Writing done entirely by AIs is currently pretty awful, but I'm sure that it will get better over time. There was a prof on Twitter (can't find the tweet now) who assigned his class to generate and then critique an essay written by ChatGPT and it caused all of them to stop using it.
The problem is that moderation doesn't scale. Newspapers, TV, radio, film all evolved to have or to conform to some kind of standards and practices department, which played well with advertiser needs. The issue with contemporary media companies is that it's not practical to do that at Internet scale -- see all of the routine stories about moderation teams at Facebook, for instance.
There are also large, structural similarities across the plays and poems that suggest a single authorship; that 'not very good' poetry has interests and stylistic tics in common with the later, more universally revered work. It's also relatively recently (1800s on) that the plays gained that the reputation that they did. The argument that William Shakespeare couldn't have written the plays is one that is extremely recent, historically-speaking, and it relies on snobbish negatives ('A son of a glove maker who had an excellent education grounded in the classics and spent his life in the theater couldn't have written the plays') rather than any affirmative evidence against. We also have text of contemporaries that are generally trustworthy sources describing his work (and offering criticisms). It's not the exciting answer, but it probably was William Shakespeare that wrote the works in question.
Hey this is a former coworker of yours from GU back in the day. Nice post! Hope you're doing well and congrats on the IIe. I grew up coding on one of these as well. :)
Amateur Shakespearean here. I've often wondered why these statistical approaches don't look for stylistic signatures. For instance, several plays are linked by their use of doubles (in Hamlet, this scales from "too, too solid flesh," up to the running joke on the interchangeability of Rosencrantz and Guildenstern). Shakespeare also seems to be especially fond of puns on "bear" and "born" -- the "to be or not to be" monologue has three or four of these. There are also ways in which scenes repeat their designs -- 1.2 of Hamlet is similar in some ways to 3.2 of Julius Caesar, etc.
I imagine two things that make simple analysis harder.
1) once an author has a sufficient library, hiring a team to emulate them is easier (Tom Clancy, Hanz Zimmer.)
2) if in a collaboration, a partner could write a skeleton or first draft, and the main author could edit and fill in the cracks. Conversely they could write scattered sentences and have their partner flesh out filler.
kennings, as this game is called in anglo-saxon, aren’t filler per se. Though they are also a memory aide (for the poet and the audience), they are primarily part of the way the skill of a scop (bard/epic poet) was judged was by how cleverly he could choose the right kenning for the mood and meter of where in the poem he was.
A saxon kenning for poet is “story-weaver”, and that’s actually very apt.
Yes, it reminds me of the "memory palace" trick that's taught for remembering e.g. numbers by associating them with a series of images. And similar tricks with "themes", audible or visual, are still used for particular characters in some modern media.
yes. and for long works like Beowulf that were told episodically, the repetition of events serves the same purpose as the 90-second “last time on dragon ball Z...” recaps of fight hilights.
There's probably a lot of literature out there. I like the scholarly essays that accompany Chickering's translation of Beowulf -- maybe start there and follow the reference trail if you get sucked in.
Wine-colored sea is a different story, I think. I've seen other languages also describe the sea as red, orange, etc. This is more of a situation with color naming: cultures separate colors on the spectrum into named categories in a more or less consistent way (e.g. blue and green come apart towards the end, and some languages still have the two merged). Red is one of the first colors to be named.
yes, that could be true. but the point still stands, that the repetition of certain phrases, even if they understood colours differently than modern audiences, still served the purpose of aideing memory and familiarity among the storyteller and the audience.
The "bard" usually also had an instrument, so it's highly likely that if he was having trouble composing/recalling the next few stanzas, he may have stalled for time by performing an interlude on his lute.
This is how the books of the Odyssey are structured:
TROY
----
1. Kikones (innocent city is ambushed)
2. Lotus Eaters (temptation, delay)
3. Cyclops (monster)
4. Aeolus (greedy crew delays journey)
5. Laestrygonians (monster)
6. Circe (temptation)
7. HADES
8. Sirens (temptation)
9. Scylla (monster)
10. Cattle/Sun (greedy crew delays journey)
11. Charybdis (monster)
12. Calypso (temptation, delay)
13. Phaecia (innocent city is ambushed)
----
ITHACA
The episodes make a perfect reflection across the Hades episode; Circe and Sirens are both forms of temptation, Scylla and Laestrygonians are both monsters, etc. Also notice the smaller reflections inside of this large one: Scylla and Charybdis are similar monsters in the same way that the Cyclops is similar to the Laestrygonians.
It's perhaps easier to remember a number like 12140904121 than it is to remember a series of random digits.
This is called a Chiastic structure (from the greek letter chi which looks like an X - i.e., the shape the chiastic structure makes if you indent each succeeding level).
Charybdis (a whirpool) and Scylla (a multi-headed monster) was the same episode in Homer's Odyssey. That was the whole point of the episode: Ulysses had to decide whether to risk his entire ship and crew or definitely sacrificing a few for guaranteed passage of the rest. The classic trolley dilemma for an audience of bronze-age seafarers.
