The Claudyssey: A line-for-line translation of Homer's Odyssey by Claude Fable 5
theclaudyssey.com12,107 line-for-line translation of the A.T. Murray Greek text (Loeb, 1919) by Claude Fable. Includes line notes and an index. HTML, epub, pdf, and YouTube Audio (by gpt4o-TTS). A dual-agent review pass was done on the entire work with a third "adjudication" agent reconciling their findings. Reviewer agnets were gpt-5.6-sol high reasoning.
It seems the traffic has made the site inaccessible. I'm enjoying reading the first page that's still in my browser.
why?
I love the poem and wanted a modern translation with annotations and index, etc., to be freely available to anyone who wanted to read it. I had a sense that Fable would do a really good job. I am pleased with how it came out. The review agents didn't change a single word of the translation, only some of the notes.
Every translator is opinionated in some way. I can see the appeal in having something like this as a neutral comparison.
LLMs are not neutral, either. They don't have "opinions", per se, as they don't want anything, but they do have biases baked into their training.
One very simple benefit is that it's unencumbered by copyright.
I do value the work of human translators; when they're really in harmony with the original work, they can be fantastic. But they can also be disastrous, like Alfred Birnbaum's translations of Murakami, which radically reshapes the protagonist and threw out nearly 20% of the original work.
If the Murakami was undertranslated then how will having more translations that crowd out the potential revenue of future translators help produce better quality translations?
The problem is that (we have every reason to believe) models like Claude were trained on existing translations. Now they're being used to generate things that are direct replacements for those translations in the marketplace.
This is basically an anti-AI author or lawyer's dream case.
And, seeing this uncanny-valley-of-slop output, I kind of hope they go for it.
How does it compare to previous (human) translations, what is the regurgitation level? In any case, lots of em dashes, so mission accomplished to bring it up to contemporary standards.
https://theclaudyssey.com/read/book-01.html#n1
The first line is a crappy mishmash of Lattimore and Fagles.
Lattimore: >Tell me, Muse, of the man of many ways, who was driven far journeys, after he had sacked Troy's sacred citadel.
Fagles: >Sing to me of the man, Muse, the man of twists and turns driven time and again off course, once he had plundered the hallowed heights of Troy.
Claude: >Tell me the man, Muse — the man of many turnings, who was driven wandering far, once he had sacked Troy's holy citadel
Combining "man of many ways" and "man of twists and turns" into "man of many turnings" is just dreck.
"man of many turnings", is a more literal translation of polytropon than Lattimore and Fagles, and and in fact the whole line is better analysed as an original, literal translation from common dictionary glosses than a "crappy mishmash". Many people who want to give a literal gloss say "man of many turns" or "many-turned man".
> ἄνδρα μοι ἔννεπε, μοῦσα, πολύτροπον, ὃς μάλα πολλὰ πλάγχθη, ἐπεὶ Τροίης ἱερὸν πτολίεθρον ἔπερσεν·
man(acc.) me(dat.) tell(pres. imp.), Muse(voc.), many-turn(acc.), who(nom.) very(adv.) far(acc.) wandered(aor. 3sg. m.p., but note active sense of "to drive off"), after(adv.) Troy(acc.) holy(acc.) city(acc.) (he) sack(aor. 3sg.)
Glosses my own, text from https://scaife.perseus.org/reader/urn:cts:greekLit:tlg0012.t...
Claude repeats man twice like Fagles, despite that not being in the Greek. Using driven and wandering is a weird repetition rather than just using "made to wander" which is a more natural way of directly translating the passive. Claude uses the "driven" in both Fagles and Lattimore. It also drops the "very" for no apparent reason (also not present in Fagles or Lattimore). Claude uses "once he had sacked Troy's holy citadel" mirroring Fagles's phasing, when you could just translate it "after he sacked the holy city of Troy". It also uses citadel rather than city, copying Lattimore.
