人物专栏 | Kwang-Sup Kim教授访谈（下）

点击上方蓝字关注我们

编者按

《理论语言学五道口站》（2021年第69期，总第203期）“人物专栏”与大家分享本站采编人员王平对Kwang-Sup Kim教授进行采访的访谈录。Kwang-Sup Kim（金光變），美国马里兰大学帕克分校博士，现任韩国外国语大学理论语言学、英语语言学教授。主要研究兴趣为句法理论、比较句法、句法与语义接口以及句法与音系接口。

本期访谈中，Kwang-Sup Kim教授首先区分了所有格和属格的强式和弱式形式，然后分析对比了双宾语结构和与格结构的差异，最后对移位和话题化进行了讨论。

Kwang-Sup Kim教授简介可参考《理论语言学五道口站》“人物专栏”2021年第64期，总第198期。

访谈内容

05.

王平：您在研究中提到，所有格和属格都受制于同一语音条件，并且依据其语音成分的不同，可将它们分为强式形式或弱式形式。您可否给我们举例说明一下二者的区别？

Kwang-Sup Kim教授：有些成分不能独立存在，必须依附于另一个成分。例如，my不能单独出现，必须依附于某些名词。而mine则不同，它可以单独出现。通常来说，以辅音结尾的词为强式，以元音结尾的词为弱式。如(ia)中的所有强式形式都是以辅音结尾的，弱式形式的末尾则是元音或/r/。

(i) a. 强式形式：mine, ours, his, hers, theirs,

     b. 弱式形式：my, our, her, their,

在英式英语中，单词末尾的/r/会被省略，也就是说它不发音。这意味着在英式英语中，弱式形式的结尾是元音，而强式形式是辅音。在美式英语中，/r/虽然发音，但似乎比其他辅音要弱一些。

06.

王平：“论元一致原则”是指如果一个谓词有两个论元，这两个论元就存在语义关系（Kim 2012）。然而，有人认为同一个谓词的两个论元未必相关。对此，您能否举例证明一下您的观点呢？

Kwang-Sup Kim教授：一般来说，很难想象同一谓词的两个论元在语义上没有关系。但如果两个论元没有语义关系，可能是有介词涉及其中。如以下两个韩语句子。

(i)  a.   Susie-ka      kom-ul  sso-ass-ta.

        Susie-Nom  bear-Acc  shoot-Past-Dec

              ‘Susie shot the bear’

       b.   Susie-ka kom-eykey   sso-ass-ta.

        Susie-Nom   bear-to   shoot-Past-Dec

              ‘Susie shot at the bear’

如果韩语“kom”（熊）被赋予宾格，如(ia)中所示，意味着熊一定被射中，而在(ib)中，它被介词“eykey”（to）修饰，则表示熊不一定被射中。但是有相当多的例子表明，同一谓语的两个论元一定具有语义关系。同一谓语的两个论元没有语义关系的这种情况很少见，我认为这种情况很可能不存在。

07.

王平：在讨论双宾语结构和与格结构的差异时，您主要用二分原则和论元一致原则进行了论述，您能否从体态结构三维模型（Causal-aspectual Structure）解释一下二者的差异？

Kwang-Sup Kim教授：在双宾结构中（DOC），两个内部论元遵循论元一致原则的语义关系。而在与格结构（DC）中，直接宾语和与格短语没有直接关系。

(i)  a. DOC: [External Argument v [VP Indirect Object V Direct Object]

    b. DC: [External Argument v [VP [Direct Object V] Dative Phrase (to-phrase)]

也就是说，在双宾结构中，动词短语VP表示状态，而在与格结构中却不是。让我们回顾一下轻动词的类型。如果一个轻动词选择一个事件或状态作为它的补足语，那么它可以被当做致使动词，其外部论元就可以表示使某个事件或状态发生。但如果一个轻动词的补足语与之无关联，它就不能充当致使动词。如下所示，相比与格结构，致使含义只在双宾结构中存在。

(ii)  a. The explosion gave Beth a headache.

    b. *The explosion gave a headache to Beth.

(iii)  a. His behavior gave Mary an idea.

     b. *His behavior gave an idea to Mary.

(iv)  a. The medicine brought him relief.

     b. *The medicine brought relief to him.

08.

王平：您认为移位和话题化有何联系呢？

Kwang-Sup Kim教授：一个句子成分要成为话题有两种方式：一个是将其移位至句首位置，另一个则是对话题进行话题标记，韩语中两种方式均有。话题分为两种：普通话题和对比话题。如果一个话题位于句中，那么即便其被标记也不能看作是普通话题。换言之，它只能是对比话题。一个成分只有经过话题化移位到句首位置才是真正的话题。尤其是很多语言中不存在话题标记，因此就必须要经过话题化移位。

参考文献

Kim, Kwang-Sup. 2012. On the Differences between the Double Object Construction and the Dative Construction. The Journal of Studies in Language,28(3): 421-447.

