[讨论]语料库在研究中应该占多大比重?

[majorlv511]

有一点关于语料库的共识是:语料库是一个很有用的工具.
那么这个工具在我们的研究中应该占多大比重,如在一篇硕士毕业论文中?
前面看到有人做时态的分析,那么除了用语料库分析出时态出现的频率,在不同文体中的使用,还可以怎么用到语料库呢?

 

[动态语法]

回复：[讨论]语料库在研究中应该占多大比重?

个人的想法：

语料库的语料和开发工具都是研究的工具和手段，用多少要根据研究目的而定，很难说有定则。
对所研究的问题提出有用的见解是任何研究的最高目的，用语料库不过是为这个目的服务的。

 

[动态语法]

回复：[讨论]语料库在研究中应该占多大比重?

另外有两件事可以作：（1）适当地看些有关研究方法的论著；（2）多看些经典论文。

下面介绍挪威两位著名语料库语言学家的研究方法论：

Writing a Paper in English Corpus Linguistics1

By Stig Johansson and Hilde Hasselgård

http://folk.uio.no/hhasselg/Metode/termpaper.htm

 

[majorlv511]

最近一直在看语料库方面的书籍.也算是准备毕业论文开题吧.总觉得当自己的题目具体到某一个小方面后,语料库的运用不可能太多,不过就是得到一下数据,作为evidence,然后就不知道怎么办了

今年听过研三的答辩,有个师姐用了统计学,给了很多表格类的,结果答辩老师一开始就说"i m amazed by so many graphics there ".

 

[动态语法]

回复：[讨论]语料库在研究中应该占多大比重?

上面说了，做研究常常有两件事要做：（1）适当地看些有关研究方法的论著；（2）多看些经典论文。
现在附上两篇个人认为是比较好的论文，权作参考吧。

1）Thompson , Sandra A. and Anthony Mulac，1991. The discourse conditions for the
use of the complementizer that in conversational English. Journal of Pragmatics 15
( 1991) 237-251.

http://www.corpus4u.org/upload/forum/2005080310570532.pdf
http://www.corpus4u.org/upload/forum/2005080310582188.pdf

2) Carter, Ronald and Michael McCarthy 1999. The English get-passive in spoken
discourse: description and implications for an interpersonal grammar
English Language and Linguistics 3.1: 41-58.

http://www.corpus4u.org/upload/forum/2005080310591793.pdf
http://www.corpus4u.org/upload/forum/2005080310593533.pdf

[本贴已被 作者 于 2005年08月03日 11时04分14秒 编辑过]​

 

[majorlv511]

回复：[讨论]语料库在研究中应该占多大比重?

以下是引用 动态语法 在 2005-8-3 10:30:17 的发言：
另外有两件事可以作：（1）适当地看些有关研究方法的论著；（2）多看些经典论文。

下面介绍挪威两位著名语料库语言学家的研究方法论：

Writing a Paper in English Corpus Linguistics1

By Stig Johansson and Hilde Hasselgård

http://folk.uio.no/hhasselg/Metode/termpaper.htm

i 've already got the article, thanx

 

[xujiajin]

语料库在研究中应该占多大比重?

Writing a Paper in English Corpus Linguistics1
By Stig Johansson and Hilde Hasselgård
http://folk.uio.no/hhasselg/Metode/termpaper.htm
This paper is real good.

 

[动态语法]

回复：[讨论]语料库在研究中应该占多大比重?

前面说的是观察别人如何做，但是我们都知道做学问如游泳，
不能君子动口不动手，光说不练不行。先从小题目做起，例如可以试试
本站上贴的研究习作题目，用语料库材料做做，做多了自然就有心得了。

随便说说，仅供参考。

 

[majorlv511]

看过网站上的习作题目了,发现这些题目都是很DETAIL的比较了哈.

 

[xiaoz]

The corpus-based approach is necessarily a marriage between quantitative and qualitative analysis, and a balance between intuititions and corpus evidence. To what extent a study should and can be corpus-based largely depends upon how amendable your research question is to the corpus-based approach. Therefore, if you want to use the corpus methodology, you should learn first of all how to formulate a research question that is suitable for such a approach.

Tables, graphs, and formulae are quite common in corpus studies. The examiner's comment "i m amazed by so many graphics there" is quite positive.

The showcases we have posted so far are very specific to some language points, examining details. Are tense and aspect abstract enough? They can also be studied using corpus data.

 

[动态语法]

回复：[讨论]语料库在研究中应该占多大比重?

以下是引用 majorlv511 在 2005-8-3 15:51:06 的发言：
看过网站上的习作题目了,发现这些题目都是很DETAIL的比较了哈.

