Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Researchers analysed novels to reveal six story types (bbc.com)
96 points by rzk on Nov 9, 2023 | hide | past | favorite | 74 comments


The original paper [1] also contains some great graphs. And less hyperbole; these are the six basic plots of ~1400 English-language stories on Project Gutenberg, not of every story in the world, ever

1: https://arxiv.org/pdf/1606.07772.pdf


There is also the Kurt Vonnegut "Shape of Stories" concept where he has 8 basic universal plot lines graphed out back in the 1940s

https://thestory.au/articles/kurt-vonnegut-story-shapes/


I hate that they talk about 8 plot lines, but only show graphics for 5.


Skimming the paper: it looks like they are performing sentiment analysis on books, effectively taking the Fourier transform of the sentiment-versus-time data, and reporting which Fourier component (up to 3rd harmonic) is strongest.


I don't see why they stop at the third harmonic, as opposed to the second or the fourth.


Looks like these are just phantom oscillations, a statistical artifact of doing SVD/PCA on smooth data: https://www.biorxiv.org/content/10.1101/2023.06.20.545619v1

You get the same patterns in music: https://pure.uva.nl/ws/files/67945245/000016_2_.pdf


Those plots are so much better. They also give a sense of the variation around the shape, and are less smoothed than in this article.

Also thumbs up for using self-organising maps. Way underappreciated idea!


For those like me who don't know what a self organizing map is

https://en.wikipedia.org/wiki/Self-organizing_map

    A self-organizing map (SOM) or self-organizing feature map (SOFM) is an unsupervised machine learning technique used to produce a low-dimensional (typically two-dimensional) representation of a higher dimensional data set while preserving the topological structure of the data.


More great "diagrams" from mathematician Jean Petitot applying René Thom's morphodynamic models to the study of the canonical formula of myths (Lévy-Strauss).

about 2/3rd in: https://www.persee.fr/docAsPDF/hom_0439-4216_1988_num_28_106...


2016, and the BBC article is 2018.


The reporting on this study is conflating sentiment with plot structure, which misrepresents the study.

See for example, Frankenstein. The sentiment rises slightly during the Creature’s narration to Victor of his circumstances - likely the narration of the French family he was “living”/stowing away with - but that’s certainly not a “rise” in the sense Oedipus rises to noble status. It’s hard to interpret Frankenstein as anything other than the protagonist’s consistent and tragic downfall (riches to rags in this analysis).

Not sure if that’s fundamentally a problem trying to extrapolate plot beats from sentiment alone, or a bit of less than accurate journalism.


If you condense every text passage into a one-dimensional sentiment score, then all texts are going to look like graphs of rising and falling sentiments to you. Nothing unexpected there.

I actually liked the charts, but I didn't see much evidence that there's anything universal at play here that isn't just an artifact of this methodology.


I did like it labelling the Ugly Duckling plot as "complex", relative to Shakespeare and Flaubert and Dante :D


There are other frameworks for storytelling:

- Booker's Seven Basic Plots

- Friedman's Story Plots

- Georges Polti's 36 Dramatic Situations

- Reich's American Narratives Tobias' 20 Plots

- Parker's Story Types

- Classic Story Conflicts

- Classic Story Types

See: https://changingminds.org/disciplines/storytelling/plots/plo...


I think my preference is Northrop Frye’s analysis in “Anatomy of Criticism”, his categories of “mythic”, “romantic”, “high mimetic”, “low mimetic”, “ironic” are particularly useful for analyzing the history of literature from mythic legends and epic poetry up to modern literature and fantasy.

Although this analysis isn’t so much for general plot structure as much as for looking at characters and particularly the main protagonist and their relationship to other characters and the environment of the novel.


Joseph Campbell's The Hero With a Thousand Faces[0] posits the protagonist's journey from "the ordinary world", through numerous adventures and challenges, to eventual redemption or victory, as a universal story template.

[0] https://en.wikipedia.org/wiki/The_Hero_with_a_Thousand_Faces


Don't forget Harvet Ismuth's 42 Essential 3rd Act Twists: https://dresdencodak.com/2009/05/11/42-essential-3rd-act-twi...


If this article interested you then I'd highly recommend checking out TVTropes (https://tvtropes.org/) (trope roughly means any convention of fiction), if you've not already!


Obligatory xkcd: https://xkcd.com/609/


The alt-text even mentions cracked.com. Makes one feel nostalgic.


Vladimir Propp already analysed the basic structural elements of Russian folk tales, down to 31 elements, and drafter how these units formed classic Russian tales.

https://en.wikipedia.org/wiki/Vladimir_Propp There's been further efforts to analyse his work https://link.springer.com/article/10.1007/s12559-015-9338-8


Time to share again my reverse-engineering of Dan Brown's novels![0]

I did this as an exercise almost 10 years ago now (wow!), resulting in my own novel[1], which you could describe as somewhat Dan Brown-like.

