Np Arguments And The Light Verb example essay topic

Orientation: Unselectiveness of Subject and Object In Mandarin Chinese In this paper, I attempt to argue that Lin Tzong-Hong!'s proposal on unselectiveness of subjects and objects in Mandarin Chinese and the theory of phrase structure is inadequate, in light of Minimalist Program, and expect to arrive at a neater and more unified explanation. Cross-linguistically, the canonical theta-role of subjects of verbs is agent and that of objects is patient. However, very often we see non-agent nominals take a subject position and non-patient nominals take the position of objects, namely, ! (R) the unselectiveness of subject and object! titled by Lin, which is very noticeable in Mandarin Chinese. For instance: (1) Unselectiveness of subject in Mandarin Chinese a. Laozhang kai-le yi-liang tanke-che.

(Agentive) p. n. drive-Perf one-Cl tank 'Laozhang drove a tank. ' b. Ga osu-gong lu-shang kai-zhe yi-pai tanke-che. (Existential) expressway-on drive-Dur one-line tank 'There is a line of tanks on the expressway. ' c. Zhe-liang po-che kai-de wo xia-si le. (Causative) this-Cl broken-car drive-Ext I scare-dead Prt 'Driving this broken car made me scared to death. ' (2) Unselectiveness of object in Mandarin Chinese a. chi niu-rou mian (Theme / patient ) eat beef noodle 'eat beef noodle' b. chi da-wan (Instrument) eat big-bowl 'use a big bowl to eat' c. chi guan zi (Location) eat restaurant 'dine at some restaurant' d. chi tou-tong (Reason) eat head-ache 'eat for [curing] headache' To account for this phenomenon in UG, Lin turns to event structure which is very much semantic in nature.

He puts such great emphasis on the event structures of verbs that he even claims that event structures determine the syntactic structure, not the vice versa. He bases his argument on the fact that inchoative's in English are quite restricted -- though some verbs undergo inchoative- causative alternation quite freely, some others do not, for example, (3) a. John broke the window. b. The window broke. c. The pigs splashed mud on the wall. d. Mud splashed on the wall.

(4) a. John put a book on the table. b. The book put on the table. c. We smeared mud on the wall. d.

Mud smeared on the wall. The problem is that there is an unexpected asymmetry between the! (R) break, splash! kind of verb and! (R) put, smear! kind of verb in that the former group of verbs exhibits transitivity alternation while the latter does not. He suggests that the eventuality predicate, which is CAUSE in this case, is a necessary element in the lexical specification of the smear type of verbs, but not for the splash type of verbs.

It is nothing other than the event structure that leads to the difference between the two group verbs. Therefore, he assumes that event structures correlate with syntactic structures in a determinant way. Turn to the main topic of unselectiveness in Chinese. Lin assumes that both the subject and object arguments in Chinese sentences are not selected by the main verb, but are introduced into the sentence via light verbs. He regards light verbs as eventuality predicates, and the account represented in (1) and (2) essentially hinges on the postulation of the specific light verbs, DO, EXIST, CAUSE, OF, USE, AT, FOR which are eventuality predicates semantically, in the phrase structures underlying (1) and (2). These light verbs host specific arguments, e. g., the agent, locative, and causer respectively in (1 a) and the patient, instrument, location and reason respectively in (1 b), at their specifier positions.

The main verbs kai 'drive' and chi 'eat' incorporate to the relevant light verbs, yielding the surface word orders that we see. The structural analyses for (1 a-c) and (2 a-d) are represented in the diagrams below. Selection of subjects in Mandarin Chinese (5)! - VP NP V V VP! -kai! - drive As shown in (5), the different subjects are licensed by different light verbs -- the agent licensed by DO, locative by EXIST, and causer by CAUSE. The subject-selecting light verb takes a VP as complement, and the main verb, in the present case kai 'drive', incorporates to the light verb. Thus, the surface unselectiveness of subject in Mandarin Chinese actually results from underlying selection by a light verb and the incorporation of the main verb to the light verb.

Next, let's look at the objects, The analysis is similar to the case of subjects: Selection of objects in Mandarin Chinese (6)! - VP V V VP NP V V VP V! -chi! - eat The case of the objects is basically the same as the subjects -- the different objects, i. e., the instrument, the location, and the reason, are licensed by the light verbs, USE, AT, and FOR, respectively. The main verb incorporates to them, and subsequently to the subject-selecting light verb. Lin further works on a principled explanation for the cross-linguistic variation seen among Chinese, Japanese, which exhibits unselectiveness of subject, but not unselectiveness of object, and English, which exhibits selectiveness of neither subject nor object. To capture the distinction, He resorts to L-syntax and S-syntax and proposes that in Chinese, the light verb structure is constructed at S-syntax, so that different syntactic light verbs can be freely superimposed on the same verb, yielding predicates of different types of eventuality, while in English, the light verb structure is constructed at L-syntax, resulting in rich inherent eventuality properties of English verbs, and Japanese falls between the two extremes. I suggest that there are one contribution but three deficiencies in Lin!'s theory.

