Researchers analysed 1700 novels to reveal six story types

http://www.bbc.com/culture/story/20180525-every-story-in-the-world-has-one-of-these-six-basic-plots

by sdeepak • 8 years ago

170 98 8 years ago

sdrothrock2• 8 years ago

Thanks to new text-mining techniques, this has now been done. Professor Matthew Jockers at Washington State University, and later researchers at the University of Vermont’s Computational Story Lab, analysed data from thousands of novels to reveal six basic story types

It seems a bit naive to say "every story in the world" when they analyzed only "thousands of novels" -- this doesn't even mention what languages they analyzed. I could imagine many Greek-influenced cultures (e.g. "the West") having similar archetypes by virtue of a common ancestor. Language analysis tools for CJK aren't, as far as I know, as advanced, so I can imagine a lot of stories from those languages being left out as well.

The reason I bring this up is that I'm reading a book called "Don't Sleep, There Are Snakes," [1] which is about a remote tribe of people in the Amazon jungle called the Pirahã. [2]

One of the most stunning points in the book is that their culture is such that all stories told by native Pirahã are based on first-person experience. When a Pirahã dies, their stories are not passed on or retold. Due to this, there's also no need to record past stories, orally or otherwise.

This kind of study would obviously exclude stories from that kind of culture and seems a bit narrow-minded: "everything I looked at says A, so EVERYTHING must be A."

[1] https://www.amazon.com/dp/0307386120

[2] https://en.wikipedia.org/wiki/Pirah%C3%A3

robotresearcher• 8 years ago

Why would it exclude Pirahã stories?

"I got up this morning and everything was fine. Mid-morning I fell out of a tree and had a hell of a time getting home. But I made it and was so happy to see my family again."

The rise/fall model is rather trivial, but your point doesn't stick.

sdrothrock2• 8 years ago

Why would it exclude Pirahã stories?

Because they analyzed "thousands of novels." There are no Pirahã novels. The point I was trying to make is that the sample set of 1700 novels, most likely from a Western tradition, is laughably small to claim it applies to "every story in the world."

smithmayowa• 8 years ago

What about fall to fall tales and rise to rise tales that abound plenty-full in several cultures like mine.

acbart• 8 years ago

Trivially reducible to the concatenation of a Rise-to-rag and a Rags-to-rise tales! Clearly nothing new there.

gaius• 8 years ago

Thanks to new text-mining techniques, this has now been done

Thanks text mining techniques! Here's an article saying the same thing almost word-for-word from 2004 https://www.telegraph.co.uk/culture/books/3632074/Everything...

stephengillie• 8 years ago

Everything new is old already.

shakethemonkey• 8 years ago

And then there's TV Tropes[1] which uses a finer measure

[1] https://tvtropes.org/pmwiki/pmwiki.php/Main/Plots

hprotagonist• 8 years ago

I wonder if I can construct a Cantor-like diagonal proof that creates a definitively non-countable plot.

tlb• 8 years ago

Here, the only tokens are rise and fall and they must alternate. So you can enumerate plots by picking a starting token and a number of alternations. Thus there are Aleph_0 plots.

If you tracked multiple hero/protagonists then their interleavings could be uncountable.

throwaway287391• 8 years ago

Wouldn't multiple protagonists still be countable by the same proof that the rational numbers are countable?

manwe150• 8 years ago

Is it required that “rise” and “fall” be boolean/digital operations? I think it would be more reasonable to characterize them as analog/continuous. That could then lead to infinite variations in rates, even just with a single character. And that’s before considering whether the characters and reader are correct in determining their current trajectory—much as in real life.

vichu• 8 years ago

A great book on this subject is Joseph Campbell's The Hero with a Thousand Faces. His theory of the archetypal hero and the theory of the monomyth is a great influence on anlyses like this one (and also on some artistic works - including a small film series known as Star Wars). It's fantastic to see these natural language processing techniques applied to this topic.

sambeau2• 8 years ago

I blame Joseph Campbell for ruining modern cinema with it's simplistic boring plots (like Star Wars's) where yet another hero battles some evil tyrant to save the world, universe or forest.

