Workshop description
The Department of Linguistics of Ohio State University and the Language and Communication Science Laboratory of the University of Tsukuba are pleased to announce a joint workshop, comprising faculty and graduate students from both institutions, to be held over February 12-13, 2019. The goal of the workshop is to share research findings and have wide-ranging discussions over various aspects of linguistics. Linguistics in Tsukuba and OSU share a view of theoretical innovation and empirical accountability as mutually indispensible. This workshop will showcase work representing the respective specializations at each institute, endeavor to relate them to each other, and exchange information and propose ideas to create innovative research across disciplines and interdisciplinary fields.
Organizers
Robert Levine (OSU, levine.1 |AT| osu.edu)
Masaharu Shimada (Tsukuba, shimada.masaharu.fu |AT| u.tsukuba.ac.jp)
Yusuke Kubota (Tsukuba, kubota.yusuke.fn |AT| u.tsukuba.ac.jp)
Date and location
Date: 2019/2/12-13
Location: 103 Oxley Hall, Ohio State University
Program
February 12
9:00-9:20 Opening remarks
9:20-10:00 Yuichi Ono (Tsukuba) "Big Data, Little Data, Right Data: Two Case Studies and Their Implications for Data-Based Linguistic Research"
10:00-10:40 Huan Sun (OSU) “Intelligent Collaboration among Humans and Machines (from Natural Language Understanding Perspective)”
10:40-11:00 Coffee Break
11:00-11:40 Ryohei Naya (Tsukuba) "A Qualitative Study on the Productivity of -ment in Present-Day English"
11:40-12:30 Lightning Talks for Posters
12:30-14:20 Lunch Break
14:20-15:20 Poster Session (Venue: Main floor corridor)
Masatoshi Honda (Tsukuba) "Focus Fronting and Emphasis: A Case Study of Particle Fronting in English"
Hiroko Wakamatsu (Tsukuba) "Borrowing of the At Sign into Japanese"
Takashi Ishida (Tsukuba) "One Head or No Head?: Two Types of Copulative Adjectival Compounds"
Shohei Nagata (Tsukuba) "Does a VP front in VP-fronting in English and Why?"
Kanato Ochiai (Tsukuba) "Connective expressions in Japanese Computer-Mediated Communication"
Kazuyoshi Ishikawa (Tsukuba) "Another Interpretation of Instrument Subject Constructions in English"
Marie-Catherine de Marneffe (OSU) "Factivity in doubt: Clause-embedding predicates in naturally occurring discourse" (with Mandy Simons and Judith Tonhauser)
Daniel Puthawala (OSU) "Investigating Ellipsis and Parsing through RSVP"
15:20-16:00 Etsuyo Yuasa (OSU) "Koto imperatives in Japanese"
16:00-16:40 Mineharu Nakayama (OSU) "Japanese EFL learners' Acquisition of Seem Constructions"
17:00 Welcoming Speeches
17:30 Reception (Oxley 100) -- Open to all; no reservation needed
February 13
9:00-9:40 Rebecca Morley (OSU) "Phonological Contrast as an evolving function of local predictability"
9:40-10:20 Brian Joseph (OSU) “Finite Complementation in Greek – A Long-Standing Puzzle”
10:20-10:40 Coffee Break
10:40-11:20 Koyo Akuzawa (Seigakuin) and Yusuke Kubota (Tsukuba) "A semantic analysis of finite control in Japanese"
11:20-12:00 Ashwini Deo (OSU) "Identifying the strongest true alternative: Marathi =c and its counterparts"
12:00-12:50 Lightning Talks for Posters
12:50-14:30 Lunch Break
14:30-15:30 Poster Session (Venue: Main floor corridor)
Björn Köhnlein and Yuhong Zhu (OSU) "Contrastive foot structure, pitch accent, and stem allomorphy in Uspanteko"
Yuhong Zhu (OSU) "Are all lexical-specific changes "lexical diffusion"? Diachronic change of Suzhou Chinese coda /ŋ/ in register-sensitive Differing Readings"
Jordan Needle (OSU) "Towards an Embedding of HTLCG into LCGϕ"
Ellen Dossey, Zack Jones, and Cynthia G. Clopper (OSU) "Interactions of social, lexical, and contextual factors on lexical processing"
Michael White (OSU) "Evaluation Order Effects in Dynamic Continuized CCG: From Negative Polarity Items to Balanced Punctuation"
Zhiguo Xie (OSU) "A Degrees-as-Kinds Analysis of Differential Verbal Comparatives in Chinese and Its Theoretical Implications"
Kosuke Kuroda (Tsukuba) "On the Optionality of Coherence Relation Marking: A Comparison between English and Japanese"
