Allow me to elucidate.
Have you read any Italo Calvino? If not, you should. Invisible Cities is such a remarkably gameable book that I've seen several authors and bloggers describe how they use it in their games, myself included. A translation team who did not have a passion for the text could not have translated it properly, and yet had the editor been as wrapped up in the text as the translator, we would naturally have seen some flaws in the translated text, he might not have spotted mistakes that translator made, idioms that fail in the second language but work in the first but the translator tried to use anyway, peculiarities of the primary language that make it into the secondary text, stuff like that.
As essential skill for a translation editor is the ability to critically read the text: to examine the text for places where there is a disconnection between what the translator intended and what the text actually says. Sure, there are standard proofreading tasks that are important as well, but just as important as these tasks is the editing of the text to make sure it flows and works the same way as the original, even if that means approximating an initial logic rather than replicating it.
For example, let's look at Kant's Critique of Pure Reason. Here's a text that's so dense, the author himself wrote a simplified version of it, the Prologomena To Any Future Metaphysic, but even that was so convoluted that even German students of philosophy prefer to read the text in English translation because the translation approximates the logical structure of the original without producing a word-for-word replication of it.
On a similar philosophical note, Jean-Paul Sartre famously derided the existential philosophy of Martin Heidegger, but Sartre's own works included elements that were so remarkably similar to those in Heidegger's, you'd think the opposite was true. While a pessimist might suggest that the problem was just Sartre's ego and that he viewed his own works as massive improvements upon Heidegger's "mistakes," but the truth is that Sartre had actually read quite poor and hastily-done translations of Heidegger's Being and Time, ones that had completely missed the mark of what Heidegger was actually saying in the work. Sartre would later read better translations and become one of Heidegger's biggest supporters, even if ol' Martin didn't support back.
While I don't watch a lot of anime these days (that's the wife's department), I remember the old Wild West days of the VHS fan sub fondly. Invariably, these fan subs would translate the text verbatim, which would often include strange idioms without any explanation and phrases whose meanings were so obtuse as to be unintelligible that a large part of watching these fan subs was learning to gloss over translation oddities and try to sort out from context what was going on. These translations were a labor of love, however, and I never paid for them, so I can't quite complain. Nor did I back then. It was just par for the course and we accepted it. It was really irritating to me when things got the English dub treatment yet the scripts would keep these sort of unprofessional irregularities and a large part of why I will always prefer a sub to a dub: I can always assume with a sub that no one has bothered to look at the language of the sub to figure out how to do it better. In a dub, you know that someone had the opportunity to improve the hitchy language and just never did anything about it.
Here's the point that I'm trying to make: bad translations aren't the result of a bad translator, they're the result of a lack of a critical reading of the text. "Bad translation" isn't even really a thing, it's more of "incomplete translation" because until a text has been critically read and edited taking into account that critical reading, it is not complete. Back in the day, when I was a German language major in college, the in vogue way of framing this conversation was "translation" as opposed to "interpretation:" just because you had translated a text doesn't mean you've correctly interpreted its meaning in the secondary language.
A translation cannot rely on specific knowledge of the thing being translated for it to be understood. For example, I should not have to have read Kant in German to understand an English translation. Nor should I require that someone explain his text to me in order to read that English translation. A translation should stand on its own without requiring recourse to any further texts.
Now for your big question: What the fuck does this have to do with gaming?
Yesterday, I bought a game via RPGNow that I'd been looking forward to the English translation of. I will not mention its title, but you can probably sort it out if you look at reviews I've written on OBS; there aren't many of them. Now, I'm used to awkward translations. Adventures in the East Mark had some peculiarities (some of which were actually charming while some were irritating), as does Shadows of Esteren (seriously, "combativeness?" Esteren's sentence structure is a trainwreck as well), but I can largely make out what's intended in both of these. (And in East Mark's case, Mr. Brown did a great job of cleaning up the less charming peculiarities once they were pointed out to him.)
So, I read this game that I had been looking forward to an English translation of. I started with high energy and enthusiasm. This enthusiasm flagged a little bit when I hit some wonky sentence structure (this really should have been sorted out in editing) and *gasp!* a double negative (I've determined that, like French, Italian -- the game's primary language -- does indeed use double negatives, so this is a straight up word-for-word translation and an oversight of the editor's) and to see that stuff in what's supposed to be a final product (I'd assume you'd only make a print option available when the book was in its final state unless you were clearly selling a beta product) irked me a bit, but I pressed on. When I got to the first bits of the game's central structure, I got excited again. This game was implementing some basic mechanics logic that I enjoy and had been positing on my own, so it was validating to see them here as well. But then, I got a little further in on the central mechanics and it all broke down. There were declarative statements. There were clarifications of those statements, but the clarifications contradicted the declarations. Then, there were examples that injected notions that I could find nowhere in the text itself. Basically, the examples of play had to be read to explain how the rules portion of the game worked because the rules were too obtuse to be read on their own. This is an incomplete translation.
I enjoy that the publisher wants to bring this game to the English-speaking world, and I think that the English-speaking RPG market has a bunch of stuff to learn from the wider body of non-English-language RPGs out there (for example, I think that the English RPG market has ignored Das Schwarze Auge for far too long). I believe that the English-language RPG community will be richer from the presence of this game within it. However, I believe that it is important to do justice to the game in translation. I want to understand the game, I want to know how it works, and I should not need any special knowledge from a source outside of the book to do those things.
To ensure that your game is understandable by the broadest amount of people when it is translated, it is necessary that a critical reading of the text be performed by someone who is not intimately familiar with it. Someone who hasn't played the game before, or had it explained to them. Someone whose perspective on the game is fresh enough that she will ask "does this mean 'X' or "Y?'" Someone whose reading of the text isn't polluted by prior knowledge and whose command of the secondary language will help bridge gaps between the intent of a passage and what's actually on the page.
In short, if you're translating a game into a language that's not your first language, you need a translation editor.
Joesky Tax: The Periapt of Imperfect Comprehension
This insidious item teeters on the border between "useful magical trinket" and "useless cursed trash." When worn, it allows its wearer to understand most of what's being said in a language he does not speak, but not quite everything. The wearer understands 80% of this periapt-translated communication correctly, but 20% comes out terribly mangled. Determine which component/s of a sentence are garbled, then roll on the following table to see how that component is confused (1d6):
- Gibberish - nonsense words and meaningless syllables
- Opposite of intended word/s
- Non sequitur - the words are real, but are largely meaningless in this context
- Wrong part of speech - instead of a noun, a verb; instead of a verb, an adjective; instead of an adjective, a preposition, and so on.
- Poor grammar - words are right, but the conjugation or declension or pluralization is off
- Roll again twice.
The periapt does not confer any ability to speak the language/s it imparts imperfect knowledge of, nor does it allow its wearer to read anything written in a "comprehended" tongue. The periapt will inevitably interfere with the wearer's ability to communicate in languages he does know, as well. Every time the character speaks, he has a 1-in-12 chance that he must roll on the chart above for the entirety of a single communication. This chance increases by one until the periapt has mangled his speech. The periapt cannot be removed without a Remove curse spell being cast upon the wearer.