There are so many more interesting, surprising stories to be told that leave the viewer with unresolved things to think about and captivating cul-de-sacs along the way. The 1970s films were full of odd, interesting plots and films where it was unclear who the protagonist was (let alone a hero). Then Lucas discovered Campbell and made two pretty-good, unbelievably successful kids' films and modern cinema was born.

Now every film (especially kids' films) are a take on a James Bond plot whether they are about loveable zoo Creatures and cheeky penguins, toys in a nursery-school or talking cars: a super-villain, a secret base, a mysterious super-weapon… yawn

Grown-up films are no better: reduced to a sorry stream of infantilised garbage about super-heros who each in turn is given super-powers only to see them quickly taken away to provide some tension before a giant battle in which cities (and budgets) are destroyed for no good reason (accept perhaps to deliver a few half-decent one-liners) before a last-minute triumphant victory in a working steel-mill (if budget allows) or some disused warehouse if not.

sigh

nprateem• 8 years ago

Agreed. Hollywood wanted a formula and found one in Campbell. After reading that book or others about the "hero's journey" most mainstream films are quite bland. Having said that, there are only 7 notes in the major scale and look what that's done for us...

yesenadam2• 8 years ago

It sounds like you're talking about US movies only. There's a whole world of cinema out here. 'modern US cinema' =/= modern cinema. every US film =/= every film. Sounds obvious, but (to take one example) the majority of "Best films of the year" lists I come across online actually have only US movies, and seem oblivious that movies are made in any other country, let alone most of the good ones. 'Grown-up films' of the world, not just the USA, are thankfully not in such a sorry state as you describe. Well, the marketers/advertisers/etc that made people in the US think that 'movies' = 'US movies' did a great job.

sambeau2• 8 years ago

My kids & I watched the Coen Brothers, "Burn After Reading" last night.

I'd love to hear how anyone could argue that it fits any one of these.

qbrass• 8 years ago

It's multiple stories driven by the same set of events.

Osborne and Harry both have riches to rags stories of varying degrees. Things get progressively worse for the both of them, but Osborne wasn't doing so great to start with.

Kate and Linda have Cinderella stories of varying degrees. They both start off thinking they're getting what they want, things get worse for most of the story, and both end up with basically what they started out trying to get in the end.

vntok2• 8 years ago

Riches to rags for Cox (from CIA analyst to vegetable in hospital).

Icarus for Chad (from aspiring millionaire to dead in mere days).

Cinderella for Linda (from aspiring millionaire to everything going wrong, losing Chad, etc. to having her surgery paid in full by the CIA)

johnchristopher• 8 years ago

I am of the opinion that the plot doesn't matter as much as the context into which the story is being told: who are you telling the story to and why ? Which view of the world are you conveying at that moment ? Why tell that particular story now and to which audience ?

Moreover, the receiver's personal ties to the story and its elements (characters, settings and then the meta:translation, origin, etc.) before, during and after getting story matters as much, if not more.

busterarm• 8 years ago

This feels pretty reductive. They chose one variable to measure stories on, fortune, and then of course all of these stories follow a short list of possible patterns.

This counts as research?

ummonk• 8 years ago

What they should research next: "every story falls into one of these basic length categories!"

projectramo• 8 years ago

Short, medium and long.

Researchers could not find other lengths although they hypothesized that if someone wanted they could create a “mega” category. “It’s possible,” they concluded.

yellowapple2• 8 years ago

I'm curious if they count television as storytelling here. If so, then the whole premise is demonstrably false; your standard soap opera will go beyond the stated one or two fortune-inflections in a single episode, let alone a season, let alone a series run.

thinkingemote3• 8 years ago

There are no television scripts in the article.

thinkingemote3• 8 years ago

They don't choose fortune, they do sentiment analysis on the words in the text. That this might be mapped or interpreted as being the same as fortune is an interpretation. For example positive words don't necessarily mean fortune in the plot, but often a plot showing fortune will have positive words.

This is an important distinction when looking at whether it's good or bad research.

ggggtez• 8 years ago

If they found that despite the fact that a story can start happy and get sad, no story had ever done so, and it actually seemed to be impossible, well that would be quite an amazing discovery.