Vis-à-vis polytropos, I just think that "man of many-turnings" sounds god-awful, but the similarity was mostly the fact it uses the construction "man of x" like Fagles and Lattimore rather than a normal adjective like in the Greek.
(Note: I'm sorry, I'm now realizing that my glosses were really bad, they provide a common basic meaning but that isn't always the sense that they have in the text, think "to bear arms" and "to bear a burden" in English having the same root and related meanings but different connotations)
> Claude repeats man twice like Fagles, despite that not being in the Greek.
This style is not unique to Fagles. In Greek poetry modifying words (polytropon, for preceding andra ...) can be in apparently insensible locations not found in prose and speech, but accepted in poetry for the sake of meter and poetic style. In Greek, this heritage makes it acceptable, and the marked cases provide redundancy that allows the very flexible poetic word order. Here, there is a choice to capture the sense putting "man" before Muse as the focus of the sentence as in the Greek, and then repeat "the man" again before "of many turnings" because it is otherwise more ambiguous to just say "of many turnings" after "Muse".
> Using driven and wandering is a weird repetition
I don't think it's negatively weird, it's a choice that tries to capture the range of senses that a Greek might associate with it.
> It also drops the "very" for no apparent reason (also not present in Fagles or Lattimore)
Lattimore doesn't really capture this. Fagles captures this by using a more expressive "time and again" for "far"/"many".
Maybe one can argue that Claude choses to reflect extremity in this by giving both "driven" and "wandering".
> once
This is a very understandable gloss that I had chosen to supply "after" instead. It is not an innovation by Fagles, but a choice Fagles from a small range of valid common choices.
> It also uses citadel rather than city
I think citadel is altogether an appropriate general translation as well. It is described as "holy", and the Greek is a Homeric "ptoliethron" that everyday Attic/Ionic speakers would consider an archaic word, where the everyday word is the "polis" that you and I are familiar with.
If your opinion is that Claude does not have a consistent editorial voice that is preferable for translation, sure I agree. If your opinion is that it translates while being aware of and perhaps taking cues from prior translations, sure I agree, with the caveat that it also sees dictionary definitions, the Homeric Greek itself, common ways Homeric is translated, and more literal glosses, and it seems to me in fact to take more influence from these (at least, just by looking at this first line). But it is difficult for me to agree that it just mashes up two prior translations.
As for the criticism of editorial voice I wonder if it can be solved by a matter of prompting with expertise.
The sort of sad but interesting thing here is that while this claims to be a more direct translation, Claude models were certainly trained on all of the pre-existing translations and this is likely going to directly influence the output in the ways you've outlined here. I'm not sure any modern LLM could honestly be claimed to produce a translation that doesn't draw on these pre-existing sources.
You're making two separable criticisms: (1) that Fable's translation is a mere "mishmash of Lattimore and Fagles" and (2) that it's a crappy one.
I'm not seeing, from what you quote, how this is more "mishmash" than "translation of the same underlying text as them".
As for whether it's crappy, that's a pretty subjective matter; for what little it's worth "of many turnings" doesn't seem particularly dreck-y to me; it seems in fact a rather literal rendering of "polytropos" and -- as Fable's footnote says -- one that maintains some of the ambiguity between "one whose paths have turned many ways" and "one with a twisty-turny sort of disposition". I am no scholar of ancient Greek and there's no reason why anyone should pay particular regard to my tastes on this; but what is asserted without evidence may be controverted without evidence.
(The one thing in those three lines that I would describe as dreck is "driven far journeys" which to me seems barely grammatical. "Secret" seems like a weird choice, too.)
>driven far journeys" which to me seems barely grammatical
"Tell me the man" and "of many turnings" sound far, far worse, not to mention the egregious em-dash. Obviously there's no way to really _prove_ what is good taste, but I would be quite surprised to hear you really think that Lattimore is worse than Claude here.
>"Secret" seems like a weird choice, too
This seems to be an OCR error from the source I copied online! Digging up a real copy, it's sacred rather than secret! I've since corrected this.
