Translation is hard

(This article assumes a certain level of Chinese proficiency from the readers)
In recent years, the Chinese translation of movie titles of movies produced in English speaking countries has become a laughing stock among the bilingual community. It is evident when articles like "超搞笑的港臺電影名翻譯", "片名的遊戲版本0.3 之一" (之二, 之三), and "香港台灣電影片名翻譯" all point to the same twisted translation practice employed by the three major Chinese speaking markets (Taiwan, Hong Kong + Macau, and Mainland). These articles have summarized that countless movie titles are merely permutations of

「魔鬼戰將戰警絕地追殺令追緝令火線無敵烈火超級終極任務悍將王牌致命終結令絕命危機總動員奇兵女兒閃靈神鬼迫切第三類戰士魔宮也瘋狂本色傳奇獵殺不可能終結者第六感情挑時空遊戲捍衛威龍小鬼赤子駭客惡夜毀滅最後英雄戰場桃色戰慄出擊擋不住攔截驚爆（點）特區帝國刺激特攻隊靈異奪命禁入時刻密碼戰拼圖我的野蠻心靈獵X行動星際霹靂特務魔法重裝殺手生死戀公敵百分百未眠辣妹控訴愛XXXX難開女（男）人香異想世界麻辣戰地」(at least this is the long standing pattern in Taiwan)

The following pirate movie and science fiction movie further illustrate the problem

Original:	Pirates of the Caribbean: The Curse of the Black Pearl (1)	Pirates of the Caribbean: Dead Man's Chest (2)	Pirates of the Caribbean: At World's End (3)
Chinese (literal)	加勒比海海盜：黑珍珠的詛咒	加勒比海海盜：死人的（寶）箱	加勒比海海盜：在世界的盡頭
TW	神鬼奇航：鬼盜船魔咒	神鬼奇航2：加勒比海盜	神鬼奇航3：世界的盡頭
Reversal from TW	Devine and Ghostly Voyage: Curse of the Ghost Pirate Ship	Devine and Ghostly Voyage 2: Pirates of the Caribbean	Devine and Ghostly Voyage 3: World's End
HK	魔盜王決戰鬼盜船	加勒比海盜：決戰魔盜王	加勒比海盜：魔盜王終極之戰
Reversal from HK	Demon Thief Lord v.s. Ghostly Pirate Ship	Pirates of the Caribbean: Duel with the Demon Thief Lord	Pirates of the Caribbean: The Final Battle of the Demon Thief Lord
ML	加勒比海盜：黑珍珠號的詛咒	加勒比海盜2：聚魂棺	加勒比海盜3：世界的盡頭
Reversal from ML	Pirates of the Caribbean: (The) Curse of the Black Pear	Pirates of the Caribbean 2: Soul Collecting Coffin	Pirates of the Caribbean 3: World's End

原名：	The Matrix (1)	The Matrix Reloaded (2)	The Matrix Revolutions (3)
Chinese (literal)	矩陣	矩陣重新載入	矩陣革命
TW	駭客任務	駭客任務2：重裝上陣	駭客任務3：最後戰役
Reversal from TW	Hacker's Mission	Hacker's Mission 2: Heavily Rearmed for Battle	Hacker's Mission 3: The Final Battle
港澳翻譯	廿二世紀殺人網絡	廿二世紀殺人網絡2：決戰未來	廿二世紀殺人網絡3：驚變世紀
Reversal from HK	Internet Murder of the 22nd Century	Internet Murder of the 22nd Century 2: Battle of/for the Future	Internet Murder of the 22nd Century 3: The Century of Startling Changes
大陸翻譯	黑客帝國	黑客帝國2：重裝上陣	黑客帝國3：矩陣革命
Reversal from ML	Empire of Hackers	Empire of Hackers 2: Heavily Rearmed for Battle	Empire of Hackers 3: Revolution of the Matrix

In my opinion, other worse examples include

Speed = distance ÷ time

• Hancock→台：全民超人、港：街頭超人、陸：全民超人漢考克
• I, Robot→台：機械公敵、港：智能叛變、陸：我，機械人
• Speed (1994)→台：捍衛戰警、港：生死時速、陸：生死時速
• The Rock→台：絕地任務、港：石破天驚、陸：勇闖奪命島

Although these translations are crappy, I must point out that good ones are hard to come by. And the 4 examples above are the perfect proof. Can you provide a better one?