Satomi Takemoto (Tsukuba) "Causative constructions of Unaccusative Verbs in Japanese"
Haruka Shimura (Tsukuba) "The Emotional Implicature in the Present Perfect Progressive"
15:30-16:10 Misato Ido (NINJAL) "The meanings of dake and shika based on their maximality and polarity"
16:10-16:50 Nobuyoshi Miyoshi (Tsukuba) "Two Types of Intentional Predicates from the Viewpoint of Relative Clause in Japanese"
16:50-17:30 Marjorie K. M. Chan (OSU) "Dr. Sun Yat-Sen’s 1924 Cantonese Audiorecording: The Interplay of First and Second Dialect Acquisition"
17:30-17:40 Closing
Abstracts
Talks (2/12):
Yuichi Ono "Big Data, Little Data, Right Data: Two Case Studies and Their Implications for Data-Based Linguistic Research"
My talk discusses issues of construct validity in dealing with “Big Data” in linguistic research. Recent linguistic research, especially dealing with corpora and quantitative data, has come to be dependent on the outputs after some analytic operations from raw data or frequencies of several indices. As has been frequently shown in the demonstrative studies, especially in second language acquisition (SLA) and corpus-based sociolinguistic studies, the results are sometimes parser-dependent, over-jumping or less validate. In this talk, through the discussion of two of my pilot research, I start with some conceptual issues of “Big Data trends” in learning analytics in linguistic research. Then, I discuss the necessity of positing operationalized constructs dealing with “Big Data” and the significance of collecting “Right Data” rather than “More Data” (Borgman, 2015). Case Study 1 is concerned with more than 20,000 log data collected on the pronunciation-training courseware to predict accuracy. Case Study 2 discusses the parser-preference on the basis of more than 200 essays written by Japanese “English as a Foreign Language (EFL)” learners. Implications for future SLA or interlanguage research will be suggested in conclusion.
Huan Sun “Intelligent Collaboration among Humans and Machines (from Natural Language Understanding Perspective)”
In this talk, I will discuss our recent work on human-machine and machine-machine collaboration frameworks from the perspective of natural language question answering and command understanding. In particular, I will focus more on human-machine interactive semantic parsing (in AAAI’19), and machine-machine collaboration for information search, particularly for code snippets search (in WWW’19). Finally, I will conclude the talk with ongoing/future work.
Ryohei Naya "A Qualitative Study on the Productivity of -ment in Present-Day English"
The notion productivity in morphology consists of the quantitative and qualitative notions: profitability and availability. While previous studies on the productivity of the suffix -ment tend to focus on the quantitative side, they do not pay much attention to its qualitative side. The suffix is described as unproductive (i.e., unprofitable) in present-day English, but it remains unclear when such an unproductive suffix is available. Seeking answers to this question, this study examines the nature of -ment-nouns coined after 1900 based on data from the Oxford English Dictionary. The main findings are the following: (i) -ment can attach to nouns and adjectives as well as verbs (e.g., memberment, foolishment), and (ii) -ment can derive nouns from converted (or zero-derived) verbs (e.g., centrement, motherment). The former seems to be strange as long as we assume that -ment is a derivational suffix that forms deverbal nouns. The latter may also be problematic given that it appears to go against Myers’ Generalization: “no derivational suffix may be added to a zero-derived word” (Myers (1984: 66), see also Nagano (2008: 16–18)). This study will show that (i) and (ii) are neither strange nor problematic by arguing that -ment does not behave as a derivational suffix but rather like a lexical noun in present-day English (Naya (2017: Chapter 5), cf. Emonds (2000)). This study will also contribute to the recent discussion over the status of “derivational affixes” (e.g., De Belder (2011), Lowenstamm (2015), Creemers, Don and Fenger (2018)).