However they only discovered that a variable can go up, down, or stay the same. Not exactly rocket science.

ur-whale• 8 years ago

I'm not sure I'm buying the theory laid out in this article, for at least two reasons.

First, I'm not even sure every good story can be shoehorned in the framework.

For example, how does the archetypal "vengeance story" fit in that framework?

In that type of story, many a time, the seeker of vengeance, once he's reached his goal after much struggling, finds he's destroyed himself in so doing.

Trying to apply the framework leads to the main character essentially never getting out of the hole ... you could argue that reaching the vengeance goal is a super short "high" of the story, but I find the argument specious.

Second, even if we assume that every story can somehow be shoehorned into a simple curve of highs and lows ... what does that actually teach you about the story?

More importantly, what kind of creative help does that buy you? I'm pretty sure I coud "reverse-engineer" any highs-an-lows type curve into a story rather easily ... will that make the story any kind of interesting? I doubt it.

apo• 8 years ago

In the early-90's I saw Vonnegut give a presentation in which he talked about his story graph idea.

However, the video linked to in the article appears to end before the most interesting part.

https://www.youtube.com/watch?v=9-84vuR1f90

Hamlet. It's never clear whether what's happening belongs more toward the good or bad side of the vertical axis.

The curve you get is a flat horizontal line. That's art. Never clearly good or bad but profound ambiguity.

Also, for those interested in the systematic (graphical) formulation of stories, check out StoryGrid:

https://www.goodreads.com/book/show/25455700-the-story-grid

8bitsrule• 8 years ago

The curve you get is a flat horizontal line. That's art.

Maybe. It's also death.

stepvhen• 8 years ago

Pretty sure most folk in Hamlet get a bad deal in the end, and Hamlet's brooding and whining doesnt outweigh that. Ophelia literally goes mad and is driven to arguable suicide. Everyone that mattered dies, and if that is outweighed by Fortinbras and his good fortune in the end. Hamlet is Icarus if anything, since it only goes well for him for a moment while everybody else is on a downward spiral.

mtVessel• 8 years ago

Aristotle came to the same conclusions somewhat earlier, using only natural intelligence and manual learning techniques.

mrob• 8 years ago

Counterexample:

A poor man finds a magic lamp. He summons the genie, and wishes for great fortune. (Rags to riches.) He makes another wish, and it horribly backfires. (Icarus.) No longer trusting genies, he wishes he'd never found the lamp. The genie rewinds time. A poor man finds a magic lamp.... (Infinite loop plot not included in the classification.)

EDIT: Downvoter, how would you classify this story?

keybuk• 8 years ago

Depends which season of Star Trek it's from.

Early in the run it'll come across as kinda corny, but mid-way through the run it'll be a strong episode and one fans will cite as a favorite forever.

MrTonyD• 8 years ago

It took me a while to figure out that on HN you get downvoted if you aren't agreed with or your opinion is disliked - quite often nothing to do with any actual intellectual argument. I've learned to ignore all the downvotes that I get - they are meaningless, and usually associated with what I consider to be my more insightful posts. (Since more insight implies that more other people haven't made the same observation - thus more downvotes.)

__s2• 8 years ago

That's alright, next you need to make the sequence normal so that it's never repeating

Another counter example: https://serprex.github.io/w/Ad%20nauseum%20ad%20nauseum

This is a story I started writing last year, it doesn't have rises & falls, just someone reminiscing on the way to the corner store for a bag of chips

If my writing doesn't count since it's just some random page on the internet, there's Bob Dylan's Tarantula, & he's won a Nobel for literature. You'll be forced to say "Every story has one these six basic plots, where a story must have a plot, & a plot must be composed of rises & falls" since anything that would go against that gets categorized as a "sub story" or "not part of the story"

kbutler2• 8 years ago

"goto start" doesn't make it a new story, it's just a repeat, or re-telling of the story (possibly with variations, a la groundhog day)

I also disagree with the contention that all "stories" have to fit the patterns, unless you narrow the definition of story to fit the patterns.

lozenge2• 8 years ago

The timeline may be an infinite loop but your story still ended, with the revelation that the man is doomed to repeat his Icarus story for everytime. An author could play that as a sad or happy ending, as they like.

jon_richards• 8 years ago

Not if it's a goosebumps novel (they have you jump to specific pages based on choices you make).

sacado26• 8 years ago

There's a traditional variant of this schema:

A poor man uses all his money to buy a horse. "You must be happy", his neighbor says. "I don't know", answers the man.