English Version

05.

Ping Wang: According to your study on the Ambiguity of the So-Called Possessive Marker ’S in English, both the Possessive Marker and the Genitive Case Marker are subject to the same phonological condition: they are realized as either a strong form or a weak form, depending on whether they are followed by a phonologically non-null constituent or a phonologically null constituent. Could you give us some examples to show how can we distinguish between strong form and weak form?

Prof. Kwang-Sup Kim: Some constituents cannot stand on its own but must be attached to another constituent. For instance, my cannot stand alone. It must be dependent on the nominal that it co-occurs with. On the other hand, mine is phonologically independent of another word or phrase. If a word ends in a consonant, it is usually strong, and if a word ends in a vowel, it is weak. Notice that in (ia) all the strong forms end in consonants. By contrast, the weak forms end in either a vowel or /r/.

(i) a. Strong forms: mine, ours, his, hers, theirs,

     b. Weak forms: my, our, her, their,

In British English /r/ is dropped in word-final position: that is, it is not pronounced. This means that in British English weak forms end in vowels, while strong forms end in consonants. In American English /r/ is pronounced, but it seems that it is weaker than the other consonants.

06.

Ping Wang: The Principle of Co-arguments means that if a predicate has two arguments, the two arguments bear a semantic relation (Kim 2012). Some linguists, however, believe that two arguments of a same predicate are not necessarily related. So, could you give us some examples to consolidate the principle and to explain the importance of such a principle?

Prof. Kwang-Sup Kim: It is hard to imagine the case in which the two arguments of a predicate are not semantically related. If two arguments are not semantically related, it is likely that there is a preposition involved. Let us consider the following two Korean sentences.

(i)  a.   Susie-ka    kom-ul   sso-ass-ta.

     Susie-Nom     bear-Acc  shoot-Past-Dec

              ‘Susie shot the bear’

     b.  Susie-ka   kom-eykey  sso-ass-ta.

       Susie-Nom   bear-to  shoot-Past-Dec

              ‘Susie shot at the bear’

If kom ‘bear’ is assigned accusative Case, as in (ia), the bear must have been shot, and if it is marked by the preposition eykey ‘to’, as in (ib), the bear may or may not have been shot. There are numberless examples that the co-arguments must bear a semantic relation. It would be interesting that there are some cases in which the co-arguments of the same predicate are not semantically related. But I doubt that there are.

07.

Ping Wang: As you have done researches to explain differences between the Double Object Construction and the Dative Construction following from the Binarity Principle and the Principle of Co-arguments. Could you explain these differences through Causal-aspectual Structure? 

Prof. Kwang-Sup Kim:In the Double Object Condition (DOC), the two internal arguments must bear a semantic relation in accordance with the Principle of Co-argument. By contrast, in the Dative Construction (DC), the Direct Object and the Dative Phrase do not bear a direct relation.

(i)  a. DOC: [External Argument v [VP Indirect Object V Direct Object]

      b. DC: [External Argument v [VP [Direct Object V] Dative Phrase (to-phrase)]

This means that in the DOC construction, VP denotes a state, while in the DC, it does not. Let us recall that there are various types of light verb. If a light verb takes an event/state-denoting complement, it can be used as a causative verb; it can denote that the external argument can cause the event/state to take place. If, on the other hand, a light verb takes a non-relational complement, it cannot be used as a causative verb. This prediction is borne out. The following pair of sentences illustrate that only the DOCs, but not DCs, permit the causative reading.

(ii)   a. The explosion gave Beth a headache.

         b. *The explosion gave a headache to Beth.

(iii)  a. His behavior gave Mary an idea.

     b. *His behavior gave an idea to Mary.

(iv)  a. The medicine brought him relief.

     b. *The medicine brought relief to him.

08.

Ping Wang: According to your research, what is the relation between movement and topicalization?

Prof. Kwang-Sup Kim: There are two strategies to make a constituent a topic. One is to move a topic to a clause-initial position, and the other is to attach a topic marker to the topic. Korean makes use of both strategies. There are two types of topics: neutral topic and contrastive topic. If a topic remains in the middle of a clause, it cannot be interpreted as a neutral topic although it is attached by a topic marker: that is, it must be interpreted as a contrastive topic. If a constituent is to be a genuine topic—a neutral topic, it must undergo topicalization to a clause-initial position. Especially, there are many languages with no topic marker. In these languages, topics must undergo topicalization.

References

Kim, Kwang-Sup. 2012. On the Differences between the Double Object Construction and the Dative Construction. The Journal of Studies in Language, 28(3): 421-447.

往期推荐

本文版权归“理论语言学五道口站”所有，转载请联系本平台。

编辑：闫玉萌 赵欣宇 雷晨 

排版：闫玉萌 赵欣宇 雷晨

审校：李芳芳 田英慧