Etsuyo Yuasa "Koto imperatives in Japanese"
Constructionalization is “the creation of a formnew-meaningnew pair.” It is distinguished from constructional changes where formal changes alone or meaning changes alone take place (Traugott and Trousdale 2013). Subscribing to cognitive construction grammar (Croft 2001; Goldberg 2006), Traugott and Trousdale explain that constructionalization is a process where a population of speakers applies a new analysis (“neoanalysis”) to a construction in a network of constructions to create a new form-meaning pair. In this paper, I will show that Japanese koto imperatives (e.g., Kichinto kōdō-suru koto! “Act properly!”) are another instance of constructionalization. The goals of this paper are two-fold. First, I will examine semantic and syntactic properties of koto imperatives. Semantically, koto imperatives exhibit well-known characteristics of imperatives (Nitta, 1991). However, koto imperatives are different from the imperatives with explicit verbs of order from which koto imperatives are considered to have developed. Syntactically, I will argue that koto imperatives are similar to prototypical modal (raising-to-subject) sentences. Second, I will provide a constructional analysis and a coherent explanation for the properties of koto imperatives discussed in the first part of this paper. My claim is that koto imperatives not only went through the process of pragmatic strengthening to gain a new sense of order, but they were also sanctioned (Langacker 1987; Traugott and Trousdale 2013) by the schema of the modal construction and inherited its syntactic information. Therefore, koto imperatives are an instance of constructionalization where a new meaning of imperative is associated with a syntactic structure of the modal construction.
Mineharu Nakayama "Japanese EFL learners' Acquisition of Seem Constructions"
This talk presents a brief summary of our recent studies on Japanese EFL learners’ acquisition of seem constructions. The seem construction has two types: one with and one without the subject-to-subject (raising) movement. The raised construction is more difficult for Japanese EFL (JEFL) learners to comprehend than its unraised counterpart, and when an experiencer phrase is included, it is more difficult than its counterpart without the experiencer phrase. Moreover, when the experiencer phrase intervenes with the raised subject and its corresponding null subject, the construction is more difficult than those with the preposed experiencer phrase, irrespective of the nature of the experiencer, either lexical or pronominal. The results of two additional experimental studies comparing the construction with passives and the construction with and without a personal pronoun in the embedded infinitive clause further suggest that JEFL learners’ acquisition of the construction seems to be affected by Relativized Minimality and negative L1 transfer.
Talks (2/13):
Rebecca Morley "Phonological Contrast as an evolving function of local predictability"
In this talk I propose a type of phoneme classification function that is highly dependent on local context. This function derives from several well-known properties of speech processing: the existence of strong top-down effects on word recognition, the ability of phonetic cues to “trade off” with one another in the determination of contrast, and the phenomenon whereby listeners hallucinate phonemes from a signal that is purely noise. This proposal will be illustrated through a re-analysis of the well-known, and well-researched, phenomenon of vowel lengthening in American English. I will argue that the observed differences in vowel duration are the result of the interaction of speaking rate with an underlying representation of voiced obstruents as inherently short-duration segments. I will also argue that the categorical perception effects found by varying vowel duration do not provide evidence of phonological contrast, but can be attributed to general properties of the proposed phoneme classification function. The implications for theories of contrast, as well as models of sound change, will be discussed.
Brian Joseph “Finite Complementation in Greek – A Long-Standing Puzzle”
A key morphosyntactic development between Ancient Greek and Modern Greek was the loss of the very robust and widely used infinitive of earlier Greek and its replacement by finite (tensed person-marked) verbs in all contexts. This includes control contexts and all sorts of complementation. That has given rise to ostensible control constructions with finite complements which further, due to pro-drop, have person-marked verbs but no overt subject (e.g. Ancient Greek “He wants to-go(INF)” è later Greek “He wants that-Ø-goes”). There is considerable discussion in the literature as to how to analyze such structures (e.g., within Government and Binding Theory, the debate is whether the absence of the overt subject means a PRO or pro empty subject. Some analysts, e.g. Miller 2002, have gone so far as to equate the finite verbal complements of Modern Greek with Portuguese personal infinitives, an analysis that fails on synchronic and diachronic grounds. In this paper, I lay out the facts concerning these finite complement structures in Modern Greek, and attempt my own solution at this long-standing puzzle.
Koyo Akuzawa and Yusuke Kubota "A semantic analysis of finite control in Japanese"
‘Finite control’ in language like Japanese (Fujii 2006; Hasegawa 1984/85; Uchibori 2000) is problematic since the distribution of PRO is standardly taken to be limited to infinitival clauses. Responding to this challenge, Fujii (2006) proposed that the embedded tense that cannot freely alternate with non-past or past tense is in effect defective, and thus exceptionally licenses obligatory control in the embedded clause. However, obligatory control verbs such as kookai-suru (regret) which allows for tense alternation is problematic for this analysis. In this talk, we will propose a semantic alternative to finite control in koto clauses. Specifically, we will claim that the empty subject is just a zero pronoun (pro), and that obligatory control status in koto clauses is determined solely on the basis of the meaning of the higher verb. Based on the two classic semantic notions of the control, de se attitude (Chierchia 1989) and responsibility relation (Farkas 1988), we will show that the control verbs share a certain type of presupposition: a causal relation between de se proposition P and volitional action V, which is the necessary and sufficient condition for inducing control in koto clauses.
Ashwini Deo "Identifying the strongest true alternative: Marathi =c and its counterparts"
Several Indo-Aryan languages contain a discourse clitic whose uses overlap with those of English discourse adverbials like only, just, right, and even without corresponding perfectly to either of them. This paper makes an empirical and theoretical contribution to understanding this family of particles, focusing on the clitic =(a)c in Marathi.
The diverse effects of =c can be unified through an analysis that extends the analytical tools from Beaver & Clark (2008), Coppock & Beaver (2011), and Velleman et al (2013), which analyzes the function of focus-sensitive discourse particles in terms of the way they comment on the strength relation between the prejacent p and alternatives p′ in the Current Question (CQ) in a given context. I propose that discourse particles may additionally comment on the strength relation between the interpretation index at a given context and alternative admissible interpretations at that context (INT). I show how the varied discourse effects of =c come from the interaction of this lexical meaning with distinct contextual conditions which are the source of (a) a relevant scale along which alternatives in the CQ or INT can be ranked and (b) expectations wrt the strength of the strongest alternative.
Misato Ido "The meanings of dake and shika based on their maximality and polarity"
In this study, I looked at the meaning of two Japanese focus particles dake (‘only’) and shika (an NPI counterpart of dake). Kuno (1999) claimed that dake-sentences convey a positive proposition as the main assertion and a negative proposition as the secondary assertion, while shika-sentences convey the negative proposition as the main assertion and the positive proposition as the secondary assertion. However, the precise nature of the notions of main and secondary assertions remains unclear. Yoshimura (2007) pointed out that “secondary assertion” should be analyzed as entailment rather than presupposition in formal semantics. Nevertheless, Yoshimura herself does not clarify the nature of the two uncancellable meaning components of dake and shika any more than Kuno does. Building on Tomioka (2015), I argue that the secondary assertions of dake and shika are entailments derived as logical consequences from the main meanings of these expressions. Tomioka pointed out that dake expresses maximality rather than exhaustivity. I argue that a parallel analysis is possible for shika that straightforwardly. Furthermore, I will show that the notion of maximality convey the secondary meanings of dake and shika as a logical consequences inevitably.