I agree that "tell me the man" feels weird. I still don't see what's wrong with "of many turnings". To me, "driven far journeys" is a bit worse than "tell me the man" (and quite similar in kind), and much worse than "of many turnings". But maybe I just have terrible taste.
(I'm glad to hear that "secret" wasn't real. It seemed quite inexplicable to me.)
I rather like "man of many turnings", on the page.
The problem I have with Claude's version - same as Lattimore, in fact - is that it's really hard to read aloud; the rhythm just doesn't hit right. Fagles is the best for that, though I'm aware of the occasional liberties he takes to achieve a high poetic standard in English. There are 19th c and earlier translations that wander even further; I love the balance that Fagles strikes.
I've got Wilson downstairs, though I've not read it yet.
Your comment persuaded me that just explaining my prompt methodology is insufficient.
With Claude's help, I ran some tests against nine different translations including Peter Green's line-matched translation, the closest method-matched human control. 36 human-vs-human pairings as controls.
Murray (the published translation of the Greek, Loeb source) and Lattimore affinities are genuinely elevated. Fagles is modest but not totally absent. The test can't exclude 4-6-word mosaic composition, and doesn't measure interpretive influence.
The longest shared passage in the 12,107 lines is 29 words. Long-fragment assembly is strongly disfavored by synthetic-splice calibration. The translation resembles Green as much as it resembles Murray, or three times as strongly as it resembles Fagles' famous one. That strikes me as being in the opposite direction from what training data-regurgitation would predict.
https://theclaudyssey.com/independence.html
Methodological corrections welcomed
The only thing Fable was given was a clean of copy of the ancient Greek. This is not a bastardization of other translations. I appreciate and value the critical feedback in any case (truly).
I find that Fagles is very majestic -wonderful, really- but he's also very liberal (poetically) and definitely did not limit himself to the same number of lines as Homer as this project did.
> The only thing Fable was given was a clean of copy of the ancient Greek. This is not a bastardization of other translations
FWIW I gave GPT-5.5 two lines of William Cowper's translation (long out of copyright to avoid any filters) without web search and not only was it able to identify it, it was able to continue it verbatim for roughly 30 lines at which point it went off course, but only by skipping like two books ahead. If I had supplied the original text as an anchor it presumably could have kept going.
I didn't try it with Fable because I'm lazy, but all these translations (presumably the ones within copyright too, given the lawsuits) are definitely well known by these models.
As another comment here points out, Fable likely had multiple translations, academic literature about the text, and maybe the original as well, in its training set. So even if only the original Greek was given to it as input it probably leveraged prior art when translating it.
> The only thing Fable was given was a clean of copy of the ancient Greek
Ok, but this clearly combines a few translations line by line if you work through it.
> The only thing Fable was given was a clean of copy of the Ancient Greek.
…and all of its training data.
You think Anthropic didn't train on some of the most famous translations of the Odyssey? Really?
No. I am saying that I had Fable do a literal word for word translation, and then did two independent review pass translations with models of different lineages and then had another adjudicate it. I am just saying what I did, that is all. It occurs to me as I am typing this that a human translator working on an Odyssey translation today would have similar "prior training," culturally. It's one of two of our oldest stories.
Yes and if a human translator produced this work, it would be viewed as near plagiarism of Lattimore and Fagles. Compare this to Wilson. Wilson had surely read Lattimore and Fagles and yet produced something genuinely different (as much as I don't personally care for her translation). Claude just regurgitated others' half-digested work.
Emily Wilson's translation is way better, more modern and adequate to present-day readers and, probably more context aware.
Maybe it's just me, but "modern" and "context-aware" are the two things I most want to avoid when seeking out a translation of an ancient text.
There's no such thing as a "timeless" translation - if you read something written by Victorians you'll get all sorts of overlaid assumptions and biases, just ones that are unfamiliar
For me, that unfamiliarity is precisely the point of reading something ancient.
Then you need to read the original. A translator is a product of their time.