Or perhaps I read a completely different "Odyssey" by a different "Homer"?
Your memory is likely faulty, or you may have read an abridged version.
Odysseus first has to choose between Charybdis and Scylla. He and his (surviving) crew pass Scylla and reach the isle of the sun-god. There, the crew kill and eat the cattle, angering the gods. When they depart, a storm pushes their foundering ship back to Charybdis, which only Odysseus escapes, drifting while clinging to the remains of his ship's mast.
You might have? Odysseus chooses Scylla, but then is punished by Zeus after his crew slaughters the Oxen of the Sun; his ship is destroyed and he is sent on its timbers back to Charybdis.
I wonder if there's a good term for when people confuse what an element in a popular work is most famous for for the actual details of that work. Call it the "Play it again, Sam" fallacy.
I'm unsure of your point here--isn't memorizing the text itself much harder than knowing this kind of high level structure, which I venture many people had/have no conscious awareness of anyway?
I don't know about this particular case, but I have some (weak, anecdotal) evidence that at least sometimes the high-level structure is harder to remember robustly than the details.
I saw an amateur pianist sit down to play (a solo piano version of) Gershwin's "Rhapsody in Blue". He took from his pocket a bit of paper containing, not the actual score of the piece, but a lot of things of the form "Starts in Bb major", "Modulates to F major", "twiddly bit", and so on. Having put that in front of him, he proceeded to play the whole piece. Correctly, so far as I know.
Caveats:
1. n=1.
2. "Rhapsody in Blue" is probably an extreme example of a piece where this sort of thing would be useful. Here's a wonderful quotation from Leonard Bernstein: "The Rhapsody is not a composition at all. It's a string of separate paragraphs stuck together. The themes are terrific, inspired, God-given. I don't think there has been such an inspired melodist on this earth since Tchaikovsky. But if you want to speak of a composer, that's another matter. Your Rhapsody in Blue is not a real composition in the sense that whatever happens in it must seem inevitable. You can cut parts of it without affecting the whole. You can remove any of these stuck-together sections and the piece still goes on as bravely as before. It can be a five-minute piece or a twelve-minute piece. And in fact, all these things are being done to it every day. And it's still the Rhapsody in Blue."
3. This was years and years ago -- either 1987 or 1988 -- and my memory may be unreliable.
4. The amateur pianist in question was a teenager; maybe memory develops and/or degenerates with age in ways that would make this less relevant for adults.
5. Music versus poetry.
But it seems plausible to me. I have never memorized anything within three orders of magnitude of the length of the Odyssey or the Iliad, but e.g. when I was a stupid child I learned ~100 digits of pi, I learned them in 10-digit chunks, and right now I think I can remember what all the chunks were but I wouldn't want to place any bets on getting them in the right order. Of course digits of pi are more or less random and epic poems have narrative structure -- but making use of that narrative structure is exactly what GP is talking about doing.
I think that the oration of these poems might be akin to jazz playing, where a musician knows the broad, basic structure of a work and fills it in with improvisation.
So then, “the” canonical version of eg the Iliad is just one version that was recorded, and it would have actually varied across the different retellings?
Also akin to freestyle rap, where the storyteller knows the beats of the tale they're telling, and comes up with a particular rhyme on-the-fly for each sentence by following the structure of the verse.
How very daring, comparing dear homer to freestyle rap.
I don't disagree, but have to say freestyle is ideally song-writing in real time. Personally, I think this means the opposite of recitation. In practice, it is just-in-time and thus very repetetive (I'm happy I managed to make this comment HN on-topic! Thoughts?).
Therefore, it's as variable as any writing, though the constraints may require certain techniques and methods. I don't know, there's a reason freestyle battles aren't hugely popular. Many battle shows are pre-written. One can easily tell the difference. I won't say that's cheating, but will judge it according to expectations.
It's practically no different from regular discourse, ranging from repetitive, casual or formulaic small talk over thought out hacker news comments--I made a draft of this one in my head and am anxious I forgot something, or add too much--or a team meeting with the boss at work, up to legal procedures where the "oral" statements are frequently read out from paper, drafted by a team of experts, or even just submitted as briefs in congress to be bundled with the plenar protocol.
The only difference is the meter, which is not even rigidly up-held in modern poetry slams, nevertheless prosody is an important part of natural language (for lack of a better word).
Prosody in writing should be an interesting topic. Perhaps that's why my writing tends to be hard to read, as I'm a second language learner without much oral experience these days.
> Personally, I think this means the opposite of recitation.
Possibly true, but my point was that Epic Poems would not have been transmitted merely by recitation of exact learned words, but by a rhapsodist learning the story and internalizing it.
Then it would be conveyed as a mixture of repeated learned verses and freestyle improvisation of parts of the story; that would explain how completely new sections appear in the different versions of epic poems.
