Examples of the man-beast myth are found among africans, japanese, and native north and south americans, not just europeans. And I read an article which argued convincingly that 'lots of eskimo words for snow' isn't true - if you talk to say, an american skiier who knows lots of words for kinds of snow, and an american meteorologist who know a different set of words for snow, you would get about as many as the eskimos have. It's hard to know who's right without studying eskimo. But I have studied comparative mythology, and it is my belief that all myths are fundamentally similar. A polar bear on an iceflow could be easily substituted for a leopard on an island in a river. A giraffe is not that different from a moose. I believe that there is no inherent meaning in specific animals or geography, all meaning is imposed on the world by the human mind, and all human minds impose meaning in fundamentally similar ways.

Consider the concept 'magic'. It is directly relevant here, since magic, having no existence in reality, can only operate according to the same 'dream logic' we have been discussing. Magic is not real, yet there must be some reason every culture has an idea of magic, and even agrees that it operates on the principles of sympathy and mysterious words or glyphs.

Quote: Original post by sunandshadow
Examples of the man-beast myth are found among africans, japanese, and native north and south americans, not just europeans. And I read an article which argued convincingly that 'lots of eskimo words for snow' isn't true - if you talk to say, an american skiier who knows lots of words for kinds of snow, and an american meteorologist who know a different set of words for snow, you would get about as many as the eskimos have. It's hard to know who's right without studying eskimo. But I have studied comparative mythology, and it is my belief that all myths are fundamentally similar. A polar bear on an iceflow could be easily substituted for a leopard on an island in a river. A giraffe is not that different from a moose. I believe that there is no inherent meaning in specific animals or geography, all meaning is imposed on the world by the human mind, and all human minds impose meaning in fundamentally similar ways.

Consider the concept 'magic'. It is directly relevant here, since magic, having no existence in reality, can only operate according to the same 'dream logic' we have been discussing. Magic is not real, yet there must be some reason every culture has an idea of magic, and even agrees that it operates on the principles of sympathy and mysterious words or glyphs.

I'd be interested in reading that article about the eskimos. It seems to me, however, that comparing words that everyone in a language group knows (eskimos) and words that only a few people with specialized skills (skiiers, meteorologists) in a language group knows isn't enough to disprove the basic assertion.

I don't agree that all myths are fundamentally similar, but I would acknowledge that many of them are. We do after all have much in common physically and emotionally. However, I think that comparative mythology tends to overlook the differences in favor of the commonality, which can produce mis-understanding. I agree that meaning is imposed on the world by human beings and that the meanings assigned to particular animals or geographic locations flows from the human mind, but I'm hesitant to agree with the notion that meaning is imposed in fundamentally similar ways. Perhaps it was at one time in the far distant past, but I think meaning piles up in ways that change how it's imposed. The simliarities of magic and man-beasts aside, a sun culture is going to construct a different social reality than a snow culture, a sea culture, an industrial culture. Meaning isn't static, culture isn't static. Things get dropped and things get added in the transmission of meaning from generation to generation. Anyway.

Yeah, anyway. It is necessary to understand this 'dream logic', to pull it from the subconscious into the light of day and encode it in algorythms if we ever hope to be able to generate fiction, because these are the rules by which the details of plot are generated. And it doesn't really matter whether all people have exactly the same logic, because any one person's will do - it only takes one human author to write a book, so a program which emulate only one human author could also conceivably write a book. So the question is, how do I identify the principles of 'dream logic', where and by what method can I research them?

I think you're on the right path.

Interesting discussion so far, although it's strayed far into areas such as cultural psychology in which I'm too uneducated in to add anything meaningful!

If by "dream logic" you mean something similar to "story logic" - the set of logic rules that define the likely progression of a story - then that's pretty much my opinion of how to go about creating a computer story generation system. Each rule set will be different for a certain story type, so I don't expect the rules to be the same for a fairy tale, a detective novel or an action movie script.

Of course, the problem is that for most story types the "story rules" deeply codify a large part of the human condition, and this is very, very hard to express in terms of logic rules, because many of the rules are so ingrained it's hard to realise they are rules until you write some of them up and notice the computer making boneheaded assumptions.

That's why some of the best computer generated story systems I've seen have deliberately chosen domains that have a simlpe set of rigidly codifed story rules, such as fairy-tales. Luckilly, I also this is the one boon we have if we aim to create a "computer game" story generator, because many computer game story types get by with a similar set of rigid story rules (such as bad guy wants to destroy the world, kidnaps girl, hero fights a series of enemies and mini-bosses, obligitory plot twist two-thirds in, hero kills bad guy and rescues girl. Or even the internet lists of amusing RPG cliches). I'm sure this aspect of having a cliched series of storylines probably doesn't appeal to you as a writer, but it does make things easier for a computer program [grin].

Of course, if you do find a good way to encode logic for the human condition (or even a good smoke-and-mirrors way to fake it), then I'd really like to know myself!

See, you always say that about each story type having its own rule set. [wink] But I think they're only as different as two kinds of sentences written in the same language, and you can't understand grammar without studying the whole language. I believe that all forms of fiction are generated from the same story logic, and my whole theory is built around identifying these underlying principles. I feel that so far I've described the high level overall structure of plot, now I just need to research the low level to figure out how the details of plot are generated. There's no point limiting myself a particular genre or theme, I will actually learn more by comparing widely different examples of fiction.

I guess I do tend to write that a lot. I guess I'm combining my opinion of how to approach the problem with the general theory - probably not the best thing to do!

While I think that there probably is a theoretical set of rules that could generate all possible story structures, I think that deriving that set is much harder than limiting the domain to a single genre - several orders of magnitude harder, if a scale of problem difficulty exists [smile]. Given that I haven't yet seen a story generator limited to a single sub-genre that didn't have serious problems, I guess I feel it's more pragmatic to pick a small subset to begin with. I suppose an appropriate analogy (going with your comparison to sentences) would be if I were to try to build a robot speech recognition system for taking orders in a fast food restaurant - it would be much easier to just get the robot to recognise key words (items on the menu, key phrases, prices etc.) then to make a system that could completely understand English.

I think it's a good idea to consider the notion of a theory for all stories as it's a very interesting subject, however I strongly suspect that in terms of practical results the next big advance in interactive storytelling and story generation will be a very specialised system, combined with a bunch of cute tricks to disguise that it's computerised (similar to how chatbots work, I guess). I feel that if a model is too all encompassing there's a danger that increasing the "breadth" of the model will decrease the "depth" of the types of stories it could create.

It's threads like these that make me wonder whether this content is above me, or if I've just never been taught the right things. Guess I'll try sticking around to find out.

I find it interesting that someone would try to break a story down into mathematics. As someone who has done relatively well in both mathematics and English, I've honestly never had the urge to relate the two. Especially with stories. Everything starts with a sentence, a line of dialogue, or a thought. I don't try to break it down past that because for me, that is its purest state.

Then again, I've always been about intuition. Whenever I come up with an idea, or write a story, I trust my intuition for what will and will not work. I fear that if you try breaking down a story to such a basic level, the writer will lose that intuition. And that intuition is critical for trying to take advantage of the psychology, and yes the soul, of that person.

That's not to say that I don't think breaking down a story to this level isn't useful. I think it's something that I, as an aspiring writer, should look into, because it may help me understand what is it that I'm doing. But I'm not sure if I agree with putting in such contents into a how-to-write book. Maybe if it's within the appropriate context.....

Anyways, this kind of analysis of literature is new to me. I may know what I'm talking about, but I don't know if I know what I'm talking about. :P

Feel free to counter my point.

The above poster is me, Oilers99. Sorry, but I'm a n00b, so forgive me if I make mistakes.....

I'm not entirely sure what sunandshadow's application is for her model of thematic arguments (edit: after scanning the rest of the forum, it might just be a study of literary theory?), but my interest in expressing stories in terms of mathematics and logic comes from my long-term goal of eventually using procedural story generation to make properly interactive storytelling within games (i.e. game stories that can change due to the choices made by the player). It's a sort of "off again, on again" project that I've been pondering about for years (and not actually getting that much practical work done, to be honest!). Presently the project is "off" while I work on other things, but I'll probably start working on it again next year.

For my purposes, I'll have to be able to represent stories in terms of logic symbols and mathematics, otherwise I won't be able to break them down into elements able to be pieced together by the computer. However I agree that computerised stories do seem more than a little "soulless" at the moment.