你的意思是说detail的没什么意思？

 

[wzli]

回复：[讨论]语料库在研究中应该占多大比重?

There are different users of corpora:
corpus referenced or supported: the corpora are used to support or illustrate some prefabricated ideas. The use of the corpus evidence is quite opportunistic as the process is quite selective: the user only incurs the instances that are supportive to the points made while ignoring all the other conflicting data. The only thing can be said of this is that the examples are 'real' as they are particularly picked up from a corpus rather than invented hot and fresh from one's own brain (I know some people do like this when they are so carried away to growing trees).

corpus based: usually one or more hypotheses are attested against the corpus data and the conclusion can be open. Most of the researches in language teaching and learning could be corpus based as we might expect our data to be as 'real' and 'authentic' as possible.

corpus applied: when one tries to apply the corpus in other fields for practical purposes, say classroom concordance, data driven learning, machine translation, one faces a wide area and a great variety of applied researches on corpus linguistics.

corpus driven: suggested by Sinclair and Tognini-Bonelli, it says something like you start from the scratch and allow yourself to be driven by the hard data. One never knows what he or she will be faced with until he or she gazes at the corpus data.

But personally I feel myself unconvinced by such a label "corpus driven". Can anyone really be driven by a corpus? All researches start from some ontological standing and certain ideas of what is going to be done. Whatelse, the perception, processing, and interpretation of the data all happen inside. But we might still use the term in our own interpretation: we may allow ourselves to be driven by a corpus to find out answers to the puzzles and questions we have on our mind specifically related to that corpus.

[本贴已被 作者 于 2005年08月04日 04时25分29秒 编辑过]​

[本贴已被 作者 于 2005年08月05日 01时55分33秒 编辑过]​

 

[xiaoz]

The sharp distintion between corpus-based and corpus-driven approaches is proposed by Tognini-Bonelli in her 2002 book "Corpus Lingusitics at Work" (Benjamins). However, the distinction is quite fuzzy in reality, as many have observed. If corpus-driven is defined as such in Tognini-Bonelli (2002), I would reject this term and use corous-informed instead. A further discussion of this distinction can be found in my posting at

http://www.corpus4u.org/upload/forum/2005070509413195.pdf

 

[wzli]

回复：[讨论]语料库在研究中应该占多大比重?

Yes, corpus-informed sounds more neutral. I love the term. Corpus is as a matter of fact one of the information sources.

 

[majorlv511]

回复：[讨论]语料库在研究中应该占多大比重?

以下是引用 动态语法 在 2005-8-4 2:27:58 的发言：

你的意思是说detail的没什么意思？

我是说这些到这种detail 的程度了,可挖掘的深度是不是很小了?
个人理解.

 

[majorlv511]

can we say "corpus can never play a dominant role in research, though it provides much real and authentic data?"

 

[xiaoz]

回复：[讨论]语料库在研究中应该占多大比重?

I bet you wouldn't say so when you have a better understanding of the corpus-based approach.

To give a simple example of using corpora in the area of English grammar, try to grab a copy of Biber et al's Longman Grammar of Spoken and Written English, or Hunston et al's Pattern Grammar, you will change your mind.

以下是引用 majorlv511 在 2005-8-4 10:51:06 的发言：
can we say "corpus can never play a dominant role in research, though it provides much real and authentic data?"

 

[xiaoz]

回复：[讨论]语料库在研究中应该占多大比重?

This may not be true. Someone has studied four very specific discourse markers (so, well, you know and like) and published a 282-page book. Can you say the four discouse markers are not "details" enough? It all depends upon your insight.

以下是引用 majorlv511 在 2005-8-4 10:47:07 的发言：

以下是引用 动态语法 在 2005-8-4 2:27:58 的发言：

你的意思是说detail的没什么意思？

我是说这些到这种detail 的程度了,可挖掘的深度是不是很小了?
个人理解.

 

[动态语法]

回复：[讨论]语料库在研究中应该占多大比重?

Personally I think small topics are more managable and can get your hands
dirty quickly. It's usually a good way to get one started.

以下是引用 majorlv511 在 2005-8-4 10:47:07 的发言：

以下是引用 动态语法 在 2005-8-4 2:27:58 的发言：

你的意思是说detail的没什么意思？

我是说这些到这种detail 的程度了,可挖掘的深度是不是很小了?
个人理解.

 

[chunling]

回复：[讨论]语料库在研究中应该占多大比重?

以下是引用 majorlv511 在 2005-8-3 15:51:06 的发言：
看过网站上的习作题目了,发现这些题目都是很DETAIL的比较了哈.

网站上的习作题目在哪里？