16th Annual JAT Contest: Judges Commentaries 2

Mar 15, 2020

One tricky aspect of the article is that it described a dispassionate academic, or scientific, endeavor in vivid, eloquent, yet measured, non-technical terms to the highly educated lay audience of Chuo Koron. Kind of the New Yorker of Japan, Chuo Koron is a “literary magazine,” a 総合雑誌 (general magazine) in Japanese. It is not an academic journal. This would suggest that adding extra information about statistics or rewording some of the authors’ text might not serve the purpose of localization. If the writers and editors of Chuo Koron put something into the article, perhaps it should be left that way.

Nonetheless, the translations by the finalists were either extremely accurate or extremely articulate or some of both. Also to qualify as finalist, the JAT members serving as “screeners” in the contest made sure that the contestant had correctly translated the four following proper nouns in the text, which were available on the web: 1) ヘイトスピーチ対策法 (Act on the Promotion of Efforts to Eliminate Unfair Discriminatory Speech and Behavior against Persons Originating from Outside Japan;” 2) & 3) 日本学術振興会 & 科学研究費 Japan Society for the Promotion of Science & Grant-in-Aid for Scientific Research (or "Kakenhi" with explanation); and 4) （株）日経リサーチ (Nikkei Research Inc.) This would be an expectation for a professional translation.

Some comments on individual finalists appear below. Tables comparing some translations of jukugo selected from all contestants in the text appear at the end of the commentary.

E59 (First Place)

E59’s translation showed an excellent combination of pleasant English usage and accuracy. She crunched almost all of the verbiage used by the authors into understandable and readable English. E59 found ways to use active verbs when there were no subjects and was very faithful to the intent and tone of the authors.

The opening definition of hate speech is nicely worded. She dealt with jukugo well, editing 対策 out of 前記対策法 (the aforementioned law) and 規定 out of 禁止規定 (prohibitions or penalties). E59’s translation reflected the tone of the piece, stating that rights were simply “established” (rather than the more inflammatory “guaranteed” or loftier “enshrined”) by the constitution and including “streets and public parks” as sites of hate demonstrations, as written by the authors.

There appears to be one place where her wording was misleading: “there has been no indication that the hate speech which continues to be propagated on the internet has subsided since the legislation went into effect.” (ネットでのヘイトスピーチは収束する気配がなく、いまだに拡散し続ける。) It appears she knows what the passage means, but the wording obscured the meaning. It looks like both verbs here go with hate speech on the net, but an English-speaking reader would think that the authors stated that hate speech in general is proliferating unabated.

Another confusing usage is “Japan should still be in the process of formulating new national guidelines” for “日本は新しい国民的な規範を形成していく途上にある、というべきであろう。” Apparently “should” is supposed to indicate conjecture, but I don’t think it strikes the (modern) English-speaking reader that way. I’m not sure what it means.

In the paragraph describing defining a “survey experiment,” she avoided using the gairaigo タイミング, using the pithy phrase “seizing the opportunity.” She also faithfully conveyed the definition of the experiment in the authors’ lay language that still created the syntax of survey questions and included the basic technical details of experiment. Like many other contestants, E59 bothered to find out that the authors’ research received multiple grants.

Many were confused by the あらかじめ at the beginning of the final paragraph. Unfortunately, E59 thought this applied to the research, rather than to first giving an overview. The final paragraph was very clear overall. However, she gave somewhat literal renditions of the jukugo人口構成比 (“composition ratios of Japanese population”) and 順次解説 (“explain sequentially below”). She included the phrase that the authors were trying to create a “microcosm” (縮図) of Japanese society, consistent with their use of lay language, while some other contestants inserted statistics terminology that did not reflect the authors’ tone and wasn’t in the original text. Otherwise, the final paragraph consisted for the most part of a series of clear, understandable, palatable sentences.

E51 (Runner-Up)

By my count, E51 produced the most accurate translation, with the least failures to transfer the meaning of the original Japanese to English. However, there were several literal translations, especially of jukugo, and perhaps some misuse of articles and other errors in English. Although much of the English vividly conveyed the meaning of the Japanese, it did so in a slightly awkward and stilted manner.

Even though several instances of misuse will be pointed out below, it should be emphasized again that this was the most accurate translation by far.

In the opening definition of hate speech, the “distinct from the rest” of “less numerous groups of people distinct from the rest” is almost lyrical in and of themselves, but they kind of strike the wrong note here.

In the third paragraph, the register is too informal in “When it comes to government regulation of hate speech.” “Concerns of” is not the best collocation here. In “others criticize that the aforementioned countermeasure act,” “criticize that” is not that natural-sounding, and “the aforementioned countermeasure act” leaves the entire jukugo intact, when it was reduced to simply “the act” in the most natural renditions. The “prohibitive provisions” jukugo was also left as-is. “Canon” in “a new national canon” seems to come out of the blue. E51 literally translated “timing” and showed improper article usage (or pluralization) in “a method called ‘survey method’” and “survey method is an approach” (“a survey method is a” or “survey methods are”).

But despite these mostly superficial blemishes, the reader obtains virtually all of the information in the article. And there were several colorful expressions such as “come under fire,” “leaning on the legislation,” and “take a tough stance.”

E05 (Finalist)

Matthew Schlecht, PhD Chemist, decades of industry experience, formidable translator, prolific Honyaku Forum contributor, and erudite conference presenter (speaking about patent translations): The clients aren't much interested in my opinion of what the patent author meant to say outside of a mention in a translator's note.

