Six reconstruction effects in Mandarin Chinese

Lip reading aims at decoding texts from the movement of a speaker's mouth. In recent years, lip reading methods have made great progress for English, at both word-level and sentence-level. Unlike English, however, Chinese Mandarin is a tone-based language and relies on pitches to distinguish lexical or grammatical meaning, which significantly increases the ambiguity for the lip reading task. In this paper, we propose a Cascade Sequence-to-Sequence Model for Chinese Mandarin (CSSMCM) lip reading, which explicitly models tones when predicting sentence. Tones are modeled based on visual information and syntactic structure, and are used to predict sentence along with visual information and syntactic structure. In order to evaluate CSSMCM, a dataset called CMLR (Chinese Mandarin Lip Reading) is collected and released, consisting of over 100,000 natural sentences from China Network Television website. When trained on CMLR dataset, the proposed CSSMCM surpasses the performance of state-of-the-art lip reading frameworks, which confirms the effectiveness of explicit modeling of tones for Chinese Mandarin lip reading.

Among many forms of A-not-A questions in Mandarin Chinese, the shi-not-shi question is found to be unique in that it obligatorily gives rise to a biased reading toward its prejacent, so-called positive bias. The previous pragmatic approach by Ye (2020) establishes a link between focus in polar questions and question bias to explain this biased reading. However, the current study finds that two other A-not-A questions formed by epistemic modals, hui-not-hui and keneng-not-keneng which are not focus markers, obligatorily produce positively biased readings as well. I propose that biased A-not-A questions are a type of high-negation questions with A-not-A residing outside of TP. shi, hui and keneng should all be analyzed as epistemic modals which are the overt realization of Goodhue’s (2019) epistemic operator scoped by the negator. The positively biased reading is derived from the resulting unbalanced partition based on general pragmatic principles. This analysis from the semantic aspect provides new evidence for the argument that the first A has reality only in PF. Furthermore, the Mandarin Chinese data lends evidence to Goodhue’s (2019) argument that there exists a doxastic speech operator between NegP and TP in high negation questions. The paper also provides explanations for previously remaining questions on bias cancelation by the stress marker daodi and factive predicates like zhidao (“know”).

Syntactic and semantic structure in the recall of orders of approximation to English

Perspective-shifting are helpful for children in Chinese passive sentence comprehension

According to the Unaccusative Hypothesis, intransitive verbs are divided into unaccusative and unergative ones based on the distinction of their syntactic properties, which has been proved by previous theoretical and empirical evidence. However, debate has been raised regarding whether intransitive verbs in Mandarin Chinese can be split into unaccusative and unergative ones syntactically. To analyze this theoretical controversy, the present study employed functional magnetic resonance imaging to compare the neural processing of deep unaccusative, unergative sentences, and passive sentences (derived structures undergoing a syntactic movement) in Mandarin Chinese. The results revealed no significant difference in the neural processing of deep unaccusative and unergative sentences, and the comparisons between passive sentences and the other sentence types revealed activation in the left superior temporal gyrus (LSTG) and the left middle frontal gyrus (LMFG). These findings indicate that the syntactic processing of unaccusative and unergative verbs in Mandarin Chinese is highly similar but different from that of passive verbs, which suggests that deep unaccusative and unergative sentences in Mandarin Chinese are both base-generated structures and that there is no syntactic distinction between unaccusative and unergative verbs in Mandarin Chinese.

The Prinmi copula and existentials are grammatically quite similar to auxiliary verbs. This chapter addresses complex syntactic structure built with the complement clause of auxiliary verbs. It focuses on meanings and functions of auxiliary verbs, covering a variety of auxiliary verbs in Prinmi: deontic modals, epistemic modals, the verbs-to-do, aspectual auxiliaries, and some minor types of auxiliary verb. The chapter discusses causative auxiliary under causative construction. Prima facie, the focus-presupposition construction remarkably resembles the obligational construction in structure: both require a nominalized clause serving as complement of the copula in an uninflected form. Finally, the chapter describes some minor auxiliary verbs in Prinmi. These are the purposives ʃɨH and ʒɨR, the desiderative ɹʒH ɹĩL, and the venturive waF. The meaning of the desiderative auxiliary is fairly similar to some modal clitics such as the optative; they differ in that the former expresses a desire, rather than an intention or volition.Keywords: auxiliary verbs; copula; existentials; Prima facie; Prinmi

In Mandarin Chinese, there exists such adverbial clause as “Li Zhen cuicuide zhale yipan huashengmi” (Li Zhen fried a dish of peanuts crispy), in which the adverbial modifies the predicate verb but semantically orients to the object. This kind adverbial clause can be formulated as “NPs+APo+De+VP+NPo=NP+VP+O and O is characterized by the adverbial”. The object-oriented adverbial clause is a mismatched syntax-semantics phenomenon, with the mapping between form and meaning distorted. Many previous studies have proposed not fully identical analyses for the syntactic distribution, pragmatic motivation and constraints. However, few researches have made syntactic and semantic analyses from the perspective of event structure in the framework of formal linguistics, which leaves wide space for further study.Event structure theory is adopted in this paper to make analysis of object-oriented adverbial clauses in event semantics perspective. This paper aims to examine the syntactic structure from the perspective of event semantic structure and explore how event structure is represented in syntactic structure of object-oriented adverbial clause.

