Open Secret 2 – Letter from a Reader

An owl swooped in and dropped a letter onto Ameranth’s lap. She smiled, unrolled the letter, and began to read.

“Hi, are you guys translating from the mandarin version of Grimgar? I noticed that some words/names are ‘slightly’ different from the original-for example, in v1 ch0, “Awaken” from your translation is “Awake” in the raw, and Altana is Oltana/Ortana(? whichever).

I’m just curious.”

Her eyebrows rose. She transcribed the letter, sent it to hikaslap, and not five minutes later her phone buzzed with hikaslap’s reply.

“Haha. No.

There is a lot of room for maneuvering when you translate, particularly when you ascribe to “sense-for-sense” translation like we do. Lots of alternative spellings and phrasings.

If someone doesn’t think that the use of “Awaken” is appropriate, or that “Altana” doesn’t sound cooler, they’re in for a nasty surprise when they compare and contrast all the small and large things we changed that, in our opinion, make the translation better while keeping the original “sense”.

If it sounds like I’m angry at you, I’m not – I figure it was just an honest question. To answer it, the presence or lack of an N at the end of a word does not mean anyone translates through the grubby window of a middle man language. Not us.

(Thanks, Ameranth.)


But Ameranth had something to add, too. After “convening” with TOM, she began drafting a reply of her own.

“Hi! Thanks for your email! Since you know some things about Japanese, you might be interested in this more in-depth response.

When the Japanese make an English word into katakana, they’re going off the way the English word is pronounced, not the way it’s spelled (because look, English spelling rules are non-nonsensical and often have nothing to do with pronunciation). Therefore, you can’t assume that any word spelled with an “a” is naturally going to be replaced with the kana and, when translating, you can’t assume that katakana words starting with will always translate to “a”. For example, here are some other actual dictionary words (not made-up fantasy place names) where which is romanized as “o” becomes another letter in English:

オーロラ – aurora

オール – all

オランダ – Holland”

Ameranth wiped her brow, and continued writing. Looks like she still knew her Japanese.

“Things like how to translate a made-up place name starting with a katakana is a judgment call, just like Grimgal/Grimgar and 幻想 being translated to “fantasy” rather than “illusion” are judgment calls. It’s based on what the translator thinks the katakana word should SOUND like in English.

Also, fun fact: The official Japanese anime website translates オルタナas “Alterna”.

Yes, the first line of the light novel “目覚めよ” does have “アウエィク” in katakana next to it, indicating that the kanji should be read that way. However, that doesn’t mean that it should be translated into English as is. The katakana is there to give the obviously Japanese kanji word “目覚めよ” an overtly foreign tone, because it’s the first thing the characters hear when they enter a foreign world. It’s NOT there as an indication of how the word should be translated into English. TOM could have translated 目覚めよ into “Wake up!” if he wanted, and it wouldn’t have been wrong.

However, if you turn to page one of the Japanese Grimgar novel, or the English Volume 1 PDF or EPUB, the entire page is blank except for the word “目覚めよ” centered in the middle of it. This indicates to me that this one word, which opens up the entire Grimgar series, is highly significant in setting the scene and establishing the tone of this multi-book story TOM’s about to spend a bunch of time, uh, reading, or whatever it is he does with it. Probably TOM purposely made the word choice of “awaken” instead of “awake” because “awaken” has a more archaic feel (actually “awake” is derived from the middle English word “awaken”). It sounds more medieval, and it’s less commonly seen in our everyday English compared to the word “awake”. This not only helps to convey what the “アウエィク” katakana is trying to do in terms of making the word sound foreign, but it also fits better into Grimgar’s medieval fantasy setting compared to “awake”. Finally, the at the end of 目覚め effectively makes the word into a command, therefore the subtly stronger, more forceful “awaken” conveys the tone better than a normal “awake”.

I hope that helps to clarify your question. Like hika said, the translation is done directly from the Japanese and with conscious effort made towards keeping the sense of the original intact. Thanks for reading!


