can anybody understand this example?

[patricx]

can anybody understand this example? i am very confused

For instance, in example 10-3, how can ungrammatical utterances be distinguished from ones that simply haven't occured yet? if our finite corpus does not contain the sentence (a), how do we conclude that it is ungrammatical, while those in (b), (c), and (d) are grammatical?

*a)He shines Tony books.
b)He gives Tony books.
c)He lends Tony books
d)He owes Tony books.

[动态语法]

Looking to corpora for grammaticality judgement is the
wrong use of corpus data. Corpus analysis is about tendencies
and patterns of language use, something that even the native
speaker may not realize. If one is more intersted in theoretically/
methmatically/logically/propositionally possible or impossible
grammatical strings, then there is no or little need for corpora.

[patricx]

thanks, this example tries to illustrate the limitation of the corpus-based methods.
i quoted from LINGUISTICS:A COURSE BOOK(2ND EDITION) P.338

[动态语法]

以下是引用 patricx 在 2005-7-26 6:44:32 的发言：
thanks, this example tries to illustrate the limitation of the corpus-based methods.
i quoted from LINGUISTICS:A COURSE BOOK(2ND EDITION) P.338


Then it is the problem of the author(s) of Linguistics. Many people
do misunderstand the utility of corpus linguistics.

[patricx]

Sorry, i forget to write the author of the book(LINGUISTICS:A COURSE BOOK(2ND EDITION) P.338), esp who contribute that chapter i mentioned. they are 胡壮麟 and 彭宣维.

[xiaoz]

The examples quoted are actually from McEnery and Wilson (1996, 2nd 2001).

Corpora are not about what is possible or correct, but about typicality and central tendency in attested language use.

Corpora do not provide negative evidence, which means that if you cannot find an example or pattern in a corpus, you cannot say such examples and patterns do not exist or are incorrect.

In this connection, many authors, including those you mentioned, have misunderstood the quantitative nature of the corpus-based approach, though this approach does not reject qualitative anslysis. As such, the corpus-based approach has achieved more accurate language descriptions which are impossible with quan or qual analysis alone.

以下是引用 patricx 在 2005-7-26 7:09:07 的发言：
Sorry, i forget to write the author of the book(LINGUISTICS:A COURSE BOOK(2ND EDITION) P.338), esp who contribute that chapter i mentioned. they are 胡壮麟 and 彭宣维.

[patricx]

exactly right. but when i read Mr. HU( Zhuanglin)'s linguistics course book, i often can't get the exact meaning, sometimes the words are unclear. such as the example above.

[oscar3]

I had exactly the same experience as patricx when I read Hu's earlier version , I even doubted my own reading comprehension. Later I found a course book of linguistics published in the USA, Which, in my point of view, is much better than Hu's. Why can't we simply use the course book published in USA or UK if we can write a better one?

[动态语法]

以下是引用 oscar3 在 2005-7-26 9:47:57 的发言：
Why can't we simply use the course book published in USA or UK if we can write a better one?


That would be my suggestion to anyone who can read English well. It's
not to suggest that there are no good books in China but that sometimes
it makes a lot of sense just to read the original text. This is especially
important if the text in question is a classic one. It's better to read the
orignal one than reading secondary sources.

[xiaoz]

Absolutely.

One difficulty though - Books published outside China may cost a fortune. Some people told me that my aspect book would cost over 1000 Chinese yuan - their pockets would be happier if they could copy the book somewhere.

[patricx]

thanks for your discussion. and it's a good news for us: Foreign Language Teaching and Research publishing house has already introduced 112 classic origianl linguistic books as Peking publishing house and Shanghai foreign language teaching publishing house do.