They didn't say they wanted a timeless translation.
Continually revising a text to be modern and context-aware would indeed make it timeless.
I love Wilson's work and take it as inspiration. Incidentally, I spotted an error in the first printing of her Odyssey that the publisher acknowledged on twitter and fixed in subsequent printings. I am not a professional, just an obsessive with a special interest. In a spirit of intellectual honesty, the fact that she is reportedly re-translating The Odyssey again is an indication that one of perfect adequacy might be elusive.
I find it to be cross between Richmond Lattimore and Ian Johnston. There are line-level annotations that discuss translation choices and a hyperlinked index w/ Latinate pronunciations was also built from scratch.
This page provides a great overview of several popular translations of The Odyssey. It lists line-by-line translations as well free translations: https://readtheodyssey.org/best-odyssey-translation
It's hard to take that page seriously when it doesn't include the marvellous translation by Peter Green.
This is a useful marker of progress in translation, not a replacement for literary translators.
Six years ago I fed Dante's Inferno to AWS translation. It was...not good[1]. Maybe its time to try again and see how things have changed.
[1] https://github.com/ffx271/dante-inferno/blob/master/english/...
> marker of progress
Unfortunately it's just a mix of a few existing translations. Elsewhere there's an explanation showing this line by line.
Read this as "Claudussy" at first.
This could allow a very opinionated and tailored translation, which I think is the appeal.
That said, I'd like to see the prompt(s) having such an effect on the subjective choices for a project like this.
Probably this will be done more and more often for works which don't have translation in some language because of low interest. E.g. old manuscripts, light novels, fan fiction etc.
Having recently read Wilson's translation, I was immediately put off at the introduction and leapt to the conclusion it was bad. It just doesn't sound anywhere near as good to my ear.
However, there may be more to it than I thought. No admittedly the following is ChatGPT's opinion, but I think it's valid and made me think again:
> The second is the better translation as English poetry; the first (Claude's) is the better translation if you want to see the Greek more directly.
> If I were studying Homer, I'd strongly prefer the first. It continually reminds you that you're reading an ancient poem whose concepts don't map neatly onto modern English.
Translation is an art form that is rapidly becoming lost, not because AI is doing it, but because there are too many translators who have apparently been trained (or likely, were never trained in it as a formal study at all) that it should be written in a way that emphasizes the translation as a standalone work more than anything else.
A good translation should attempt to communicate what the author meant, as closely as possible, even if there is no equivalent in the target language.
A good translation should be beautiful all on its own.
It's the balance of the 2 that makes a translation successful.
If you erase the author's true meaning just because you don't want to slow the reader down in having to take a moment to read an explanatory footnote if they're confused, you have failed to translate it, and have likely editorialized it.
Why do many English-speaking people know 'wabisabi' as a concept? Because someone bothered to explain it to them, and they bothered to listen. That act enriches the reader.
And don't even get me started on "localization".
I don't remember so many emdashes in the original.
They called them mudashes back then.
I'm just curious why you'd put "Claud[e]" in the name of this project. Is it an Anthropic-sponsored one? If so, may you please disclose it? Additionally, I think it puts a date to a wonderful project that otherwise could be very timeless. In 5 years we'll probably be like, "what's Claude?" Anyway, really cool work and thanks for sharing!
I truly doubt it will be timeless. There are already a few excellent English translations of the Odyssey.
It is just a domain for the project's github page. All the files, YouTube uploads, etc., are The Odyssey or Homer's Odyssey.
In 5 years we'll probably have locally run AI which could do something like this in a millisecond.
This took a lot more tokens and time than I expected it to. Even before the review passes.
A language model does language model things? That might actually be... an appropriate use of the technology!
What is the point of this?
Now have it make it into a movie.
Christopher Nolan beat you to it
I mean let Claude make it into a movie. A task which is quite possible.
Ok, but why would we want that instead of a movie made by people?
Christopher Nolan beat you to it
Oh, then Elon beat you to it.