When translating Chinese to English, most translators today can do little more than copy and paste a machine translation as their target text. I’ve helped many corporations and government agencies clamp down on this issue and have translators attempt to do human translations, which exposes a new problem for this population: reliance on hyper-literal, word-for-word translations. While it is tempting to say that the root cause of the problem is a lack of intelligence, I believe that the fundamental issue is rooted in the medieval Universal Grammar theory, which Noam Chomsky attempted to validate through scientific experimentation (before later revising many of his earlier conclusions). I believe that ideology, not intelligence, is the problem because I have also taught a number of translation students who, within as little as a year or two, are able to transcend the “Chinglish” phenomenon. Competent Chinese translation services require overcoming the universalist ideology and adopting contemporary science-based translation practices.
Case Study: Out-of-Context Cherry Picking
A recent case study on the problem of cherry picking comes from translators tasked with translating legal documents related to local eminent domain law in China. The local law, much like American local law, requires posting highly visible public notices outside the property and publication in local newspapers. The intent for the particular locality was to attract foreign investors by incorporating legal protections similar to NAFTA’s anti-expropriation clauses and allowing factory investors to be secure in their properties. Nonetheless, the translators working on the translation described providing notice of an eminent domain action as an “announcement.” The reason for their decision was that they were looking at a White House press release of Joe Biden’s announcement of forthcoming executive orders to protect against COVID-19.
Mandarin Chinese does indeed have distinctions between announcement, publication, and notice (xuanbu vs. gongbu, 宣布 vs. 公布). Still, the precise meaning of each of these words in various contexts does not overlap perfectly with particular English words. However, Chinese translators generally express a strong belief that there are linguistic universals in which there is an absolute 1-to-1 correspondence between these words (this population excludes a handful of the most famous translators in China). When completing translations between Chinese and English, this results in the “Chinglish” phenomenon when translated to English or “Translationese” when translated to Chinese. This belief in universals is leading translators to produce translations that are completely unlike language seen anywhere else in the universe.
Ancient Universals Theory
The theory of linguistic universals formally goes back to the 13th century with Roger Bacon’s early books on linguistics that investigated ancient languages and their connection to English. The theory has a long history you can read about on Wikipedia and is a tradition that has been very influential on foreign language instruction and translation. Translators in East Asia seem to be first indoctrinated into this theory in primary school. They are made to memorize word list pairs associating Chinese (or Japanese) words with English words and are falsely taught about the absolute equivalence between the words as if each word directly referred to a concept in Plato’s Theory of Forms.
Chinese translators applying this theory like to make statements that one word in Chinese is the exact equivalent of another word in English. Simultaneously, they seem to recognize widespread public criticism that translations generally make little sense but also believe their own translations are highly perfect. As in the eminent domain case study described above, these translators never look at things like genre or context to determine how different communities use words. They are usually unaware of things like the fact that stock exchanges use the word “companies” while the internal revenue service refers to “entities,” even when talking about the same things. Those translators will always call both “enterprises” — because that’s what the US Bilateral Investment Treaty says.
More than a discrete ideology, universal grammar among translators seems to be a default and intuitive assumption about how languages work and are developed by societies. A good case study of this is how the anthropological school of linguistics got started during field work on small Pacific islands. Anthropologists could not understand anything of local discourse through word-for-word translations that assumed universal equivalence through words. Only when they studied the daily experience and context of these Pacific Islanders were they able to decipher the meaning of these words.
Example: Government Agency Drift
A useful case study of just how extraordinarily misleading and confusing the intuitive universal grammar theory is can be found in the translation of governmental subdivisions in China. In particular, Chinese has the words jigou and jiguan (机构、机关) that refer to subdivisions of government and exclude natural persons. These words are usually translated to government “organ,” which is rarely used in English laws except in international treaty law, where an “organ” is used to include natural persons. Basically, it is the opposite of the Chinese word it purportedly stands for as an exact equivalent under the universal theory.
More extraordinarily, the Chinese words misidentified as “organs” were actually loanwords borrowed from English in the 19th century. Much like the children’s game of Telephone, where a message gets increasingly corrupted as it is translated, the original 19th-century English has become extremely corrupted by Chinese translators. You can easily see this for yourself on a highly convenience-oriented resource such as Wikipedia. A search for government agency on Wikipedia gives the following definition: “A government or state agency, sometimes an appointed commission, is a permanent or semi-permanent organization in the machinery of government that is responsible for the oversight and administration of specific functions, such as an administration.” The words jigou and jiguan (机构、机关) are actually metaphors for parts of the machinery of government. These were translated into Chinese from British and American texts and adopted by Chinese public policy experts in the 1940s.
However, when the English language was updated in the 20th century to incorporate a highly standardized language of government agencies, bodies, offices, and their supervising departments and ministries, Chinese translators failed to take note of the new English vocabulary. Rather, they relied on highly specialized treaty language that, relying on dictionaries, they misunderstood and created translation guidelines that created a false equivalence between unrelated terminology. More shockingly, they did it for Chinese language that was originally introduced from English. That is to say, despite the common origin, Chinese translators still missed it and identified a totally unrelated word as an absolute equivalent.
Apply Contemporary Science
Like other sciences, linguistics has achieved remarkable advances throughout the 20th century, and some of it is directly pertinent to Chinese to English translations. I’ll provide a very brief summary — this material is worth mastering for any translator — before getting into the specific application. In the mid-20th century, MAK Halliday followed up on WVO Quine’s challenge, specifically about the ontological relativity of word meanings specific to Japanese-Chinese and English, while studying at Peking University on an exchange from Cambridge. Halliday’s answer to Quine’s challenge about the incongruence of Chinese measure words created an entirely new science, Systemic Functional Linguistics, wherein Halliday discovered that, while there is no overlapping vocabulary where Chinese measure words are deployed, a series of words could establish indexicality. Thus, providing an answer to the “three cows” language reference challenge that Quine proposed exists in Chinese and Japanese. Subsequently, neurologists investigating Halliday’s answer to Quine studied Chinese schoolchildren and compared their brain development to English-speaking children. They discovered that not only did the Chinese children’s brains develop differently, but it also took their brains longer to develop to acquire the capability to use Chinese measure words. That is to say, not only were the words not different, but the corresponding brain structure responsible for using the words did not exist for non-Chinese speaking populations in England.
When applied to actual translation work involving legal and business documents, the nature of the translation task changes dramatically. Under a universal theory, translators in practice simply look words up in a popular dictionary and attempt to match definitions to other words. This may work for languages within a single language family, but for East Asian languages with their own distinct traditions, this led to the creation of a variety of sham Englishes such as “Chinglish” and “Engrish.” Looking at Department of Justice statistics, it appears that reliance on this translation approach causes a 70% failure rate for attorneys whose cases otherwise fail no more than about 1% of the time. Organizations place great trust in translators, who are sabotaging their work without much notice.
The theories derived from Halliday’s work at Peking University and Cambridge led to a highly specific analytical approach devised by Norman Fairclough in the 1990s, which has been adapted specifically to legal translation by Catherine Way. Instead of trying to look at universal definitions of words, which is pseudoscientific, contemporary science-based theories look at the context, social relationships, history, culture, and indexicality. For example, in the above case study on the machinery of government, I look at the shared history of the English and Chinese terminology and the context in which the terms are used, i.e., whether used in the treaty context or the domestic bureaucracy context. Finally, I also consider the indexicality issue, specifically what the word is intended to signal in each case. The word “organ” is meant to include individual natural persons, and the Chinese word jiguan is intended to exclude individual natural persons. To make this point even more explicit, the Chinese word is a metaphor for a machine part – artificial and not natural, therefore not a natural person. Words like agency or instrumentality in English correspond much better; the word “organ” used as its translation has no historical, contextual, or indexical correspondence at all. Little wonder, then, that native English speakers complain they cannot understand Chinese administrative law regulations. It is not because they are ignorant of Chinese culture, but rather that the translation vocabulary is wrong.
Conclusion
The linguistic universals theory, while highly intuitive and widely adopted among translators of Chinese to English, is ultimately pseudoscientific, as many critics have pointed out. As Quine has pointed out, and neuroscience later verified, Chinese has grammatical forms that do not exist in English and therefore talks about abstract entities not recognized by English. After decades of relying on this theory, consumers have consistently reported that the output is unintelligible, unusable, and downright misleading. Instead of looking for universals, modernize your perspective on translation to focus on the history, culture, context, and indexes of the language being translated. Only this kind of translation service can be genuinely valuable to a client.