The Value of Differences: Jennifer Lindsay on Noticing Translation | Hacker News Reader

The Value of Differences: Jennifer Lindsay on Noticing Translation

sydneyreviewofbooks.com

31 pointsThevet26 comments

As a French speaker who has done quite a bit of paid translating between French and English (so an uneducated translator rather than an amateur one I guess...), I have found that a lot of translation (in non-fiction, but also sometimes in fiction) have a feel to them. You might not care or notice at a glance, but the text does feel unmistakably translated. That is a constant reminder of that fact that translation is a job that takes real skill, not just knowledge of both language.

My layman take on this is I don’t really want to read a book where “There isn't a paragraph or turn of phrase which feels like it didn't originate in English”. I want the original language to punch through a bit and be celebrated especially with idioms.

Don't you want to appreciate the book at its best, as it was appreciated by readers who read it in its original language? Why demand a compromised work? If you want to soak up untranslated idioms, you can just learn another language.

"just learn another language."

Lol Just learn many other languages and cultural idioms that are sometimes regional and dense with meaning?

The reason to do it is precisely because it is hard.

I sincerely believe everyone should learn at least one other language. At least make the effort, you'll learn a lot in the process and it can be quite fun. It also enriches your understanding of your primary language once you've seen how other languages express particular concepts.

Absolutely, but there are good books written in many different languages.

Also, as I mentioned, some idioms have regional meaning which raises the complexity greatly

fun fact: Gabriel Garcia Marquez said in public that he believed the english translation of "100 anos de soledad" was better than his spanish original.

A book with that title would surely be very different indeed.

Sorry but translated works do not mean compromised work. Don’t disrespect translators like that.

Second, Learning a language is not a binary variable. It takes decades to master a language.

Finally, Your ability to appreciate a book is both a function of the text and your ability to comprehend the text. A translated book will give you better experience than the book in a language you are unfamiliar with.

"I want the original language to punch through a bit and be celebrated especially with idioms." is asking for a compromised, incompletely translated work, no?

Not at all. Take the example of the Bible. Most people don't want a Bible with everything translated to perfectly modern English. They have expectations of the work that are better served by using some archaic/historical terms, or even leaving terms like the tetragrammaton intact.

Well if I really wanted to appreciate the book at its best I’d learn the language and read the original. But that has its obvious impracticalities.

I think it’s a trade off then. I’d prefer authenticity, richness and nuance over accessibility. It is just preference. Did you prefer the modern accessible versions of Shakespeare or the originals in their Elizabethan glory?

fun experience: I've read two english translations of The Glass Bead Game by Hermann Hesse. The first one honored the general vibe of its Germanic ancestor, and made it extremely hard to read. The second one had its own vibe, not uniquely English nor obviously German, and was a masterpiece.

I'm not even sure what this means for a language to punch through in a way the author or reader might find desirable. Do you have an example?

I'm not the original commenter but I think some of the personal epithets in homer just smack you with your distance from the original. "White armed andromache," "mouse god apollo", "ox-eyed hera" and of course the famous polytropos odysseus etc.

These have as much to do with ancient greek culture and maybe mindset as language per se. But even still, the choice is to elide them or let them be strange. There's no real way to carry the meaning through to english without it being striking, calling attention to the original.

I'm as big fan of "rose-fingered dawn" (forgive my lack of imagination) as the next nerd, but that always struck me as less about greek itself and more about the evolution of western language before the widespread use of writing. Little of the rhythm and rhyme can translate directly; it's more of a conceptual translation, which could just as easily be done with Mandarin or Yoruba, if one gave them a greatly reduced diction compared to contemporary speakers. English is simply poorly suited to convey how nice these epithets roll off the tongue and why they're leaned into so heavily.

Granted, the languages I speak have so little overlap in text it's actually quite difficult to imagine translating between them without a great loss of meaning and tone. And if you look at something like Tang chinese poetry (let alone something truly ancient) translation becomes a game of "which aspect of the linguistic dynamics here are worth communicating to english speakers?"

So, I'm not sure the "language" punches through so much as you see innovation in the use of english to convey virtually-impossible-to-translate tone and rhythm and wordplay—but it's still relatively contemporary and idiomatic english, or it would simply not reach most readers. Though somehow superlative translators manage more than I thought was possible.... sometimes it feels like reading shakespeare at great effort and difficulty is the closest english-only-speakers will come to understanding how constraining a language modern english is for formal poetry.

Prose is much easier.

It is down to the judgement of the translator to choose what to preserve, I'm not suggesting they should carry everything over. Some things will be too difficult for the reader and not worth the effort.

However, there will be cases where the original text will have lines that shaped the language and literature it came from and it would be remiss not to let the reader feel that texture, even if it is a challenge to read.

