Rethinking Wh-island Effects in Chinese
Barry C.-Y. Yang
TingTing Christina Hsu2022, Studies in Chinese Linguistics
Sign up for access to the world's latest research
checkGet notified about relevant papers
checkSave papers to use in your research
checkJoin the discussion with peers
checkTrack your impact
Abstract
The traditional observation that Chinese wh-arguments do not exhibit wh-island effects may be only apparent. With new evidence from ‘how-many’ phrases it is demonstrated that Chinese has wh-island effects even with wh-arguments. What nullifies such effects is in fact the disguise of D-linkedness. Although the lack of wh-island effects seems to pattern Chinese wh-construals with Japanese ones, further tests show that these two languages are still different with respect to strong island effects, (anti-)crossing effects, and multiple wh-interpretations. The finding leads to the need to reinvestigate the mechanisms underlying the scope-taking wh-elements of wh-in-situ languages on the one hand, and those triggering wh-island effects on the other.
Related papers
Multiple Wh-Construction and Its Interpretations in Chinese *
This paper investigates the syntax and semantics of multiple wh-questions (MWHs) in Chinese, which, being a clear case of wh-in situ language, qualifies as a useful tool to reexamine earlier proposals on the same issue in wh-movement languages. On the syntax part, based on Subjacency conditions and other blocking effects, we argue that even in Chinese, one of the wh-phrases in the MWHs must move at LF in order to trigger the pair-list reading (PL), a proposal that reflects a generalization in Bošković . On the semantic interface, inspired by a proposal in Beghelli (1997), we propose that the nature of the LF movement for the pair-list reading is focus-related. We show that through the focus semantics, we can straightforwardly derive the 'question-on-question effect' in Hagstrom (1998) and the 'conjoined speech acts' analysis in . Furthermore, since focus is by definition exhaustive, we also derive the exhaustivity of the pair-list reading without any ad hoc stipulations. As for single-pair readings (SP), the ability of Chinese wh-phrases to remain in situ provides a mechanism for them to be unselectively bound by the Q-morpheme, which in semantics is interpreted as a choice function, as in . The result is that a single-pair answer suffices as an answer to MWHs.
Theoretical implications of the parsing of Japanese wh-scrambling constructions
2003
Grammatical theories differ with respect to mechanisms used to relate displaced categories to the predicates from which they derive their argument roles. Transformational theories are multi-stratal, and relate displaced categories to their underlying canonical positions, from which argument interpretation proceeds, as in a declarative clause. By contrast, mono-stratal theories do “not assume the existence of an extraction site, so there is no sense in which the processing of this construction requires the formation of an underlying” representation.
On nominal islands and LF extraction in Chinese
Natural Language & Linguistic Theory, 1994
Evidence from Chinese Wh-questions strongly suggests that LF locality is a matter of referentiality and nominality, while locality in overt syntax involves the requirement of head government. This observation echoes the SVLIT ECP approach of WAHL (1987). Specifically, Chinese arguments and referential adjuncts (when, where, instrumental how and purpose why) contrast with nonreferential adjuncts (manner how and reason why) in allowing wide-scope construals out of islands in LF. Overt Wh-fronting, on the other hand, displays an argument/adjunct asymmetry. This paper proposes to deal with the LF asymmetry within the Generalized Binding framework, in association with the referentiatlnoureferential distinction among Wh-etements. With the ECP reduced to the OENERALlZED BINmN(~ VmNCIVLES (GBPs), it further argues for a type of locality employing the notion of CHECKPOINTS instead of BARRIERS, according to which the [N] feature is checked along with the [WH] feature through Comp-indexing. Consequently, nominal clauses are always islands for nonnominal/nonreferential adjuncts, because either selectional restrictions or the GBPs will be violated by longdistance extraction. This move is independently motivated by a parallel asymmetry in Chinese (non)bridge-verb constructions.
Contradiction, Why-Questions and Complex Islands in Chinese
University of Pennsylvania Working Papers in Linguistics, 2015
This paper proposes that the complex NP island effects (henceforth: CNPC) in Chinese whinterrogatives receive a semantic explanation. I argue that the mainstream treatments of Chinese CNPC in terms of movement constraints fail to address the interpretational distinctiveness associated with different wh-phrases. On the other hand, island facts follow naturally from an independently motivated constraint on relative clause's propositional content, motivated by this distinctiveness. The rest of this paper will be structured as follows: Section 2 reviews a standard assumption that motivates syntactic-based theory of CNPC in Chinese and other wh-in situ languages, namely there exists a fundamental argument-adjunct asymmetry; I show that positing such an asymmetry is not empirically sound and instead argues that the real underlying differences of island-inducing abilities in wh-in situ languages should be reduced to the different semantics between reason adverbial why on one hand and other wh-phrases on the other. Section 3 goes on to discuss the idiosyncratic semantics of reason adverbial why; Section 4 proposes a crucial semantic requirement in the interpretation of wh-phrases within relative clauses. I show that CNPC effects arise when the semantics of why is necessarily incompatible with this semantic requirement for embedded wh-phrases, thus always creating contradiction; Section 5 concludes the paper. 2 Rethinking the Argument-Adjunct Asymmetry 2.1 Previous Theories Chinese in situ wh-questions have been standardly claimed to obey the complex NP constraint (Huang 1982, Huang et al. 2009). This is supported by examples such as the following:Numbered examples can be formatted like the following sentences. Leave a blank line above and below a block of examples, but do not leave extra space between examples. 1 (1) *Ni xihuan [ta weishenme xie _] de shu? You like he why write REL book #'Whyi do you like the books that he wrote _i?' This infelicity, according to Huang, is reminiscent of the canonical island effects in English, illustrated as below. *I am grateful to the three anonymous reviewers for the PLC for their useful suggestions. I also thank Rui Chaves for discussions with me. I am particularly indebted to Jun Chen and Lihua Xu for their overall support and for supplying their judgments. The usual disclaimer applies. 1 These wh-phrases also obviate other strong island constraints, for example, adjunct islands and subject islands, illustrated as follows (i) a.Ta [yinweishenme jiegu yuangong] yihou bei laoban piping=le? He because.of what sack employees after by boss criticize=ASP 'For what reasoni was he criticized by the boss after he sacked employees _i?' b.[Ta yinweishenme cizhi] zui hao? He because.of what resign be.most good 'For what reasoni will that he resigned _i be the best?' These phenomena can follow from the analysis laid out in this paper. However, I will leave their exact formulation to future work.
OVERT Q-FEATURE MOVEMENT IN JAPANESE WH-CONSTRUCTIONS
2006
in terms of Richards's (2001) Principle of Minimal Compliance, which is basically the notion that once a particular constraint is satisfied, it may be subsequently ignored.
The nature of Chinese Wh-questions
Natural Language & Linguistic Theory, 1994
The scope properties of Chinese Wh-elements, which are in situ in the syntax, are usually accounted for by assuming that they are raised at LF as operators. Not only does such an account rely on costly stipulations but it also fails to generalize to Whelements which are not XPs. The proposal here is that Wh-eiements be treated as variables bound by a OUESTION (Q) operator which assigns scope to the Wh-elements in its domain. The Q operator can be attached to a verb and realized as the A-not-A form of the verb when there is no Wh-element in its domain. The Q operator, including the one realized as the A-not-A verb form, is raised at LF to satisfy selectional restrictions and its movement is regu!ated by the standard constraints. The scope properties of Wh-elements and the distribution of A-not-A verbs are accounted for within the current theoretical framework without further stipulation.
Additional Evidence for LF: Wh-Words In Mandarin Chinese
dspace.wul.waseda.ac.jp
Studies of wh-words in Mandarin have added to discoveries in the syntax literature. The discussion of movement of wh-words in Huang (1982) lent additional evidence to the syntactic level of Logical Form, and also contributed to the notion of barrier (Chomsky 1986). However, there have ...
Why Island Constraint Is Weaker in Japanese than in English: A Processing Perspective
Open Journal of Modern Linguistics
The constraints on discontinuous dependency in a sentence have been discussed as "island constraint". Island phenomenon can be observed in many languages; however, while the island effect is quite noticeable in English, it seems obscure in Japanese. This study thus experimentally evaluates the strength of island effect in Japanese complex sentences quantitatively and demonstrates that the island effect in Japanese is weaker than that in English. Furthermore, we attempt to attribute the difference in the strength of the island effect in these two languages to their processing characteristics; that is, the syntactic relationship between two discontinuous elements in English must be computed at the head of an island, with one of the two elements unreceived, while the syntactic computation at the head of an island in Japanese is performed with both elements received. We assume that the processing of discontinuous dependency is thus more costly in English than in Japanese, and that the strong island effect in English reflects the relative difficulty of the syntactic computation. We obtained evidence for our hypothesis from an experiment asking participants to make grammatical judgments on Japanese sentences in which we manipulated their word-order in three ways. We suggest that part of the apparent syntactic phenomenon can be attributed to processing factors.
Wh -Expressions in Mandarin Chinese
The Handbook of Chinese Linguistics, 2014
The Intervention Effect as a Syntactic Phenomenon in Korean Wh-questions
Studies in generative grammar, 2016
‘Multiple Sluicing’ in Japanese and the Functional Nature of wh-Phrases
1998
This article discusses the syntax and LF properties of 'Multiple Sluicing' in Japanese. Built on the LF-Copying analysis, we examine the condition imposed on the elements figuring in the IP-Copying site, and show the relevance of the functional interpretation of wh phrases. We also look at some quantificational properties of multiple wh questions. Finally, the possibility of 'Multiple Sluicing' in English is examined. 'John said someone bought something.' b. Mary-wa [dare-ga nani-o ka] siri-tagat-te iru. Mary-Top [who-Nom what-Acc Q know-want is 'lit. Mary wants to know who what.' Takahashi argues that in sentences like (1), two wh phrases are present, one adjoined to the other, in Spec CP, in which they get licensed by the C head which dominates the interrogative marker ka, with IP being empty (due to deletion, in Takahashi's analysis).
English Speakers ’ L 2 Chinese Wh-topicalization : Movement or Base-G eneration ?
2011
In contrast to Chinese, a wh-word in English must obligatorily move to Spec-ForceP in whquestions and no discourse requirement is imposed on wh-movement. Furthermore, topicalization does not apply to wh-words in English (Stepanov and Tsai 2008). Availability of wh-topicalization in Chinese raises a question as to whether a wh-topic like shenme shu ‘what book’ in (1b) is the result of movement or base-generation. Taking complex NP island, reconstruction and superiority effects as a diagnostic of movement, the current study aims to provide empirical evidence that the topicalized wh-element is derived by movement in English learners’ L2 Chinese grammars. The sensitivity of Chinese wh-topicalization to syntactic constraints has not been investigated in L2 acquisition research, presumably due to the general assumption that wh-words in Chinese stay in-situ and no overt movement is involved; thus no constraints on movement can be tested in L2 Chinese. This study will, therefore, fill a gap...
The Syntax and Semantics of Korean Wh-Quantification: Wh-NPIs and wh-in-situ
2020
This dissertation investigates the syntactic and semantic aspects of long-distance dependency involved in Korean wh-Negative Polarity Items (NPIs). Korean wh-NPIs consist of a whitem and a focus particle -to meaning ‘also’ or ‘even’. In all reported cases, the wh-item and -to appear adjacent to each other. However, the wh-item and -to can appear discontinuously when the NPI licenser is located in a different clause from the wh-NPI. I will call this new form of wh-NPIs “split wh-NPIs”. The central claims of this dissertation are the following. The first, long-distance dependency displayed in split wh-NPIs is attained via overt movement of a focus particle -to ‘also/even’ in an effort to form a local relation with an NPI licenser. Second, -to is interpreted with its associated wh-item via point-wise semantic composition without forming a local relation at LF. These conclusions are motivated by finding two sets of new data: 1) split wh-NPIs cannot be formed across island boundaries, 2) split wh-NPIs are susceptible to focus intervention effects. This analysis has important implications for another wh-in-situ construction in Korean, wh-questions. By comparing split wh-NPIs with wh-in-situ questions, I argue that wh-in-situ questions do not involve any type of movement.
Linguistic Form and Function: Expression of Countability in Chinese wh-Phrases
Language and Semiotic Studies
The present paper investigates the relationship between form (morpho-syntax) and function (semantic interpretation) as involved in the expression of countability in the Chinese whpronoun system. We focus on three wh-pronouns: duo-shao 'much-little', duo-shao-ge 'much-little-Classifier', and ji-ge 'how many-Classifier'. Based on the examination of the interpretation and distribution of these wh-phrases in three distinct linguistic contexts (questions, bare conditionals, and negative statements), we generalize that while duo-shao functions as an indeterminate determiner, duo-shao-ge and ji-ge function as two count determiners. This generalization invites the conclusion that the morpho-syntax of these whphrases, particularly, the presence/absence of a classifier, determines and shapes their semantic interpretation in countability. From a more general perspective, the present study suggests that a linguistic form determines its own linguistic functions.
Wh-Conditionals in Vietnamese and Chinese: Against Unselective Binding
Annual Meeting of the Berkeley Linguistics Society, 2006
The Relative Constructions and the Intervention Effect in Chinese
2008
Two types of analysis of relative constructions have been proposed in the literature: the matching (or wh-operator movement) and the promotion (or head-raising) analysis. The matching analysis has been proposed by Lees (1960, 1961), Chomsky (1965, 77), and Kuroda (1968) among others. 1 In this analysis, nominal expressions, to which relative clauses are adjoined, are base-generated in the head positions of NP, and wh-phrases or invisible operators move to the spec of the relative clauses to match themselves with the nominal expressions. Thus, the nominal expressions in the head positions do not move for the sake of deriving relative clauses. According to the promotion analysis, on the other hand, which has been suggested by Brame (1968), Schachter (1973), Vergnaud (1974) among others, nominal expressions are base-generated inside relative clauses and then, they are moved to the head positions of NP. Thus, it is not a wh-phrase but a nominal expression that moves into the head position in order to derive a relative clause. In other words; nominal expressions not only head relative clauses but also serve as (wh-)operators. Following Carlson (1977) and Afarli (1994), Aoun and Li (2003) show that both the matching and the promotion method are necessary to account for all types of relative constructions in English. Specifically, Aoun and Li claim that although wh-relative constructions (i.e. relative clauses in which wh-phrases are used) can resort to only the matching method, that-relative constructions (i.e. relative clauses in which "that" is employed) can choose either of the two methods. They have come to this conclusion on the basis of three types of evidence: (i) idiom chunks, (ii) binding, and (iii) scope interaction. Aoun and Li (2003) extend their analysis to Chinese and argue that although Chinese also allows the promotion as well as the matching-69
Information Structure Constraints and Complex NP Islands in Chinese
2013
This paper presents an analysis of the complex NP island effects in Chinese. I follow Ginzburg & Sag (2000)'s analysis of in situ wh-interrogative construction and propose that feature percolation from the non-head clause daughter to the head daughter is required for a proper treatment of in situ whrelative. A semantic analysis of the idiosyncrasy of weishenme 'why' reveals that a definite reading is forced for a wh-relative when weishenme stays in situ. This requirement causes feature percolation into relative head to fail. In this way I show that the island effects in Chinese can be independently ruled out in the grammar as a case of contradiction.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.