I was celebrating Victoria Day last weekend at the beach with a few friends, discussing the anticipated fireworks show that was to come that evening, when I said “our tax dollars paid for the fireworks” in Japanese. My friends understood what I meant, but said that it *has* to be recast in the passive (or active with a null subject). Why?? What’s going on??

So there’s this little thing in semantics called metonymy, where a word is used to represent another word. A popular example is “I read Shakespeare”, where it’s understood that what’s being read is the works of Shakespeare, rather than the decaying corpse of the Bard. Metonymy comes in various forms such as container-for-contained (e.g. “the White House” for the civ. servants that work in the White House; “dishes” for the food on the dishes, etc.); tool-for-work/accomplishment (e.g. “hand” for “throwing ability”; “pen” for “writing style”, etc.); part-for-whole (e.g. “wheels” for “a car”; “head” for “person” (when counting), etc.), etc.

And apparently, metonymy isn’t necessarily extendable cross-linguistically. According to Kamei and Wakao (1992), “he read Mao” is acceptable in English and Japanese, but not in Chinese. (My mother disagrees, but she’s also been a naturalized Canadian for over three decades, so her grammaticality judgements in Mandarin may be prejudiced.) This means that just because something is metonymic in English, doesn’t mean it’ll necessarily be metonymic in Japanese — which clearly is the root cause of my miscommunication with my friends about fireworks being paid for by the city through our tax dollars.

So maybe metonymy is a lexical property that differs across languages, making it a syntactic property. After all, some languages include things such as number, gender, and case with their nouns, while others don’t. But, for one thing, any metonymic reading is selectional and restrictive. While the number, gender, and case parameters are _always_ obligatory in the languages that use them, a metonymic reading is contingent on the verb being used, and (to some extent) the context. Although we can understand “lend me your ears” to mean “give me your attention”, you can’t achieve the same judgement in “do I have your ears yet?”

At the same time, a lexical substitution on “ear” for any other word relating to the organ also fails to obtain that metonymic reading: *lend me your cochlea; *lend me your lobes, so to an extent, there is some lexical property at play. The metonymic reading can also be cancelled by new information: “* I’m parked out back and may not start”.

So there is demonstrably some idiomaticity to metonymy, but also one of selectional restriction. The idea is that if the literal meaning does not compute (for lack of a better word), then a related salient concept must be accessed in order to achieve meaning, whence the metonymic interpretation. But if that were true, why is it that such a selectional restriction violation should successfully trigger metonymy in some cases, but not others? From what I’ve read, there’s some statistical data to suggest that this is linked to how closely linked the metonymic form and the metonymic reading are.

So back to the issue of “our tax dollars” in Japanese. After a quick Google search, I found that all subject-marked phrases of “our tax dollars” were almost always accompanied by passive verbs, which indicates to me that money, unlike in English, is not interpreted metonymically as an agent for purchase. Another way of putting is, in Japanese, the semantic link between “money” and “user” is not as strong as it is in English, which is why even under the literal reading fail, the metonymic reading doesn’t surface.

Addendum (2013/06/09): I’m suddenly reminded of an early misinterpretation I had on a somewhat well-known slogan for a dairy-based candy: 「ママの味」 (lit. “taste of mommy”). Between the fact that the confectionery in question was dairy-based, and the slogan likening its flavour to that of one’s mother, I naturally drew unpalatable conclusion that this Japanese slogan aimed at children was intimating that the candy tasted like breast milk. Ew. Of course, I now know that it’s in fact another example of metonymy, where “mother” is the trigger for “mother’s cooking”. So really, what the slogan is saying is, “it’s just like mother used to make!” rather than “it tastes like those first __ months of your life!”

(English vers top)

先週末、ヴィクトリアの日[1] をお祝いするため、友達と海へ行って、その晩に期待された花火の話をしていた時、「私たちの税金が花火を支払った」と私が言ってしまった。相手の二人が理解してくれましたが、受身形じゃないといけないと言われた(若しくは主語無しに能動動詞と)。どうして??何ゆえぢゃ!


それに、換喩は他言語に伸ばせるわけでもなさそうです。亀井と若尾 (1992) によれば、「彼が毛(マオ)を読んだ」は英語と日本語で通じますが、中国語では通じない。(うちの中国語母語の母さんが通じると思ってますが、三十余年間にカナダで住んでつつあったので、中国語の判断が偏ったかも。)こうなると、ある言葉が日本語で換喩になれても、英語の換喩になれるとは限らない。だから先週末、花火が税金で払ったことを友達に誤解された原因です。



上記どおり、換喩って慣用的なところがありながら、制限的な範囲もあります。意味論によって、語義通りがあってないなら、連想されたことを接近して、換喩の 意味が取ってこれる。でも、本当そうだったら、何で、時に換喩的意味が取れるか、時に取れないか?諸説の論文を読んでみれば、それは「換喩の言葉」と「換 喩の意味」の距離の問題です。


付録(2013/06/09): 突然、私が有名な飴の会社の標語の誤解を思い出した:「ママの味」。この飴が乳類と標語が自分の母の味覚性の間に、私が語学学習性として当然に、この子供向きの標語はこの飴が母乳の味という結論を引き出しちゃった。ぶぇー。勿論、今はもう一つ換喩の例だと知っています。「ママ」は「ママの料理・手際」の代表だ。即ち、この標語は本当に「生まれてから何ヶ月間の口にした味」じゃなくて、「母が昔作ってくれた料理と同じ味」の意味をしています。

[1] カナダが昔、大英帝国の領地の証明、ヴィクトリア女王の誕生日だった。今は帝国って連想がなくて、学校や仕事から休んで、花火を見ることになった。