[0] https://docs.google.com/spreadsheets/d/1HdlD_tmmm1D0zX1JgXzF...

[1] https://www.amazon.com/dp/B00QPBYGFI


I met Dan Brown at an event where he stated that (Writing the Blockbuster Novel)[https://www.amazon.com/Writing-Blockbuster-Novel-Albert-Zuck...] was the formula he used to write his books.


I wonder how I missed that book. Thanks!


I always felt like we could mathematically define stories but I wasn’t sure how until I learned about Chaos Theory. There are some stories that seem to be very simple geometric, discrete shapes, but I think if you look at any story you’ll find a level of noise and chaos in there somewhere. The more a story approaches real life, the more chaotic it gets. I think Cassavetes is a great example of this. Recently I’ve been thinking about narrative in terms of lossy compression, which can then be connected to our everyday perception, which uses compression to understand reality since we can’t ingest all the information we receive. Narratives differ in how much detail they delete, which then makes each narrative a compression algorithm. Movies with more detail like Cassavetes or the Italian Neorealists are usually considered “artsy” while movies with very little detail are considered “trashy.” This realization helped me talk about movies without prejudice.


Borges: "There are only four stories: the siege of the city, the return home, the quest, and the sacrifice of a God"


I like David Lynch’s approach to storytelling, which mostly doesn’t follow these basic plots:

Accepted into the institute’s Center for Advanced Film Studies in 1970, Lynch studied with the Czechoslovak film maker Frank Daniel, whose course on film analysis shaped his writing and directing habits. “It’s a simple thing he taught me,” says Lynch. “If you want to make a feature film, you get ideas for 70 scenes. Put them on 3-by-5 cards. As soon as you have 70, you have a feature film.” Except that he now dictates to an assistant, Lynch still works this way.


There’s a lot to unpack in the word “scene”. Every scene itself is either a needed part of a story, or a story in and of itself, except that the meaning relies on the context of the whole work.


I don't remember if it was Rogue One or another of the SW franchise, but they actually backed their way into the plot but first picking scenes from other movies and then stitching them together.


This led me to an amazing NYT profile of Lynch from 1990: https://www.nytimes.com/1990/01/14/magazine/a-dark-lens-on-a...


That is maybe because everyone’s life falls into a single basic plot:

They are born, they develop, they die.



I guess you need SVD to figure out that if you have two emotional trajectory states (rise or fall) that these can only toggle (since "rise, rise" is just a bigger "rise" and "fall, fall" is just a bigger "fall"). So you pick the initial emotional state and then toggle however many states are in the story. If you have at most three states then you get 2*3 trajectories.

1. rise 2. fall

3. rise, fall 4. fall, rise

5. rise, fall, rise 6. fall, rise, fall

Sorry, I don't find this particularly insightful. The insight would have been if certain sequence lengths were more important or excluded. Or other parameters about the states.


According to John Gardner (or possibly Dostoyevsky, or Tolstoy, or someone else entirely -- attributions differ), there are only two basic plots:

1. Someone comes to town

2. Someone leaves town


Dostoyevsky and Tolstoy are exemplary here. It's not whether the characters rise or fall, it's _how_. The plots of both Anna Karenina and "Crime and Punishment" seem pretty unoriginal, but my gosh, the journey while you read them is incredible.


Quote Investigator can't track down the original words but the earliest attribution is to Gardner: https://quoteinvestigator.com/2015/05/06/two-plots/


Change is a catalyst. And all stories are stories of change.


You can shoehorn “all stories” into all sorts of frameworks. I personally find this really limits what stories can be and leads to decade after decade of lazy writing done by people who believed these commentators. Eg, heroes journey is actually a bunch of techniques, not just one, and some stories don’t even have characters at all


I think all stories will find themselves in these patterns, but it's important to let the story find the pattern rather than the other way around.

I doubt that the person who came up with the story of Icarus was attempting to do RISE, FALL explicitly.

Although, it's also important to be open to new paradigms. If you believe that there are only two stories: RISE and FALL, you can find yourself stretching metaphors to their breaking point to make stories fit in a box.


Yes, but... so what? We see this type of analysis often, but I never see where it gets used to help understand something new, or make anything more clear. How do stories written with each of those plots differ? Do they vary in popularity over time? Why only those six, why not Rise-Fall-Rise-Fall or Fall-Rise-Fall-Rise? Etc.


https://en.wikipedia.org/wiki/The_Seven_Basic_Plots

Even 7 isn't quite correct, technically there are loads of others but they aren't successful plots.


It could be even more simplified to define stories as a series of SETUP, PAYOFF.