PS: Funimation translations (from where I assume you’re getting the “O” in オルタナ and “awake” for 目覚めよ) are the epitome of lazy. Their subtitles might be official and authorized, but that doesn’t mean the quality is good. Actually, their translations are often bland. They retain very little of the tone, writing style, and artistry of the original Japanese script. It seems like their translators just don’t care enough or don’t get paid enough to put in any more effort than making the translation accurate on a technical level, which is the absolute minimal required. And while their subtitles do a decent enough job of conveying the meaning (which is enough for most viewers, I suppose), it hardly does the Japanese scriptwriters justice.”

She put her pen down, rolled up her reply, and went back to the fireplace to warm up. It was cold where she lived.


8 thoughts on “Open Secret 2 – Letter from a Reader

  1. Hey !
    I just wanted to comment that in the transformation of japanese words to romaji, they actually use exactly the spanish pronunciation of the latin-alphabet-written word, for example the word HOUSE we spanish-speakers (Im from Argentina) would read something like o-uh-SE (just watch this vid ). I think this has something to do with the first Catholic missionaries to disembark in Japan, who were Spanish and Portuguese, and who said that the japanese writing system was made by the devil (lol) so they tried to teach latin characters, using the pronunciation they had. Btw portuguese is very, very similar to spanish, almost like a dialect).


    • That’s not entirely true. They do something like that for some languages, like Greek, and some words in other languages. For English, the main current seems to be adaptation through British official accent, with some recently added words coming from US accent, and random words coming from the accent of wherever the first person that used that word before a Japanese was from, direct reading as Japanese romaji/Spanish-portuguese writing or mishearing (like “tagger knife”, which is misheard “dagger” + “knife”).
      For languages like Russian they go by ear always, which makes it a total chaos to research the words back (changes in vowels, consonant simplifications…).

      PS. While Portuguese and Spanish, as Iberian romance languages, do have similitudes, there are two recognized midpoints in the Western Block continuum that make the intelligibility drop a lot unless you have enough proficiency in both languages to notice similitudes or know one of the midpoint languages (Galician or Bable). Just take into account that both languages have a different ratio and assortment of root languages (to say something, Portuguese has words that come from Suevian and a higher percentage of Celtic terms), and Portuguese changed further after Portugal became independent.


      • Are you talking about loanwords? The main point isn’t really loanwords, the point is that Spanish and Japanese romaji have similar pronunciation schemes. This at least is audibly true, though one can debate the influence of Spanish missionaries – Japanese is recognized to be in only its own language family, after all.

        There is one instance in Southeast Asia where French missionaries (exactly one man, in this case) revolutionized the Vietnamese language by inventing a way to describe it with Latin characters, and the use of Chinese characters in Vietnamese today has been almost entirely discarded.


      • Well, Japanese romaji is closer in pronunciation to Latin (Spanish h is mute), but the reason would be that it were the Portuguese the ones to bring the letters there (the oldest bilingual dictionary Japanese-[any Western Language] is Potuguese, after all).


      • Romaji doesn’t really matter anyway. The existence of equivalent sounds is mysterious enough and not easily explained (not that any explanation could exist, I think.)

        It is true that the earliest attempts at romaji were based on Portuguese, but the current one in use was developed by an American (Hepburn), which is based on English phonology.


  2. Examples of Funimation and Crunchiroll (i.e. “official translators”) downright laziness/weirdness:
    -GATE 2: the Japanese ambassador is referred to as “-sama” by the nobles, while the subtitles consistently read “-dono”.
    -Durarara!!: Sonohara Anri addresses Karisawa Erika as “Karisawa-san”, but the subtitles say “Erika” (which even in English is a big leap of treatment).
    -Many, many others: have you ever seen any anime where the name of someone, something or somewhere is written in romaji (normal letters) in-screen while at the same time it appears subtitled for both spellings to be different (off the top of my hat, you have Gakusen Toshi Asterisk, which repeatedly shows in announcements the names of the contestants and their schools in romaji, and Heavy Object, that as a “foreign story” has romaji as the only writing system, while in this last one only the names of the main characters have been properly romanized).


    • The biggest example of how Funimation doesn’t care about the quality of their translations should be Aquarion Logos, where they translated “migi” as left.


Got something to add?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: