※ 清華語言所新034次專題演講公告 ※
題目：I have a V. I have an O. Um, V-O compound!
In this talk, I will address the issue of two-way V-O compound verbs﹝兼類述賓複合動詞﹞in Mandarin Chinese. The two-way V-O compound verbs have been observed to involve two syntactic patterns. That is, the object can remain in the standard postverbal position, or to be introduced by a preposition preverbally. However, I argue that the preverbal PPs in two-way V-O compound verbs do not behave like typical PPs in other V-O compound verb cases whose objects can only be introduced by prepositions. The seeming prepositional phrase in two-way V-O compound verbs, on the other hand, is proposed to be a combination of differential object marking and an internalized object NP. In addition to the theoretical issue, I will also discuss the relevant V-O compound verbs in the field of teaching Chinese as a second language. The characteristics of identifying different V-O compound verbs will be proposed, and the teaching order regarding the two syntactic patterns of two-way V-O compound verbs will also be suggested based on the search results in the Sinica Corpus.
※ 清華語言所新035次專題演講公告 ※
題目：State / event contrast and aspectual marking in Mandarin
This talk provides evidence for a non-unified treatment of the semantics of eventive and stative predicates. We investigate Mandarin sentences containing eventive vs. stative predicates and show that while stative construals of stative VPs do not require aspectual marking, eventive (episodic) construals of eventive VPs require overt aspect.
We adopt the (classic) Davidsonian approach assuming that states and events have different argument structures (Davidson 1967, Katz 1995, 2003, Kratzer 1998): bare states are properties of intervals (type <i,t>), true or false for a time, while bare eventives are properties of events (type <v,t>). So how do we derive properties of times from properties of events? Via aspect, an operator that maps properties of events onto properties of times. This well-established analysis together with our claim that aspect must be overtly realized in Mandarin in single-predicate root clauses automatically predicts the difference in terms of aspectual marking with stative/eventive predicates in Mandarin.
The argument structure hypothesis (in the sense of Katz and Kratzer) is based mainly on English, where the perfective aspect is not morphologically realized (1-2). In Mandarin, the contrast predicted by this analysis is much more straightforward, since the perfective aspect is overt (4). This makes Mandarin a perfect illustration of this analysis.
(1) a. John loved Mary. b. [ti PAST] [John love Mary]
(2) a. John kissed Mary. b. [ti PAST] [PERF [John love Mary]]
(3) a. Nèi-shíhou Lǎomóuzi xǐhuān Gǒnglì b. [TP that-timei [TP tj [VP L love G]]]
that-time Laomouzi love Gongli
‘At that time, Laomouzi loved Gongli.’
(4) a. Lǎomóuzi qīn *(le) Gǒnglì b. [TP tj [AspP le [VP L kiss G]]]
Laomouzi kiss PERF Gongli
‘Laomozi kissed Gongli.’
The event argument of “action” verbs was proposed by Davidson and defended by Katz to explain asymmetries in adverbial modifications. We show in this talk that aspectual marking in Mandarin provides evidence for a non-unified semantic treatment of states and events, thus favor a classic Davidsonian approach.
Finally, we discuss sentences with a bare eventive predicate that are nevertheless well formed. We show that when it is the case, they necessarily yield habitual readings. (Habitual sentences refer to sentences describing regularities or make a generalization over instances of events (Carlson et al. 1995) In the presence of a quantificational adverb (Q-adv.) (5a), a locative PP (5b) or other adverbials (5c), sentences with an eventive predicate become well formed without aspect marking in Mandarin. They describe regularities rather than episodic events. Some bare activities allow habitual readings in the absence of any adverbs (5d).
(5) a. Mǎlì hěn-shǎo xiào.
Mali very-few smile
‘Mali rarely smiles’.
b. Tā zài zhèi-jiā miànbāofáng mǎi tiándiǎn
3SG at this-CL bakery buy dessert
‘He buys his dessert in this bakery.’
c. Zhèi-jǐ-jiā diàn hěn wǎn guānmén.
this-several-CL store very late close
‘These stores close late.’
d. Gǔlóng chōu-yān.
‘Gulong smokes.’/ *‘Gulong is smoking.’
The data above suggest that habitual readings of eventive predicates do not require aspectual marking. We claim that the habitual readings of eventive predicates in Mandarin are derived from a quantificational element in the form of an overt Q-adv. or a covert Q operator. Overt Q-adverbs, such as jīngcháng ‘often’ and hěn-shǎo ‘rarely’, take predicates of events and return habitual properties, which are properties of times (6). (Sun 2014)
(6) ⟦hěn-shǎo⟧g,c =λP. λQ<i,t>.λt: t is relatively long. Few Q-intervals in t contain a P-event
The covert quantification operator Q plays a similar role to that of overt Q-adv., and it also gives properties of times. The only difference between overt Q-adverbs and the covert Q is that Q is an underspecified quantification. That the operator Q is morphologically null in Mandarin comes as no surprise: habitual/generic sentences are minimally marked crosslinguistically, independently of their tense-aspect system (Dahl 1995).
Carlson, G. N., & Pelletier, F. J. (1995). The Generic Book. Chicago: University Of Chicago Press. Davidson, D. (1967). The logical form of action sentences. In N. Rescher (Ed.), The logic of decision and action (pp. 81-120). Pittsburgh: University of Pittsburgh Press. Katz, G. (1995). Stativity, Genericity and Temporal Reference. Ph.D. dissertation, Department of Linguistics. New York: University of Rochester. Katz, G. (2003). Event arguments, adverb selection and the Stative Adverb Gap. In E. Lang, C. Maienborn, & C. Fabricius-Hansen (Eds.), Modifying Adjuncts (pp. 455-474). Mouton de Gruyter. Kratzer, A. (1998). More Structural Analogies Between Pronouns and Tenses. In D. S. Lawson (Ed.), Procedings of SALT VIII (pp. 92-109). Ithaca: Cornell University. Sun, H. (2014). Temporal Construals of Bare Predicates in Mandarin Chinese. Ph.D Thesis, Université de Nantes, Leiden University. Utrecht: LOT.
※ 清華語言所新036次專題演講公告 ※
講者：Michael Yoshitaka ERLEWINE (National University of Singapore)
題目：Extraction and licensing in Toba Batak
I investigate patterns of preverbal fronting in Toba Batak, a predicate-initial Austronesian language of northern Sumatra. Contrary to the claims of previous work on the language, I show that multiple constituents can be simultaneously fronted, though only in limited configurations. I argue that the heads C and T are present in Toba Batak, with their traditional division of labor (wh/focus-probing by C and subject licensing and fronting by T), but extraction patterns are restricted by the limited means of nominal licensing (abstract Case) in the language. In addition, the features of C and T have the option of being bundled together on a single head, inheriting properties of both C and T and probing together for the joint satisfaction of their probes. Further evidence for this organization of C and T comes from the distribution of the particle *na* in two varieties of the language. This study sheds light on the relationship between western Austronesian voice system languages and the clause periphery in other language families.