The next day, the horse escapes. "Your hard-earned money, and your horse, both gone. You must be sad", his neighbor says. "I don't know", answers the man.

A few days later, the horse comes back, with a female horse. "Now you have two horses, and very soon a third one. You must be happy", his neighbor says. "I don't know", answers the man.

The next day, the man's son tries to care about the new horse, but the horse rebels and the son has his leg broken. "That's your only son, and know he won't be able to work or provide for you in your old days anymore. You must be sad". "I don't know", answers the man.

A few days later, this is war. All young men must leave and go to war. Since the man's son is injured, he doesn't have to go. "Your son won't have to face the horror of war. How good for both of you ! I wish my own son was still there with me".

Some versions of the story stop there, but it can keep going, and end either on a positive event or on a negative one. Basically, this is a story about how you can never know if an event will be good or bad, no matter the first impression. It cannot fit any of those story patterns.

TJSomething• 8 years ago

That sounds a lot like some kind of episodic serial.

jessaustin• 8 years ago

This is a parable, not a novel. It's hard to imagine expanding this narrative to anything novel-length and also readable, so it's unrelated to TFA.

Of course, GP post is similarly OT.

atentaten• 8 years ago

This is from the Tao Te Ching. This verses teaches one to be balanced and not to be attached to outcomes because in life there are endless highs and lows.

talltimtom• 8 years ago

it cannot fit any of these story patterns.

Of cause it can, just because you have a character proclaim he doesn’t know doesn’t suddenly make it so. The curve goes down when his son breaks his leg and it goes up when he doesn’t have to join the army. There is no requirement to integrate all future development into the current state of the curve.

pssflops• 8 years ago

"There are two tragedies in life. One is to lose your heart's desire. The other is to gain it.” ― George Bernard Shaw, Man and Superman

brenschluss• 8 years ago

Also see: Finnagans Wake, which is a loop of a story. https://en.wikipedia.org/wiki/Finnegans_Wake

SubiculumCode2• 8 years ago

If you reduce to 1 axis all plots then yes all changes are on that axis.

blueboo• 8 years ago

lol!

tldr:

Every story has change.

* Positive

* Negative

* Positive, then negative!

* Negative, then positive!

* Positive, then negative...then positive!

* Really negative...then positive!

if that's what you learn after studying 1700 novels, i'd give up on fiction entirely

SubiculumCode2• 8 years ago

Even if we were to allow that the story can be described on one dimension over time: Real stories do not have one arc.

A lone melody works on one dimension also: Absolute rises and falls.. This can be pleasing but is limited in it expressiveness. Add another note however, and you get another kind of rise and fall: Relative. One note is falling, but the other note is falling faster. One note rises while the other note falls, but then later falls when the other rises.. Now add a third note, and you start to get to hint and deceive where each note will go, fulfilling and violating expectations.

Now that sounds like a real story.

falien• 8 years ago

The blog post this was based on (http://www.matthewjockers.net/2014/06/05/a-novel-method-for-...) which is from 2014 was widely panned when it was first publicized, and some of the many problems with the methodology pointed out by scholars with more experience and competence with the statistical methods involved. That's how scholarship is supposed to work. However, it will never stop media from seizing on a great click-baity headline.

simtam• 8 years ago

Every time I see such a list on the internet, I ask myself "So, now, which of those basic plots is 'Master and Margarita'?", which leads to a conclusion that these lists bear a negligible amount of cognitive value.

stewbrew3• 8 years ago

Following this typology: Man in the hole? The "Master and Margarita" isn't that complex with respect to its basic plot. The plot doesn't tell you much about the story though.

vntok2• 8 years ago

"Master and Margarita" seem to follow the BBC post's Oedipus storyline: Fall (rejection of the Master's novel, his desperation, separation from Margarita and internment into a psychiatric hospital), Rise (Margarita's offer from Satan brings her power over the elements, the Master finally gets out of prison and the couple gets back together) & Fall again when they are poisoned and die in disgrace both from Satan and God.

