Pattern Language For Writing Fiction

I've been interested in Patterns for a while now and have wondered about combining them with my main hobby - fiction writing. There are many patterns at different levels that assist in the craft of fiction; Pattern form would seem a useful way to encapsulate these.

Having recently attended an excellent two-day Patterns course given by JimCoplien et al I'm convinced that a PatternLanguageForWritingFiction is not only plausible but desirable. AFAIK no-one is working on this, so I've started on my Web site at http://www.fnapf.demon.co.uk/muse/index.htm. Currently there's only one pattern there - I'm working on others. BrokenLink as of 2003-04-09

-- TrevorMendham

TellingLiesForFunAndProfit (Lawrence Block) ISBN 0-68-813228-6 contains much advice in pattern-y form.

A work worth reading on writing is Writing For Story by Jon Franklin. ISBN 0452272955 (c) 1986 by Mentor Library of Congress Catalog Card # 86-063378. It is sub-titled "Craft Secrets of Dramatic Nonfiction by a Two-Time Pulitzer Prizer-Winner" -- despite the "non-fiction" claim the methods described are as applicable to fiction as to non-fiction. This is a good read and practical advice on writing as a craft. Embedded in the text, not quite in pattern format, are prescriptive methods for writing "story" -- and that's what all writing is about. -- RaySchneider

Unless what is being written doesn't contain any stories.

When you've finally written something, HowToGetaPublishingDeal?

CategoryBooks

Hey, I've been working on one of these privately for a while, off and on. It's neat to see there's some interest in this.

WarAndPeace is my source text, but I have looked at TheTaleOfGenji? with similar eyes. A lot of it ends up looking like LISP or APL -- but ThatComparisonIsPurelySuperficial?.

Some descriptions: hebrew/greek/cyrillic or other letter before command specifies character/place/motif/etc-specificity...

I do this when I transcribe recipes : it saves tons of writing to map bijectively, say, the set of all ingredients, and then simply use whatever letter or series of letters to refer to them in the steps...

The general command for introducing a main character is simply CD (which stands for character development) which you nest outside of New. So to introduce us to Pierre:

beyt(CD(New))

or perhaps Pierre has been introduced with an unattractive description:

beyt(CD(New(Desc:Unatt)))

or perhaps Pierre has been introduced from the perspective of A. Scherer, for whom Pierre causes tension:

aleph(TEN:beyt(CD(New(Desc:Unatt))))

or to introduce us to the cargo hold of a vogon ship, use a different alphabet and define it as the one used for all settings, say greek, then:

beta(SD(New))

or to specify it as being introduced from the perspective of ArthurAndFord?:

alephbeyt(Persp:beta(SD(New)))

The whole point of it right now is not so much to write fiction but to see how it is written... Like creating a periodic table of the different atoms which make up the story. I realize that any attempt to ecumenically describe the structure of fiction would be foolish : this is more just my way of enjoying the books I read, being able to see deep patterns in them, regardless of the fact that their existence may not spread far beyond my own emotional responses...

Some patterns I think I have found in fiction ... it seems sometimes that a good story is at its root merely good use of counterpoint ... for this reason I define all functions to have an opposite or a moral opposite. For CD this would be ACD, since I have noticed that often the best character introductions seem to work like this:

gimel(TEN:aleph(ACD(Desc:Att)));aleph(gimel(CD(Dial)));r-gimel(aleph(ACD));r-aleph(gimel(CD))

The next step will be to develop a shorthand for processes. I.e., the process above gets mapped onto, say, a letter of the Devanagari alphabet, or is perhaps a given Chinese character. This shorthand-for-shorthand stuff continues until thousands of pages have been captured by a few looming symbols and parentheses... -- AlecSinger?

Or, perhaps you should go back a step and use MeaningfulNames?.