves. Because critics have this penchant for dividing plots into ever broader categories and types, it was fairly easy to organise them into some kind of coherent diagram.

It's a mess of cultural theories lashed together, but it actually seems to function pretty well as a prototype! The user randomly selects one option at each stage (number between 1-3/ number between 1-8, etc) and at the end selects a position in the Aarne-Thompson index, a fantastically arbitrary system that gives Folk tales a position based on genre-type, principal characters, interactions and plot-lines. The Wikipedia article is worth a look, just for how entertainingly obscure the whole list is.
It's been an interesting experiment, and it's made great steps towards what I was aiming for, which was an uncanny, unnerving story; something that riffs off the same rules and formulae of a folk story, but combines the elements in unusual ways, jarringly, disconcertingly muddling archetypes into a story full of symbols but seemingly devoid of meanings. I suppose this is quite a psychoanalytic take on the Fairy-tale as a cultural item, but approaching it from the angle of a rigid structuralist.

Well, whatever.
I'm not completely happy with it yet, and I have a horrible feeling that I will have to add whole new levels of complexity to get the type of results I want, probably involving Tarot cards and and a great deal of shuffling and relocating and those polygonal-dice most commonly favored by Dungeons and Dragons enthusiasts. Which I'm cool with, I guess, I'd just prefer it if it didn't have to be this huge dramatic undertaking with all the paraphernalia. I'd like to keep it simple, like a computer program, but would this ruin the cobbled together aesthetic? More work needed.

Elliot