Many bilingual people like to mock these translations. But has anyone analyze what makes good translation so hard?

One important factor is the differentiation between "literary" (L) and "colloquial" (C) in the Chinese language. There's a phenomenon among southern Chinese dialects called the "literary and colloquial readings" (文白異讀), that is the same character has a reading in a literary context, and another in a colloquial context. Using Min Nan which has the most number of such differentiations as an example, the reading from 1 to 10 are

Character	一 1	二 2	三 3	四 4	五 5	六 6	七 7	八 8	九 9	十 10
Lit.	it	jī/lī	sàm	sù	ngó͘	lio̍k	chhit	pat	kiú	si̍p
Col.	chi̍t	nn̄g	saⁿ	sì	gō͘	la̍k	(as above)	peh/pəeh	káu	cha̍p

(Note: Some special diacritics cannot be displayed in IEor Google Chrome. Firefox can)

When describing monetary amount, the colloquial reading is used. The "一" in "一百塊" is read as "chi̍t". When describing a telephone number, like "0912", it is read as the litarary reading "空kiú it lī". In addition, even a character as simple "人" (person/people) is read differently in "有人" ("there's a person") and "人民" ("people" in a socio-political context). The primary cause such differentiation is that colloquial readings are inherited from some time between the Three Kingdom (三國) period and the Southern and Northern Dynasties (南北朝) (220-589 AD), while the literary reading is a result of southern Chinese trying to imitate the northern standard dialect during the Tang (唐) Dynasty (618-907).

So what does the "differentiation between L and C" (文白異讀, DbLC) have to do with translation of modern English movie titles? Although DbLC is (almost) non-existing in Mandarin, the distinction still exists in people's mind. Otherwise, there wouldn't have been a New Culture Movement (新文化運動) some 90 years ago. That is to say it was not long ago when the educated Chinese population during the Ming (明) and the Qing (清) Dynasties (1368-1911) wrote literary text by imitating classical Chinese from the Qin (秦) and Han (漢) Dynasties (221 BC - 220 CE).

The title "I, Robot" is translated as "機械公敵" (Robotic Enemy of the States. Yes, 公敵 is part of it only because Will Smith played the protagonist in Enemy of the State) in Taiwan, and "我，機械人" in Mainland. Although "機械公敵" is not exactly the same as the original title, any Chinese native speaker will tell you that this title is more suitable as a title of an artwork; Although "我，機械人" is exactly the same as the original title, any Chinese native speaker will tell you that this title is not as suitable as a title of an artwork. Hancock is another good example. There are many more examples, the easiest being titles with only one noun, like The Mask, The One, or The Pianist.

The distinction between literary and colloquial is more pronounced in Chinese than in English. Therefore, if the original English title is a very commonly used word, then you can only choose between something that is faithful but too colloquial or something that is polished but distorted. (Of course, commercial incentive is another important factor for this chaos.)

So being bilingual doesn't really qualify a person to laugh at these translations, albeit not ideal, without examining the fundamental differences between the two languages and the collective habit on the choice of words between the two groups of speakers. In addition, there's no point with mainlanders mocking Taiwanese translation such as "捍衛" and "總動員" as a ghost of the past from Chiang Kai-shek's era, nor is there point with Taiwanese mocking HK translation as some cheap erotic or mainlanders' translation too colloquial. None of them are good at this.

---

Now I'm done with English to Chinese. Let's move on to Chinese to English.

There's this one time when I found out from newspaper that a certain media or group of people make fun of "unofficial" translations of Chinese dishes before Beijing Tourism Administration published 中文菜單英文譯法 (English Translation of Chinese Dishes) during the Olympics. (Another news report, CNN report). The most frequently mocked examples of unofficial translations include

• 童子雞 → Chicken without sex life
• 麻婆豆腐 → Bean curd made by a pock-marked woman

Although such translation is indeed inappropriate, I must yet again question what the root cause is. I propose the following

1. Chinese is a strongly "semantic-oriented" language
2. English is a highly hybrid language. When borrowing foreign words, the phonetic values are usually taken without looking at the meaning. (And if a borrowed word is written is Roman letters, then it is borrowed as-is)

Chinese is a strongly "semantic-oriented" language

When the Chinese world sees something alien, they usually first look at what it is, what its functions are, and what it means. Only when all those fail would the Chinese resort to phonetic borrowing. This phenomenon is quite unique in modern language transfusion. Even in Korean and Japanese, the two languages most influenced by Chinese, they both tend to borrow as-is in comparison to Chinese. These two languages are not as semantic oriented as Chinese.

JP	reading	KR	reading
エ	e	ㅅ	s
レ	re	ㄱ	g
ベ	he	ㄹ	l
ー	(long vowel)	ㅂ	b
タ	ta	ㅌ	t
コ	ko	final ㅇ	ng
ビ	bi	ㅡ	eu
ニ	ni	ㅏ	a
ス	su	ㅣ	i
ト	to	init. ㅇ	(silent)
ア	a	ㅔ	e
		ㅓ	eo

For example, elevator, invented in the West, is 電梯 in Chinese (tele-strair), "エレベーター" in Japanese (erebētā), and in Korean, it is sometimes "승강기", cognate to "昇降機" Chinese (elevator translated semantically), and sometimes "엘리베이터" (el li be i teo). Convenience store is "便利商店" in Chinese, a semantic translation, and "コンビニエンスストア" in Japanese (konbiniensu sutoa). Similar examples (where it's a semantic translation in Chinese and phonetic in Korean and Japanese) include escalator, remote control, allergy, and BlackBerry, the phone.

Because Chinese is a strongly semantic-oriented language, many Chinese use the same logic while exporting Chinese words. This is why 麻婆豆腐 (mapo tofu) becomes "bean curd made by a pock-marked woman".

English is a high hybrid language. When borrowing foreign words, they are usually borrowed as-is)

Sources of English words. Latin takes up 28%

On the contrary, there are numerous borrowed words. English is a member of the Germanic language. But as you can see from the pie chart, words from Germanic sources only take up 25%, while words from French and Latin encompass 28% each. Both of which are Latino-Faliscan languages, which are not so close to Germanic. Despite that, there are still many loan words from French and Latin. As a result, a concept can be represented by many different words. For example, "kingly", "regal", and "royal" all mean "of or relating to a king", and "kingly" is from Old English, "regal" is from Latin, and "royal" is from French (although ultimately also from "regalis" in Latin); "Water", "aqua", and "hydro" all represent water, and "water" is from Germanic, "aqua" from Latin, and "hydro" from Greek.

More examples:

Eng.	Chi.	Related	Chi.	Latin root	Meaning
body	身體	corporeal	有形體的	corpus	body
hand	手	manual	手動	manus	hand
fire	火	ignite	點火	ignis	fire
light	光	photon	光子	φως(希臘)	light
death	死亡	mortality	死亡	mors	death
year	年	annual	每年的	annus	year

What about Chinese things? "風水" is feng shui, not "wind water"; "豆腐" is tofu, fewer use "bean curd"; Even "炒麵" (fried noodles) is sometimes written as "chow mein" or "chao mian".

All the examples above point to one thing, that is English often borrows words directly as-is. Those that are unaware of this probably doesn't know "mapo" is acceptable.

---

White ____ ?

Diplomatic tussle: 'Aobama' or 'Oubama'? by The Washington Post is an unusual article about not politics, but linguistics. It shows the differences of opinion between the Americans and the (mainland) Chinese over some of the Chinese translation, which can be summarized as the follwing:

1. The Americans wants the Chinese to call White House "白屋" (lit. "white house"), and not the established "白宮" (lit. "white palace") because the U.S., as a federal republic, has no palace.


2. The Americans wants the Chinese to cal the current president "歐巴馬" (Ou-bama), where as the Chinese prefer "奧巴馬" (Ao-bama) because "奧" can mean "profound and deep" (深奧), and "歐" can mean Europe (歐洲). The Americans' rebuttal is that "歐" is closer to the actual pronunciation.

What do we see from this example? Although the Americans translators have learned the meaning of Chinese words and sentences, they are unaware of the distinction between literary and colloquial Chinese and the collective habit of native Chinese speakers. Second, both parties over-analyze the meaning of "歐" and "奧". In fact, either word carries only the phonetic value.

This article is quite long, but I have only 3 take-away points

1. The distinction between literary and colloquial words is stronger in Chinese than in English
2. Chinese is a strongly semantic-oriented language
3. English often borrows words directly as-is

Because of these reasons, translation can be a difficult task.

Learning a language is more than just knowing the meaning of words, mastering the grammar, and being able to live a life in that language. There's another layer of the cognitive process behind a language that needs to be understood.