Print

Print


On Mon, 14 Jan 2013 20:23:51 -0500, MorphemeAddict <[log in to unmask]> wrote:

>On Mon, Jan 14, 2013 at 8:18 PM, George Corley <[log in to unmask]> wrote:
>
>> On Mon, Jan 14, 2013 at 7:09 PM, David Peterson <[log in to unmask]> wrote:
>>
>> > I found a paper that does a fairy summary of the types of compounding,
>> and
>> > provides some interesting examples of natlang compounds (e.g. apparently
>> > the word for "thing" in Mandarin is a compound of "east" + "west"). You
>> can
>> > read it here:
>> >
>> >
>> >
>> http://www3.nacos.com/lsj/modules/documents/LSJpapers/journals/135_scalise.pdf
>>
>>
>> Yes.  I'd forgotten how odd that was when I learned 东西.
>>
>
>I suspect that the word was coincidentally pronounced the same way, and so
>the characters for "east" and "west" were arbitrarily associated with it.
>I doubt it's originally really a compound, although now, of course, it is,
>at least in writing.

What do you mean by "really a compound"?

Anyway, my money would be on that this compount really was semantically "east" + "west", originally, and I can imagine a development like "east and west" > "from east to west" > "the whole range" > "every (kind of) thing" > "any (kind of) thing" > "thing".  Your coincidence hypothesis could be ruled out if we have early enough attestations of the compound when fewer characters have merged.  In theory it might could be ruled out by attestations in other varieties with different mergers, but graphic borrowings might also cause that so I'd be cautious (maybe depends whether it's a colloquial or literary sort of word in those other varieties).  

-- Ah, someone purports that http://zh.wikipedia.org/zh-cn/%E4%B8%9C%E8%A5%BF contains the actual etymology.  But I can't read it to judge.

Alex