It's difficult to pick examples but an influential line like 'all that glisters is not gold' could be translated to the equivalent of 'don't be fooled by appearances' but that would just kill the imagery and poetic quality. I would want the equivalent of 'glister', I'd want it to express the shine and glow, I'd want the word to be archaic and I'd want the rhythm of the line to be there.

We see this a lot in the manga/anime communities. Readers usually get used to the Japanese honorifics, name suffixes etc. and have a pretty good understanding of them.

An adaptator getting rid of them to have more natural English will come up with way worse phrasing, while losing a lot of the nuance and meaning of the original text.

In a way, the reader undertands that it all happens in a non-English country, so getting unnatural English phrasing isn't much of an issue.

That, and I want the translated work to reflect the original vision of the author as much as possible. I don't want cultural references changed for my country, nor do I want things changed because my country might find them offensive when the original audience didn't. In my opinion, the best compliment you can give a translation is to say that it fades into the background and lets the original shine through.

But what is the original vision of the author? It may not be possible to fully to fully capture without compromise.

For example, many people in this thread are insistent they want a "phrase by phrase" translation. But what if the author wrote the book in a colloquial spoken style, that is specifically supposed to come across as neutral contemporary language? If you translate the work into English in a stilted style that carefully preserves the structure of the original even when it produces unnaturalistic English, you have lost a key aspect of the original vision of the work. It won't hit the reader the same way.

Strongly disagree. Translation is like taking a 3D object and trying to convey it in 2D. You can be overly descriptive and accurate but lose all sense of flow. You can try to capture the flow of things but then lose the precision of meaning. Or you can approximate both to make the translation as strong as possible on both front, then get slammed from both sides.

Sometimes translators can highlight a less considered aspect of the work, like when you hear a great cover song that makes you appreciate the original that much more.

I understand the ideal of the perfectly translated thing, by definitionally it can't exist. I suggest going to a library and read the first page of 5+ translations (let's say, Don Quixote) and you will have a pretty clear sense of what the original text must have conveyed as well as several conflicting voices, some of which you will hate and others you may even appreciate. But you can never hear the original through a translation so it might as well be one that is interesting in its own right.

If original vision involved the language in which the book is written to be natural to natives, then weird turn of phrase in the translation is not accurate. It does not let the original shine, it makes the original appear odd.

This is very close to the traditional discussion of translation vs adaptation[0].

> Perhaps there is a larger prejudice here against scrutiny of translation as overly academic, an exercise in smug pedantry hampering creative freedom

Yes. And while the discussion is about literary work, I think this is an extension of the stance people have toward translation in general.

We see that with product designers demoing AI translation as "transparent", and something that allows users to not have to care about the language barriers anymore.

While being able to roughly understand foreign contents is better than nothing, too many people IMHO are assuming that auto-translation is good enough. It seldom is, and will probably never be if we care even a little about what people are really trying to say. You can survive with translation, but the bare minimum shouldn't be the end goal, or at it we should keep in mind the missing parts.

[0] https://acculing.com/difference-between-translation-and-adap...

AI translations are getting really usable lately. I'm using it a lot for legal translations between German and English. This went from barely usable ten years ago to basically flawless today. I also use it to localize our company website in German (from English). I've had some native speakers double check my work and while there are occasional mistakes that need fixing, it's basically good enough that I trust it enough at this point to publish and then fix things on an incidental basis rather than getting a manual review first. Just faster and easier. We've had zero comments on this from our German speaking audience. They don't seem to be aware.

I recently added French for some of our pages as we were getting some inbound from France. You still need to be careful but mostly it does the job. And I get a kick out of seeing my writing translated into what looks (to me) pretty elegant French. I have some dormant memories of high school French. Enough to appreciate it. I generally just use OpenAI's o4 model currently. I might experiment with 4.5 to see if it does a better/different job. I generally just paste markdown and let it translate the whole thing with the instruction to preserve the structure. Works well enough.

I'm guessing there are going to be a lot of AI assisted book translations in the near future. I appreciate that they might not be amazing but they are probably getting good enough. And increased context sizes and improved instruction following should also make it easier for translators to stay on top of and direct/oversee the quality; e.g. fine-tuning certain wording and stylistic touches.

Probably a bigger factor here is original authors gate-keeping quality here. They wouldn't necessarily want a low quality translation misrepresenting their work. However, I would expect that especially authors targeting relatively minor languages might have an incentive to self-publish their work in other languages targeting larger audiences (English, Spanish, etc.) and that a few successes with this might make them more confident. If you did the hard work writing the book and it's moderately successful, why not increase the potential audience by 10-20x? Of course publishers might pick up on this faster than starving authors.

And the standards for self publishing on e.g. Amazon are pretty low. Probably a lot of AI generated garbage there to begin with.

This made me think a lot about Sharon Choi, so I'm sure must be just insanely talented. https://www.washingtonpost.com/gender-identity/you-probably-...

Reply on news.ycombinator.com