stewbrew3• 8 years ago

Or maybe, every story has one of this one basic plot: (1) setup/intro/beginning, (2) stuff happens, (3) things somehow come to an end (for now).

sacado26• 8 years ago

Not sure you even need the first part. A character has a problem and tries to solve it, until he either succeeds or definitely fails. Lots of stories start with a conflict at the very first line, without any kind of setup (especially thrillers).

hexane360• 8 years ago

(1) stuff happens

stewbrew3• 8 years ago

Well, most stories somehow try to establish who is the protagonist of what follows.

jVinc• 8 years ago

I'd love to see someone drawing out the graph for A song of fire and Ice. Not only do you have several simultansus character arcs to draw out, but through the story arch they are also discontinuous and mix between being interwoven and separate in quite interesting fashions.

Ptyx• 8 years ago

What about Waiting for Godot? The play in which nothing happens, two times.

weeksie3• 8 years ago

Plot structures are tools. All else being equal it's a good idea to pick a plot that's as simple as possible and hang the interesting stuff off that. The classic mistake of beginning writers is to go for complexity in plot and then spend their entire time serving that plot to the detriment of their characters.

That said, of course not every story fits into some basic set of plot outlines without some serious mental acrobatics. Plot isn't even necessary, what's necessary is getting inside people's heads. The classic plots (however they're formulated) are all good ways to do that but you can also do that with meandering unstructured prose, if you're good enough, though few people are and the further you get from universal tricks the more likely you are to run into serious disagreements over who writes really good nonconformist stories.

Maybe a good example is dialog. The beginner's mistake is to write as if people are listening and responding to what the other has said. e.g.

"Hey Joe, what's up?"

"Just whippin' up some breakfast."

versus

"Hey Joe, what's up?"

"Jesus, you look like shit. Want some breakfast?"

In the second example the reader fills in the blanks mentally. It creates engagement. Do that in the small throughout your book and few people will complain that you lack a plot. Easier said than done.

Plots are good. Use them. But they're not the whole, uh, story.

jacobkg• 8 years ago

Can you give reference(s) for that beginner’s mistake? I’ve never heard it before and it’s definitely something I do. Would love to learn more about this!

__s2• 8 years ago

Philip K Dick's Gather Yourselves Together had pretty banal dialogue

thrusong• 8 years ago

In journalism/media school, we're taught "show, don't tell" in both our news and creative writing courses. You want to use detail to set the atmosphere/tone without flat out saying it.

weeksie3• 8 years ago

I can't think of a specific ref off the top of my head but there are great books out there. Also, I used dialog as an example but what makes a story addictive is the act of filling in the blanks. You can do that with emotional transitions or imagery and that's what I meant by "do that throughout your book and few people will complain about a lack of plot"

Plots are a great tool in that they have that part built in.

I'm several champagnes in at this point on a transatlantic flight and we're landing but I'll go through my list of fave writing books when I get settled

jungler• 8 years ago

It's just a general clear communication principle that can take many forms. If you have filler content for the sake of padding out the work, it isn't communicating anything, and the audience will know it.

Take that thought in the other direction - how to communicate more efficiently - and you end up with snappy movie dialogue and non-verbal interaction that can push a plot forward really quickly.

With any kind of communication the challenges involve a mix of both knowing what techniques are possible, and having messages worth focusing those techniques on.

whydoineedthis• 8 years ago

Wasn't this already well known from Kurt Vonnegut lectures? https://www.youtube.com/watch?v=oP3c1h8v2ZQ

thinkingemote3• 8 years ago

As indeed the first and second paragraph states

keiferski• 8 years ago

Kafka doesn’t seem to fit these plots. Most of his work is about conveying a mood or feeling.

crazygringo• 8 years ago

This abuses the word "plot".

These aren't plots, they're highest-level structures. They're also merely all six alternations of up/down across two and three combinations.

"From bad to good fortune" or "From good to bad to good" isn't a plot.

"A young man leaves his town in search of adventure, his talent is quickly recognized by the military and he leads a platoon in battle which wins the war and he finds the meaning in his life he'd been missing" is a plot, albeit a very basic one.

dang• 8 years ago

Ok, we replaced the title with the plotless subtitle.

nwellnhof• 8 years ago

Right, the six "structures" they found are:

    /  \  /\  \/  /\/  \/\

rayalez• 8 years ago

There are dozens of writing mentors/gurus, and every one of them has discovered the story structure, or the 3/7/10/50 basic plots. All of them are different. From vague and meaningless "beginning-middle-end but not always in the same order" to Blake Snyder's Beat Sheet or Dan Harmon's story circle(Campbell's hero's journey).

They all are kinda true, for the same reason you can find the proof of UFO in the Bible, or determine that there are "two kinds of people", or discover 4 personality types. Anything can be shoehorned into any kind of pattern if your definition is vague enough and you're willing to stretch the meaning of things hard enough.

Different writers are familiar with different structures and use ones they prefer, bending and modifying them as they wish. Some writers break the rules on purpose. Some writers get high out of their minds and write stream of consciousness novels in a weekend. Any of that can result in a great/famous/successful story. Finding the pattern that fits all of that is only possible if it's so vague that it can fit literally anything, at which point it becomes meaningless.

Plots and story structures can be picked up and used when it's convenient, it's easier to write when you have a paint-by-numbers structure to hang your story upon, but there isn't the one story structure or the six story types, that's silly.

CuriouslyC3• 8 years ago

You misunderstood the research in the article. They aren't saying that there are only 6 plots, rather that there are 6 basic progressions of valence (i.e. mood of the story) that almost all stories are minor variations of. There are of course variations that are outliers, and the way in which these valence transitions occur differs from story to story, but the emotional "ride" these progressions create should be very similar.

madcaptenor• 8 years ago

So if you figure that stories are composed of rises and falls, and they have to alternate, don’t you get this result automatically?

nl• 8 years ago

This is the paper if anyone is interested: https://arxiv.org/abs/1606.07772

I've done some work with one of the authors (Lewis, who is based at Adelaide Uni). They do some interesting work generally.

Theodores• 8 years ago

I think people's motivations for writing fiction - novels - could be reduced to six.

Some of these motivations are simple to understand - 'get rich', 'get recognised', 'sheer pleasure', 'because your failed author parents want you to repeat their folly' and so on. I have seen 'there are only n types of plot' stories before, I would much rather have motivations for writing novels being the categorisation as then I would be able to better identify what I did want to read, i.e. the 'banned books'.

There is a big difference between what some Eton educated aristocrat dictates to their secretary so they can be lauded in society as an 'author' to what someone in jail with no education writes to get themselves their freedom. This applies to words beyond novels too. For example the 'left wing' writings of Tony Benn (UK anti-war movement, aristocrat, dictated to secretaries) matter very little to me compared to what Malcolm X had to say.

New stories that don't necessarily fit into pigeonholes can come from real life where that period of history is a bit of a lie and the truth can't be told under any circumstances. During these times the only truth can be found in novels that transpose the names, times and places for something that the audience won't self-censor. So we end up with 'Oceania' or 'Animal Farm' with it not being clear what regime is being critiqued exactly.

Such works of fiction are written because something needs to be said, or at least the author thinks so. It is not like the author has decided 'I want to become a writer' and googled a plot from the internet to rip off. In this way literature is written, not just novels to make money with.

amp108• 8 years ago

Ignore everything that makes stories different from each other, only look at the beginning and the end states, and hey-presto, you can reduce every story to one of six types.

somberi• 8 years ago

Hemmingway, when challenged that simple words and short sentences do not work well, wrote:

Infant's shoes for sale. Never worn.

Apparently Ray Bradbury, when asked what science fiction is, supposed to have written:

Last man in the world heard a knock on his door.

hiisukun• 8 years ago

This reminds me strongly of a blog post [1] I read recently (probably on HN!) that listed an abstract form for 'all' the RPG plots. The list is unfortunately much than the six given in the bbc article, but the feeling I got from reading it is the same.

[1] http://rolltop-indigo.blogspot.com/2018/10/the-big-list-of-r...

Researchers analysed 1700 novels to reveal six story types

Related Stories