The contribution is the introduction of light verbs in accounting for the facts and the instantiations of various types of light verbs, such as DO, CAUSE, EXIST etc. On the other hand, however, there are three deficiencies in Lin!'s theory that motivates revision and relativization. Firstly, Lin!'s framework is somewhat obscure. Although he claims to assume the Principle-and-Parameter approach proposed by Chomsky, he seems to have a foot in either camp syntax as well as semantics.

For example, to distinguish the smear-type verbs and splash-type verbs, which have identical categorical features, he resorts to event nature of verbs. Actually, this is a rather ad hoc point in that the difference in syntactic structure is the only evidence available, how can we put it aside and resort to some other aspect and even claim the eventuality structure determine the syntactic structure? To put it more explicit, just consider the Chinese counterpart of the two verbs, 'I? and 1/2 | which make the Chinese version of sentences (1) and (2) both acceptable, how can the event structure be accountable for this? Can the event of smear and splash vary cross-linguistically? Obviously not. Secondly, I observe there is a need to abolish the so-called L-syntax adhered by Lin.

Lin!'s adherence to it is motivated by the ability to account for the distinction between Mandarin Chinese, and English, with respect to the locus in the grammar where event structures are constructed. Since the decisive effect of event structures over syntactic structures does not stand firmly, the need to abolish L-syntax is just to follow the natural line. Thirdly, the tree diagrams proposed by Lin are rather confusing. Consider (5) and (6), there are two maximal projections in (5) which is concerned with subject selection, and three maximal projections in (6) concerning object selection. The main verb incorporates to subject-selecting light verb once in subject selection and twice in object selection.

This is an unexpected and undesired asymmetry as we have two separate structures in order to license subjects and objects. Uniformity is one of the major concerns in our research work. To obtain a more desired explanation, I make use of Niina Zhang 2002!'s observation: Zhang observes there is neither agent internal argument nor patient external argument, and if an external argument is not an agent, the internal argument must be patient, whereas if an internal argument is not a patient, the external one must be agent. That equals to say, providing agent and patient are canonical thematic roles of subject and object respectively, in one predicate structure, there are two possibilities: one is that all the arguments are of canonical thematic roles; the other is that one and at most one is of non-canonical thematic role. In pursuit of a more simplified as well as unified account, I make the following approximation in the frame of Minimalist Program, adopting the VP-shell structure, with a head vs. and a head V, for the account of both subjects and objects. In my analysis, Lin!'s various light verbs are considered as features of the light verb, which means the light verbs proposed such as DO, CAUSE, EXIST, OF, USE, AT, FOR etc are just bundles of features of the light verb.

Suppose the agent and patient selecting light verbs are of DO and OF feature, and they are interpretable features of light verb and cannot contribute anything to the crash of a sentence. Thus, there is no need for them to be checked or eliminated. On the other hand, the non-agent and non- patient arguments selecting light verbs such as CAUSE, EXIST, USE AT and FOR turn to be uninterpretable features in my assumption. They need to be checked and eliminated, otherwise, the sentence will crash. Correspondingly, the NP arguments have features too.

They are agent, patient, in accordance with DO and OF, and causer, locative, instrument, location and reason etc. in accordance with CAUSE, EXIST, USE, AT, FOR. The former two are interpretable while the latter are uninterpretable. Thus, in a certain domain of the syntactic structure, the NP arguments and the light verb may have the match relation. For a sentence to converge, the appearance of an light verb with an uninterpretable feature acquires an NP argument with an corresponding uninterpretable feature to match with it and eliminate the feature.

The features are! (R) one to all! Once one feature is checked and eliminated, all the other uninterpretable features eliminate together with it. For a sentence to converge, these features are to be checked before Spell-Out. The subject and object arguments also carry one of such features.

Thus, feature-elimination is possible in this argumentation under the agreement of light verb and NPs. And as Zhang observes, the non-canonical thematic role may exist in either subject or object, not both, the feature checking between vs. and one of the NPs, be it subject or object, is once and for all. the syntactic structure of predicates: (7) vP NP 1 vP vs. VP NP 2 VP V NP 3 To account for cross-linguistic variation, we set a parameter on vs. features. They can be strong, which need to be checked before Spell-Out, as in Chinese, or they can be weak, which may be procrastinated till LF, as in English. Instead of resorting to other aspects of grammar, we just do with the light verb proper.

Bibliography

Lin, Tzong-Hong. Light verb syntax and the theory of phrase structure. Doctoral Dissertation Niina, Zhang. Thematic Dependencies Between External and Internal Arguments of Transitives. Ouh alla, Jamal. 1999 Introducing Transformational Grammar: From Principles and Parameters to Minimalism.
P 139-142. Edward Arnold (Publishers) Limited Parsons, Terence. 1995 Thematic Relations and Arguments.