Nobuyoshi Miyoshi "Two Types of Intentional Predicates from the Viewpoint of Relative Clause in Japanese"
This presentation, from the viewpoint of relative clauses in Japanese, clarifies environments that have intensional readings and argues that there are two types of intentional predicates. Previous studies of intensional predicates have considered that predicates in intensional or nonspecific contexts have the modal operator. In reality, however, there are phenomena to which this generalization cannot apply. In this study, we classify predicates into a type to refer to a mental world and a type to refer to the real world. In addition, through an observation of restrictive relative clauses in Japanese, we further subdivide the former type of predicates into a type to only refer to possible worlds and a type to refer to both possible worlds and time domains. The analysis mentioned above can explain not only the counterexamples of intensional readings, but also behavior of exceptional restrictive relative clauses in Japanese.
Marjorie Chan "Dr. Sun Yat-Sen’s 1924 Cantonese Audiorecording: The Interplay of First and Second Dialect Acquisition"
In 1924, a year before his death, a Cantonese audiorecording was made by the China Evening Post in Guangzhou of Dr. Sun Yat-sen 孫逸仙 (Sun Zhongshan 孫中山; Nakayama Shō 中山樵), known as the father of modern China. Although only 7 minutes in length, the recording—in which Dr. Sun advocated a policy for national salvation—provides an interesting corpus to study the interplay of his first and second dialect acquisition, his first being Zhongshan Cantonese from early childhood, and his second that of standard Cantonese acquired later in Hong Kong and Guangzhou. An instance of self-correction reveals his conscious choice (albeit with mixed success) of his second dialect for greater appeal to a broader, regional audience. Three basic patterns can be found in his speech production: 1) adoption of some standard Cantonese segments and tones, 2) retention of some Zhongshan segments and tones, and 3) variation and fluctuation between the two varieties of Cantonese. These three patterns will be analyzed in this study.
Posters (2/12):
Masatoshi Honda "Focus Fronting and Emphasis: A Case Study of Particle Fronting in English"
Within the cartographic framework, Rizzi (1997) hypothesizes that focus fronting targets a unique functional projection dedicated to focus in the CP domain. In contrast, observing non-contrastive focus fronting phenomena in Romance languages (e.g., Sicilian), Cruschina (2011) proposes his original split CP hypothesis, according to which the CP domain includes two functional projections for focus: Contrastive Focus [CFoc] and emphatic Information Focus [IFoc] (i.e., new information and emphasis). The assumption of the latter focus projection provides new theoretical tools for scholars to examine focus-fronting phenomena cross-linguistically (e.g., Trotzke and Quaglia (2016) for particle fronting in German). Extending the emphatic IFoc approach to particle fronting in English (e.g. Round and round spins the fateful wheel! (Emonds (1976: 29)), this study demonstrates that particle fronting in English is derived by emphatic IFoc fronting. The proposed analysis is further extended to comparative substitution (e.g. Most embarrassing of all was losing my keys. (Emonds (1976: 35))).
Hiroko Wakamatsu "Borrowing of the At Sign into Japanese"
In the field of language contact, Muysken (2000, 2013) identifies distinct and fundamental patterns of code-mixing, i.e., the cases where lexical items and grammatical features from two language appear in one sentence. One of them is Insertion, in where "well defined chunks of language B into a sentence that otherwise belong to language A (2013:712).” Another pattern is Alternation, in which “[t]he succession of fragments in language A and B in a sentence, which is overall not identifiable as belonging to either A, or B (2013:713).” Using these notions, I analyze the recipe names containing the at sign, posted to a popular website in Japan. I argue that the at sign is at times used as in English, relating to Insertion; and also used as a marker of direct modification, or adjunction, relating to Alternation.
Takashi Ishida "One Head or No Head?: Two Types of Copulative Adjectival Compounds"
This presentation focuses on adjectival compounds: compounds with an adjectival head. This type of compound has been largely neglected by scholars in comparison to the group of nominal compounds (cf. Conti (2010)). The current study, inter alia, is interested in copulative adjectival compounds (e.g., bitter-sweet emotion, an Afro-American girl). Based on the discussion of N-N coordinate compounds (cf. dvandva (cf. Bauer (2008), Shimada (2013, 2016))), I will point out the syntactic and semantic differences observed in copulative adjectival compounds from the view of headedness. I will further argue that the relevant compounds can be classified into two types.
