So, over the last few weeks I’ve been playing through Namco-Bandai’s Tales of Vesperia with a friend. We haven’t gotten particularly far yet, but the game is quite pretty and very amusing so far. I’m not here to write a review, though — not right now, anyway — and there’s been one thing about this game that’s been bothering me since we started playing it: the translation.
Now, don’t get me wrong, it’s entirely competent, it’s just really quite dry considering the playful, light-hearted nature of the game itself. Characters speak in stilted, unnatural ways, and I can almost feel the original Japanese text oozing out from right beneath. They have almost no defining characteristics to their speech, which almost negates the surprisingly good English voice work.
I wouldn’t be surprised if the localization team was eight or ten people: There are no personal voices to be found here. I feel like Vesperia has too conservative a translation; like it needs Vic Ireland to come pepper in some Bill Clinton jokes.
The way I see it, there are three major approaches to translation of media like games or anime: translation, localization, and adaptation. Each deviates successively more from the source material, and each has its uses depending on the material and the target audience. Let’s look at them one by one:
First of all is direct translation. In the most direct translations, you’re writing to communicate the meaning of the original words first and foremost. Any changes made are solely for the sake of grammar. Rather than modifying an unfamiliar cultural element to something more familiar to the audience, these often include translation notes to explain the concept to the reader instead.
This is the most conservative type of translation. It keeps the literal meaning of the source material intact, making it particularly suitable for things like educational or business materials. The result generally doesn’t feel natural in the target language, though. If the translator can write decent English, it will make sense, and the reader will be able to understand it, but it is also generally quite obvious that it was not originally written in English.
I find many fan-produced translations, when they aren’t veering in the opposite direction, are guilty of relying on this far too much. It is difficult to engage and entertain a reader when they are constantly reminded of the foreign nature of what they are reading. It feels dry and academic, which is not what you want when you’re aiming to entertain. There are certainly times when it is appropriate, though: Atlus’ Megami Tensei franchise is rooted in Japanese culture and mythology, and to strip that out is to completely miss the point, resulting in abortions like the English release of the first Persona.
Moving a bit away from the source material, we get to localization. Despite most commercial translations describing themselves as this, it is not quite as common as you might think. The priority of a localization is to maintain the intent and the “flavor”, so to speak, of the original material rather than to stick directly to the literal text itself. Changes are made when the original content would not work naturally in the target language; most commonly this affects humor, but anything culturally or linguistically idiosyncratic can be affected by this.
An original Japanese text might make a groan-inducing pun. While explaining the pun itself in English may still induce a groan, it would be distracting, and probably not the right kind of groan. In a localization, it would likely be replaced with a similar, but different, joke that works for the situation in the target language.
For example, in Mother 3, there is a character named “Yokuba”. His name is a play on the Japanese word “yokubari”, which means “greed” — given his character, the name fits him quite well. The unofficial English version changes his name to “Fassad”, a play on “façade”, describing a different but still unpleasant element to his character. The literal meaning may have changed, but the effect is the same: through the character’s name, we already understand that he is not the most honest man.
The most liberal type of translation is an adaptation. In some cases, an adaptation is not even a translation at all; there are often significant changes made to the source material, and in the most extreme cases may be a complete rewrite using the original materials simply as a guideline. Robotech is a particularly famous example of this, repurposing footage from several Japanese robot shows, mashed together and rearranged to create something almost completely different. If you aren’t careful, these are a great way to piss off an entire fanbase. That said, there are definitely cases where this kind of approach can be used to great effect. In particularly, I believe this approach can be useful for absurd comedy-based material, as well as for material aimed squarely at younger children.
Let’s say that for example you’re tasked with producing an English version of the classic childrens’ cartoon Doraemon. While the content is, for the most part, pretty universal — many kids would be able to relate to Nobita’s follies — there are still distinctly Japanese elements within that most kids would neither understand nor particularly care to understand. Since your goal is foremost to entertain, it’s entirely acceptable to throw out the original dialogue and rewrite entire scenes from scratch to suit the target audience better.
However, as I said before, you must take extreme care when deciding to take this approach. Robotech only gets a pass due to its vintage, and if someone today were to try the same thing, they’d likely not only get an incredible negative reaction from the fans, but likely threats of legal action from the original producers as well. Tread lightly.
To be perfectly honest, I find that the slightly more liberal translation found in a localization is preferred in most cases. There are of course situations where even that may be too much, so careful consideration of both the source material and the target audience is necessary when figuring out the approach to a project. A translation that would please the otaku audience might put off a more general audience, while an adaptation designed for children would likely be torn apart by the otaku audience.
And of course, there will always be someone out there decrying your translation because it isn’t close enough to the source material. To them, I say: always remember what happens when your translations are too literal. You get a Castle Shikigami 2, and nobody wants that.