- Person becomes rich/poor: Setup to what is about to happen

- Person becomes poor/rich again: Pay-off

Albeit, it has equally low value on what actually makes it a story...


Vonnegut was right! Here’s a great lecture where he explains the 8 shapes of stories.

https://www.youtube.com/watch?v=oP3c1h8v2ZQ


Good luck fitting the Game of Thrones book(s) into this.

Also, in general a narrative suspense/drama wouldn’t fall into this, for example “twelve angry men”.


Is this the type of thing that warrants research? I think that we're able to discern these types on our own pretty well


I suppose it's these simplified views of storytelling that brought us Netflix quality entertainment.


Sounds like a paraphrasing of Kurt Vonneguts 'Shape of a story' presentation.


For movies it is seven types:

1. Man vs Man.

2. Man vs Dog.

3. Dog vs Zombie.

4. James Bond.

5. Stories of Kings and Lords.

6. Women Over 50 Finding Themselves After Divorce.

7. Car Commercial.


on a deeper level, every good story has contrast.


In summary, resorted:

1. Rags to riches: RISE

2. Riches to rags: FALL

--

3. Icarus: RISE, FALL

6. Man in a hole: FALL, RISE

--

4. Oedipus: FALL, RISE, FALL

5. Cinderella: RISE, FALL, RISE

--

So every story can be framed by putting brackets somewhere in this sequence: RISE, FALL, RISE, FALL

Now....what to do with this highly-abstracted level of information...?

It's like saying all music in the world consists of some form of "lower note, higher note, lower note, higher note"


Yup, and these are certainly not "plots" by any definition of the word.

They're one aspect of story, along with a hundred others.

> Now....what to do with this highly-abstracted level of information...?

If you're a reader/watcher, nothing.

If you're a writer... ensure that it's clear. So much of writing is about showing your draft to a friend/teacher/etc., finding out that they didn't "get it" because what you wrote was confusing, and rewriting to make the thing clearer.

If somebody reads your script and they're like, well the protagonist clearly FELL but then the only KINDA ROSE but KINDA STAYED THERE and I don't get what the point is? Are you trying to say their efforts were worthwhile or futile? And then the writer is going to do much better if they change the story to become clearly either "FALL" or "FALL, RISE" -- instead of a muddy in-between.

Also: this pattern is fractal. Every scene is generally either a FALL or RISE for the main characters in it (if one character rises, the other falls). And then, depending on how you define them, each act is similarly a reversal (especially e.g. in a 5-act TV episode). This is what keeps story interesting. But over the course of all of these ups and downs, the author needs to decide what the biggest pattern is going to be.


> If somebody reads your script and they're like, well the protagonist clearly FELL but then the only KINDA ROSE but KINDA STAYED THERE and I don't get what the point is? Are you trying to say their efforts were worthwhile or futile? And then the writer is going to do much better if they change the story to become clearly either "FALL" or "FALL, RISE" -- instead of a muddy in-between.

Plenty of good stories have ambiguous endings. That’s actually one of the issues with this article - even when things are abstracted to the point of absurdity like they are here, it still doesn’t cover all stories.


> Plenty of good stories have ambiguous endings.

But they still generally follow the pattern. Probably the most common "ambiguous" ending is where protagonist gets what the WANT (seems like a RISE) but not what they actually NEED (so it's actually a FALL). Or vice-versa, the don't get what they WANT (seems like a FALL), but they do get what they NEED (so actually a RISE).

So the "ambiguous" ending is not usually about not following RISE/FALL structure -- it's leaving you to argue about whether the RISE/FALL was good or bad. (E.g. he lost his Wall St job but gained a family, he conquered the mountain but his best friends died along the way.)


Are you a writer or are you guessing?

Craig Mazin and John August have often argued on Scriptnotes[1] that this type of structural perspective can be useful for analysis, but not when writing. For example, you can often think of a movie as having three acts, but knowing that doesn’t help you write the movie. It’s more a symptom of good writing than a recipe.

1: https://scriptnotes.net/


I'm adjacent to it.

One of the first things you usually do as part of planning a season of a dramatic TV show is to figure out the arc of each character, up or down. You can't get anywhere unless you know where you're going. And then yes, it's similarly an extremely conscious part of the writing process at the episode level, act level, and scene level. It's very much "recipe" in that sense -- but that's not to take away from any of the creativity. It's just the basic structure for creativity, the frame.

Now, of course, it's recipe because of lots of trial and error in writing stories and seeing what stories worked and which didn't. If you're writing something shorter like a movie, you may very well implement the pattern on your own through intuition. But you also may very well not, and the script will have problems, and rewriting will be a lot more successful if you're aware of structure and not just relying on intuition.


