Hi there! I’m Shannon, a Language Specialist in the JA-EN language pair. Going forward, I will be making periodic forum posts targeted at translators who work in this pair. I hope that these posts will facilitate better communication with the LS team and help you produce top-notch translations for our customers.
The first post will focus on source text ambiguity.
The Ambiguity Problem
Japanese is full of sentences that lack key details that must be clarified before it is possible to craft proper English matches. Consider this simple Japanese sentence: 学校に行きます。
It is ambiguous in at least three ways:
- No explicit subject
- Object has no article or quantifier
- Ambiguous tense
Therefore, the speaker may be trying to say any one of the following things, just to name a few:
- They go to school.
- The team is going to the schools.
- He will go to a school.
It is essential to resolve each source of ambiguity if your translation is to communicate the speaker’s intent accurately. But how? Here’s a two-pronged strategy for tackling an ambiguous source text without introducing erroneous information or deleting vital nuance.
The Solution, Step 1: Resolve the ambiguity.
A. Check the context.
Let’s return to the example given above: 学校に行きます。Based on this sentence alone, it is not possible to figure out which combination of subject/object/tense is intended. Stripped of its context, the sentence can mean many different things.
Put another way, no matter how plausible a candidate translation may be in a vacuum, it may actually be objectively incorrect in light of the text that surrounds it.
Here’s the same sentence, this time embedded in a dialogue:
A: 君は毎週月曜日に何をするの？
B: 学校に行きます。
A: 学校名は？
B: 渋谷小学校です。
Now it is clear that Person B is saying, “I go to school.” The ambiguity has been resolved, and we can see that all of the candidate translations above would actually have been wrong!
When translating an ambiguous segment, always zoom out and ask yourself whether the source text as a whole provides any clues as to how that segment should be interpreted.
But what if you remain stumped after looking at the context or, worse, you are wrestling with a tiny job that is entirely lacking in context?
B. Rewrite the English sentence to minimize extrapolation.
In some cases, it may be possible to rewrite the translation to get around the ambiguity and minimize the extrapolation.
Consider another example: ゆきさんは昼ご飯を食べに行った。
This sentence can be translated in at least two different ways:
- Yuki went to eat his lunch. / Yuki went to eat her lunch.
- Yuki went to eat lunch.
Unless you can glean Yuki’s gender from the context, Option A will force you to extrapolate this fact and supply information to the text. Option B, though, allows you to translate the segment without ever knowing (or guessing at) Yuki’s gender.
So even if there is ambiguity in the source text, playing around with the wording of the translation may allow you to avoid supplying information that you are unable to confirm.
The Solution, Step 2: Submit a provisional translation.
Even after attempting the techniques above, you may still find yourself needing to assume certain details that the source does not make explicit and that cannot be reasonably omitted in English, such as number or gender.
In cases like these, it may be necessary to incorporate provisional translations into your workflow, while remaining in close contact with the customer about your thought process.
A. Supply the missing information, flag the issue, and submit the job.
Imagine encountering the following source segment in a business email: ご提案を検討させていただきました結果、残念ではございますが、今回はご縁がなかったということでご容赦くださいませ。
What is the subject of the first clause, “I” or “we”? You cannot tell for sure.
One option would be to make a provisional translation that keeps the English sounding clear, such as "We considered your proposal, but…" You would then leave a comment flagging the choice of “we” over “I,” which would allow the customer to ask questions, request a revision, or make edits on their end if your choice turned out to be wrong.
B. Eliminate the troublesome nuance, flag the issue, and submit the job.
At times, it might be simpler to drop an ambiguous term from the target text, again making sure to flag the issue for customer review.
Consider the following source segment: 彼が本を返しに来るとのことです。 The とのことですconstruction indicates that the information has come from a source other than the speaker. However, translating it smoothly in English would require background information that is missing here, namely information about both the source of the information and its recipient. There are many different ways to render this construction in English, depending on the facts (e.g., "I/we have been told that..." / "apparently" / "he said that..." / “He told Lucy that…”)
You might translate this segment as "He will come to return the book(s)." You would then leave a comment informing the customer that you had not included a match for とのことです, with a request for the relevant background details in case they felt strongly that the translation should reflect this nuance. (You would also note that the customer should confirm the number of books and edit the text accordingly.)
C. Ask before submitting.
If the source text is on the longer side and there is sufficient time before the submission deadline, also consider using the comment section to seek clarification from the customer before submitting the job.
If the customer does not provide a timely response, you can always fall back on the provisional translation approaches described in 2.A. and 2.B. above.
Important Note: Always flag provisional translations for customer review.
It is not appropriate to add information to, or omit it from, the text without comment. While some customers may be less responsive than others, your job as a translator is to keep them empowered and informed.
—
Any thoughts or questions about the above? Feel free to chime in below! I look forward to getting to know you better.
0 comments