E05 showed enormous aptitude, skill, and potential for growth as a translator and produced a literate and eloquent translation that was a pleasure to read. Although there were some misunderstandings, the translation conveyed the meaning of most of the text. However, the liberal interpretation of the text strayed from the tone, or register, of the original article, aimed at a highly educated lay readership. The translation is an excellent piece of writing, but perhaps not the most faithful representative of the original text. I don’t know if this was entirely intentional or the result of a last-minute effort that was never checked against the original text (some of the actual errors looked more like checking errors).

I comment on this because I think there is a natural tendency to, er, rigorously interpret the source text. In many cases, a working translator is required to clean up the original text in the process of localization. However, in this case, the authors’ manner of expression appears to be intentional and should have been conveyed in the translation. Past contestants have taken this approach, and some of my first paid translations made too many “improvements” on the source text, to the vocal consternation of the money-paying client.

But first, the heading for contest passage was a difficult challenge (「ヘイトスピーチ」を学術的に検証するために). Most of the finalists’ translations didn’t really sound like something an English speaker would say. However, I thought E05 gave an excellent rendition with “Examining hate speech under an academic lens,” although Contestant E60 came up a pithy “Scientifically examining hate speech.”

The use of “marginalized groups” rather than just “minorities” is economical, but incendiary and editorializing. That freedom of expression was “guaranteed” by the Japanese constitution rather than just “established” (定められた) is also incendiary.

“路上や公園でのヘイトデモ” was reduced to “public spaces.” This rendition certainly conveys almost all of the meaning, but there have many hate demonstrations and counter-demonstrations on roads and in public parks, and the authors may have specifically wanted to cite those venues. (They are the experts.)

“Japan still has substantial ground to cover before it can set a new national standard” once again conveys the meaning, but inverts the positive of tone “日本は新しい国民的な規範を形成していく途上にある.”

E05 wrote “regarding the extent to which the government should regulate discrimination and verbal abuse,” but the authors intentionally mirrored the categorizing style of an opinion survey with “should (should not) regulate” (規制すべきである（規制すべきでない）).

Too much license was taken with the following simple lay description of the survey, which centered on the introduction of “experimental variables.” It’s pretty hard to imagine that the authors would not want that crucial phrase to appear in the translation. In this instance, the basic meaning as well as the tenor of the sentence was changed.

Oddly, E05 omitted the definition of the experimental element of the survey (to introduce variables) in the sentence: “This large scale public opinion poll is an opportunity to gain scientific insight into the reasons why people adopt different actions and attitudes regarding hate speech.” (サーベイ実験とは、大規模な世論調査に実験的要素を組み込むことで、人々の行動や態度に違いを生じさせる要因を特定しようとする方法です。Perhaps this is a checking error, rather than taking license.

The passage すなわち国民全体の縮図となるような形で回答者を抽出することはできない。was translated “it was impossible to secure a sample representative of the whole population of Japan.” No information was lost, but it appears the authors were attempting to be eloquent and non-technical and it would have reflected the tone of the article to include a rendering of 縮図.

Forgive me for not pointing out the many excellent examples in this translation.

E62 (Finalist)

E62’s transition was especially well written, but unfortunately contained several errors in the transfer of meaning and was somewhat freewheeling in a way that distorted the tone of the original text.

Good turns of phrase and word choices included the first sentence of the second paragraph, “some have voiced deep concerns,” “no prohibitions or penalties,” “law came into force,” “local authorities”…“have strengthened their stance,” “presented with this opportunity,” “to pinpoint factors,” and referring to the limitations of the survey as “caveats.”

However, major errors of meaning transfer included “they are against hate speech” for “should (or should not) be regulated,” “may get carried away by emotions or sentiments when discussing certain themes or topics” for “due to the nature of the topic, people get carried away,” “age…60” for “age…69” [probably a typo], and “we will discuss a study that we conducted earlier” for “first, we will discuss the [present] study that we conducted.”

Nonetheless, the quality of the English writing that E62 used to convey the Japanese cannot be understated.

E29 (Finalist)

E29 wrote in very natural English, with little evidence of the original Japanese grammar and syntax, but committed a few errors of transfer of meaning.

Good turns of phrase and word choices included “[speech] which incites discrimination…,] “not taking sufficient action,” “what is known as the ‘Hate Speech Act,’” “does not prohibit or penalize,” “the mechanisms behind how people form their attitudes,” “discussions tend to be driven,” and “the Act” for “前記対策法.”

However, major errors of meaning transfer included translating “収束する” as “converging” (Weblio and Eijiro contain the definitions “settle” and “blow over”), “it is questionable whether the country is establishing new norms” (it appears to be establishing them), “inflammatory speech invites danger” for “there is a risk of inflammatory rhetoric,” “we will discuss the overview of our study” (they’re giving an overview).

Otherwise, however, E29 extracted very natural English from the Japanese text, offering one of the most pleasant translations to read.

Lost in Translation – Jukugo

Pithy expressions in one language aren’t often pithy in another. Jukugo, combinations of two or more kanji, whether literal or idiomatic, sound good in Japanese, which is an agglutinative language anyway. However, they tend to be awkward and mask or distort the intended meaning if left intact when translated into English. All of the following jukugo from the contest passage are literal (mean what they say), but most could use reworking to sound good or convey their intended meaning in English. These jukugo aren’t that formidable, but they do illustrate this challenge. Translations by the finalists and several of the other contestants are presented in tables below.

態度決定 Needs a verb.

順次解説 Get rid of the 順次? Simply use a plural?

人口構成比 Japan’s population?