Parts and boundaries

Entrainment is a natural inclination of people who interact with each other to develop similar, matching, or synchronized forms of behavior. In spoken interactions, this was observed at many linguistic (syntactic structures, referring expressions, dialectal features) and para-linguistic (intensity, speech rate) levels. Numerous studies have shown that the degree of spoken entrainment correlates with several social conversational attributes such as task success, likability, attractiveness, power distribution, and others. This feature of human-human interactions promises great potential for improving the naturalness and social acceptability of voice applications in human-machine interactions. However, despite many findings of various forms of entrainment in various languages and cultures, there is very little work on analyzing the effect of language or culture on speech entrainment in comparable spoken corpora. This is a significant gap since many applications utilizing spoken interaction and entrainment will be deployed to be used by native and non-native speakers of many cultures and languages. A welcome exception filling this gap is a recent paper by Xia et al. (2014) comparing entrainment in Standard American English and Mandarin Chinese. In this paper we provide a first analysis of speech entrainment in Slovak based on comparable corpus to the ones analyzed in Xia et al. This work contributes to our understanding and future modeling of cross-cultural aspects of behavioral entrainment.

The point of departure of this paper is the hypothesis that there is a general format common to all representations in the human cognitive system. There is evidence from cognitive psychology that this might be frames in the sense of Barsalou’s. The aim of the paper is an exploration of the consequences of this assumption for natural language. Does natural language provide evidence in favor of Barsalou frames being the general format of representations in human cognition? The paper discusses two levels of representation of linguistic gestures: syntactic structure and meaning. The first part deals with syntactic structure and compositional meaning. It is argued that specific universal uniqueness constraints on the syntactic and semantic structure of complex linguistic gestures provide positive evidence for the assumption. The second part investigates lexical semantics, in particular the emergence of abstract attribute vocabulary. Observations in this field, too, corroborate the hypothesis.

Current lexical semantic representations for natural language applications view verbs as simple predicates over their arguments. These structures are too coarse-grained to capture many important generalizations about verbal argument structure. In this paper, I specifically defend the following two claims: verbs have rich internal structure expressible in terms of finer-grained primitives of meaning, and at least for some languages, verbal meaning is compositionally derived from these primitive elements. I primarily present evidence from Mandarin Chinese, whose verbal system is very different from that of English. Many empirical facts about the typology of verbs in Mandarin cannot be captured by a flat lexical semantic representation. These theoretical results hold important practical consequences for natural language processing applications.

Data involving epistemic modals suggest that some classically valid argument forms, such as reductio, are invalid in natural language reasoning as they lead to modal collapses. We adduce further data showing that the classical argument forms governing the existential quantifier are similarly defective, as they lead to a de re–de dicto collapse. We observe a similar problem for disjunction. But if the classical argument forms for negation, disjunction and existential quantification are invalid, what are the correct forms that govern the use of these items? Our diagnosis is that epistemic modals interfere with hypothetical reasoning. We present a modal first-order logic and model theory that characterizes hypothetical reasoning with epistemic modals in a principled manner. One upshot is a sound and complete natural deduction system for reasoning with epistemic modals in first-order logic.

The multi-media archives are very difficult to be shown on the screen, and very difficult to retrieve and browse. It is therefore important to develop technologies to summarize the entire archives in the network content to help the user in browsing and retrieval. In a recent paper [1] we proposed a complete set of multi-layered technologies to handle at least some of the above issues: (1) Automatic Generation of Titles and Summaries for each of the spoken documents, such that the spoken documents become much more easier to browse, (2) Global Semantic Structuring of the entire spoken document archive, offering to the user a global picture of the semantic structure of the archive, and (3) Query-based Local Semantic Structuring for the subset of the spoken documents retrieved by the user’s query, providing the user the detailed semantic structure of the relevant spoken documents given the query he entered. The Probabilistic Latent Semantic Analysis (PLSA) is found to be helpful. This paper presents an initial prototype system for Chinese archives with the functions mentioned above, in which the broadcast news archive in Mandarin Chinese is taken as the example archive.KeywordsAutomatic GenerationNews StorySemantic StructureLatent TopicSpontaneous SpeechThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

This book contains chapters on epistemic modality by top researchers in the area. Several chapters about the question of what kind of possibilities epistemic possibilities ask: What kind of a possibility space do we need in order to model epistemic possibility? Is it a different kind of space from the one we need to model, for example, metaphysical possibility? Others are about the workings of epistemic modal expressions in natural language, for example, should we be contextualists about epistemic modals, or relativists, or should we go in for some sort of expressivist theory?

On the semantic content of subcategorization frames