This growing meaning potential warrants a fresh, unmechanical conceptualization of China English. The language should not be understood mechanically as bearing a number of Chinese syntactic and pragmatic norms or as having “normative English as its core” plus “Chinese characteristics in lexicon, syntax, and discourse.” This characterization assumes that Chinese and English elements are easily separable, as the inference model seems to imply. Since English is used by numerous Chinese in new contexts and domains, it will undoubtedly develop a rather sophisticated, self-sustaining linguistic system. Rather than viewing the new variety of English against a native-speaker norm, it may better to view it as a new system based on “elements, structures, and rules drawn from both English and from one or more languages used in the environment” (Kandiah, 1998: 99). These elements, structures, and rules will be fused so seamlessly that it might at times be difficult to pinpoint what the Chinese characteristics are in this new variety of English. In my analysis of the bulletin board threads, I have identified patterned rhetorical strategies. Which ones can we be certain are truly influenced by Chinese discourse, and thus can safely call rhetorical strategies with Chinese characteristics? Therefore, identifying Chinese characteristics becomes less important than observing and describing the meaning potential of China English – what Chinese people can mean and can do with English in new contexts and domains.
This is what I've been thinking for a while. Searching for Chineseness (cf. Margie Berns' AILA paper) will only get you so far -- in fact, it won't get you far at all, except in essentializing, or if you want to stay on the "Chinglish = L1 interference that must be eradicated" bandwagon.
I highlight You's last point because what one "CAN" mean or two has two different possibilities: first, there's what you "can" do in the sense of literally what you are able to do. Like, if you want to say "I love you" in English instead of Chinese, because of whatever reason, you are actually able to do that. However, there is also what you're allowed to do, or what is (broadly) accepted by (educational/linguistic/etc) gatekeepers. There's an analogy here to acceptability and the "error vs innovation" distinction. You CAN write or say whatever you want in English. How other people will take it up is not up to you, yet you can use your knowledge of acceptability to your own advantage when choosing how to express yourself.
Thus, knowledge of 'acceptability' latches on to Canagarajah's pedagogy of "shuttling between communities" ... knowing how people are going to view your use can help you make decisions. See also the Matsudas' recent piece on L2 writing pedagogy and WEs.
Of course, I'm still assuming a certain stability, if not in usage, then at least in acceptability. And I think the ideology of acceptability is probably a lot more stable than usage, because what is usage anyway? You can always say whatever you want. It's always changing.
I might be making acceptability into too much of a binary --- acceptable vs unacceptable -- because obviously people react to things in idiosyncratic ways. This actually goes all the way back to J.R. Ross whose 1979 article "Where's English?" attempted to quantify acceptability and came up with some pretty weird attempts to answer that question.
Anyway. Enough rambling.