But then you have books of rules like "Save the Cat!" that identify a lot of beats that you do usually find in screenplays. Maybe not all the ones are in every script, but it's uncanny how the midpoint is almost always as described to where you can pause it when you feel the beat and you're halfway through the film.


Sounds like a good subject of Fourier analysis.


It's a fractal. You don't have complete freedom across the entire space of "stories" to just pick one; most such randomly selected stories would be just random things that happen for no in-story reason, then the putative story simply ends. A top-level RISE/FALL/RISE structure can add interest to the story. But then within that there is a fractal of other options, because if that's literally all your story has, once again it's not really that interesting. You can keep descending down gather up more and more detail.

Sometimes someone discovers a little undiscovered tributary, or does a really good job with something like Memento.

On the one hand, there is order to the space more than just "things happen, then they stop". On the other hand, like a good fractal, you really can't classify the whole space of stories with metaphorical rectangles on the diagram. No matter what rectangle you draw around romance, you're going to get every other genre mixed in at least a little bit, and so on for all the other things you can try to draw.

As one gets more experienced with stories, one's interest tends to "zoom in". A kid may be satisfied with a simple hero story. As you get older, the hero story itself may be of little interest, but you are more interested in the fractal bits within it; the quality of the characterization, twists, plays on expectation. Later you may find interest in underlying philosophical themes and so on. Or, you can miss these finer details and become exhausted with stories believing you've seen them all, which despite how it may sound like I'm characterizing it here, isn't necessarily the worst thing ever. There's all sorts of things in the world and you can't become master connoisseurs of them all.


> So every story can be framed by putting brackets somewhere in this sequence: RISE, FALL, RISE, FALL

Even more abstract: CONFLICT, CONFLICT, CONFLICT, …

Perhaps any story not rooted in conflicts would be horribly boring and so not re-told?


You can make a really boring series of events interesting, but without a payoff it doesn’t feel like it’s worth it.

Example:

I woke up this morning really really early… I checked my charging Apple Watch and noticed it was 2:15. I got up and noticed how quiet it was… like incredibly quiet. I put on my shoes and walked toward the stairs, and it was eerie how little sound there was.

When I got to the bottom of the stairs, it was still even quieter.

—-

This is literally what happened this morning, but the mention of “quiet” puts a question in your head: why is the author mentioning how quiet it is!? Isn’t it always quiet at 2am??

That will get you to invest in the story and there’s no conflict. But then if the ending was “yeah it’s just always quiet” you wouldn’t have learned anything new, and there’s no reason to share / recommend / reread the story.

Basically this is why JJ abrahms movies are fun to watch until they’re over.


Your episode does have an implicit conflict, though, and it's precisely the one you're saying: you have set up an implicit conflict between your auditory perception and expectation.

(Actually, there's another implicit conflict in there which I would argue is perhaps even more important: why did you wake up so early, and why did you choose to go down the stairs instead of back to sleep?)


So what makes the story interesting here is the encounter of something unexpected.

Reminds me of That's interesting! (1971) (https://news.ycombinator.com/item?id=36156400)


If you carried on this story though, without any conflict (note: I agree with sibling comment that there is actually implicit conflict between actual vs expectations), wouldn’t it be incredibly boring or lead to a “so why are you retelling this?” response?


> Even more abstract: CONFLICT, CONFLICT, CONFLICT, …

How about this: alternating sequences of THINGS CHANGING and THINGS NOT CHANGING.


Another way to explain it is that popular stories are a chain of buildups and payoffs for the reader. The core of it is that unlikeable characters get misfortune and likeable characters get fortune.

Amateur authors and stories based on real life events can break the mold. I think that's why both can be quite popular.


> The core of it is that unlikeable characters get misfortune and likeable characters get fortune.

I don't think that's really true, though. Tragedies are a very popular type of story.

Indeed - a big part of the reason for the success of "A Game of Thrones" is because they violated genre tropes and killed off likable main characters.


And to reinforce your point, the later seasons suffered (in part!) due to the plot armor that suddenly enveloped the main characters.

Among other reasons.


Explain don quixote then.


>It's like saying all music in the world consists of some form of "lower note, higher note, lower note, higher note"

https://en.wikipedia.org/wiki/Parsons_code


OK, but that's actually useful for information retrieval.


> It's like saying all music in the world consists of some form of "lower note, higher note, lower note, higher note"

Tell it to John Cage.


John Cage is not all the music in the world.


I think you're mixing up your universal and existential qualifiers.


Every story? Now there's a false title. _Project Lawful_ for example has a sufficiently complex plot with enough opposed character perspectives that you'd have to really work to hammer it into this pigeonhole. Maybe one particular character arc could be said to have this structure.

More generally, any story that changes between character perspectives and evolves in who holds narrative force will have a hard time being analyzed this way.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: