Chinglish: Officials have been trying to wipe out Chinglish, a quirky, nonsensical language combining both Chinese and English, off menus and road signs. One example? “Racist Park” signpost at a park celebrating ethnic diversity. Crimes against grammar range from spelling mistakes to paragraphs of total gibberish.
This is egregiously misleading on at least three counts. (It's also rather poorly written.)
First of all, the term "Chinglish," at least as it is used here, does not refer to a language; it is a (rather unfortunate) name for English text that has been badly translated from Chinese. It's the sort of translation you get when the translator is largely ignorant of either the source language or the target language, and is thus not in a position to do sanity checks to make sure that the translation actually means approximately the same thing as the original. (This happens a lot with unsupervised machine translation; if you're using a computer to translate text into a language you don't know, you might even end up mistaking an error message for a translation.) You look in your Chinese-English dictionary for an adjective meaning something to do with race; you find racist; you don't know what it means, and don't bother looking it up in the English-Chinese section of the dictionary to find out; and then all of a sudden there are a bunch of English-speaking tourists laughing at the sign outside your well-meaning park.
It's a bit like John Searle's famous Gedankenexperiment in which a person who speaks no Chinese is confined to a room containing a book with detailed instructions for manipulating Chinese characters. Chinese text comes in, and the person looks up what to do with it, performs the necessary operations, and sends out the reply. The person in the room appears to be communicating in Chinese, but in fact has no comprehension of the content of any of the messages. Searle's point was to suggest that even if we someday manage to make a computer that fluently manipulates natural language, that still won't mean that it understands anything. But if we replace the wonderful book in Searle's room with an ordinary bilingual dictionary (and swap languages), this approximates the situation of the hapless translator rather nicely.
There's nothing particularly special about "Chinglish" as compared to bad translations between other pairs of languages. (This blog, for example, is named after a mistranslation of my own that could perhaps be called an example of Czenglish.) If bad Chinese-to-English translations are more common than some other types, this may be because (1) the two languages are not related to each other, so the translator gets no help from cognates, (2) the two languages use different writing systems, so it takes more work to learn to read both of them, and (3) there are more Chinese-to-English translations than there are English-to-Chinese translations out there on the menus and signposts of the world. (Englinese is no less risible than "Chinglish"; it's just less common.)
In any case, "Chinglish" is not a language in the formal sense, because it would be absurd to ask whether a given sequence of words is or is not a well-formed utterance of Chinglish. And it's not a language in the functional sense, because the people who write it don't know what it says, and the people who read it don't know what it means.
Secondly, "Chinglish" doesn't really combine English and Chinese. The words are English, and they're influenced by Chinese only indirectly. There are certain patterns of substitutions and other errors that tend to come up because of ambiguities in possible correspondences between Chinese characters and English words—for example, 'dry' is often rendered as 'fuck'—but that's not at all the same thing as "combining
both Chinese and English."
Thirdly, spelling is not grammar, dammit. (And I don't much care for this rather tired metaphor of "crimes against grammar" in any case, but it seems especially inappropriate in a context where the infractions are clearly unintentional.)
Why does any of this matter, apart from my general preference that my morning paper report truths rather than falsehoods? Aren't these errors fairly trivial? Well, it matters because this sort of nonsense has the potential to confound the way people think about real languages, and here I am thinking in particular of Singlish. Singlish really is a language:2 people speak it, on purpose, and use it to communicate with one another, and fluent speakers can readily distinguish sensible Singlish sentences from gibberish. Singlish combines elements of English and Chinese (and Malay and Tamil and a few others besides). I believe it could reasonably be characterized as "quirky." (What natural language could not?)
So part of the trouble with the newspaper's account is that their description of "Chinglish" makes it sound very much like Singlish (except, of course, for the part about being nonsensical). If Reuters and the Toronto Star can't be bothered to differentiate between "a language that combines elements of other languages" and "a collection of bad translations from one language to another," then they are contributing to the misconception that mixed languages are somehow degenerate or inferior.3 To mistake a mistake for a language makes it easier to mistake a language for a mistake.
1. The main article to which it was adjoined was about attempts to reduce air pollution in Beijing, also for the Olympics.
2. In calling Singlish a language, I don't mean to get into such vexed questions as how to distinguish between a language and a dialect, or when to say that an interlanguage has become a pidgin, or a pidgin has become a creole. All I mean is that Singlish is very clearly something that "Chinglish" is not.
3. Consider, for example, the Singaporean government's effort to get people to speak English instead of Singlish; what I read in the newspaper obscures the differences between this and the Chinese government's effort to improve the quality of the English translations on public signage.