Shohei Nagata "Does a VP front in VP-fronting in English and Why?"
I address so-called VP-fronting (hereafter VPF) in English (e.g., John promised to pass the exam, and pass the exam he did ) . In the generative literature, VPF has traditionally been captured by the movement of a verb phrase to a certain sentential position (Lechner (2003), Phillips (2003), Huang (1993) among others). On the contrary, some authors have recently argued that a fronted VP in English should be analyzed as a dislocated constituent rather than a moved one (e.g., Ott (2018), Thoms and Walkden (2019)), from the observations that VPF exhibits (1) morphological mismatches, (2) the interpretation of pronouns which is not observed in general moved constituents, (3) parallelism with a fragment answer of a predicate (e.g., A: What did John do? B: Win five medals. ),and so on. However, the parallelism does not seem to straightforwardly lead us to a conclusion that VPF is an outcome of bi-clauses in that the pragmatic effect imposed on a fronted VP is distinct from that on a fragment answer (i.e., given vs. new)). With in mind such observations arguing against movement approaches, this presentation nevertheless explores the possibility for an account employing movement and aims to provide a novel view of do-support.
Kanato Ochiai "Connective expressions in Japanese Computer-Mediated Communication"
Nowadays, the communication media "LINE" has become widespread among people of a wide generation in Japan and it is difficult to avoid focusing on LINE for grasping recent spoken Japanese. This presentation takes up the difference of the usage tendency of connective expressions (Conjunction / Conjunctive particles) between LINE conversations (LINE), "Feature phone e-mail comversations"(Fp e-mail), and "personal conversations". As a result of the investigation, I found out the next 3 points: 1)The usage frequency of all conjunctions is widely different between LINE and Fp e-mail but the usage order of each conjunction is similar. 2)The usage tendency of the conjunctive particles used in the middle of the sentence is similar in LINE and Fp e-mail. 3)The usage tendency of the conjunctive particles used in the end of the sentence is widely different between LINE and Fp e-mail. Also I will discuss the usage tendency above from the viewpoint of "kedo" used in personal conversations, and Topic juxtaposing occurring in LINE.
Kazuyoshi Ishikawa "Another Interpretation of Instrument Subject Constructions in English"
Instrument subjects in English have been mainly analyzed based on a semantic role: agent (e.g. The stick hit the horse. (Schlesinger (1989: 196))). Instrument subject constructions are currently required to be considered from a pragmatic perspective because they are marked expressions (cf. Alexiadou and Schäfer (2006), Mack (2010), Fellbaum and Rapoport (2013)). In previous studies, instrument subject constructions are used to indicate what kind of instruments they are in comparison with other instruments. In this presentation, I will point out that there is another interpretation, in which case instrument subjects are contrasted with human agents in context. This study contributes to reveal a new aspect of instrument subject constructions.
Marie-Catherine de Marneffe (with Mandy Simons and Judith Tonhauser) "Factivity in doubt: Clause-embedding predicates in naturally occurring discourse"
Clause-embedding predicates, like “know" and “believe", are a central case in the study of projection. Several factors have been proposed to explain the projection behavior of clause-embedding predicates, but often on constructed examples. This paper describes a resource to study which linguistic and extra-linguistic factors are involved in understanding the extent to which utterance content projects. We built the CommitmentBank, a corpus of naturally occurring clausal-embedding predicates, with judgments of speaker commitment to the content of their complements. Using the CommitmentBank, we evaluate the contributions of several factors of projection put forth in the literature. While these data support the claim that predicates give rise to projective readings to different degrees, they challenge the existence of a categorical distinction between ‘factive’ and ‘non-factive’ predicates. Our findings instead support analyses of projection that do not assume conventional specification of presupposition (e.g., Heim 1983, van der Sandt 1992) but instead derive projectivity from a range of lexical and contextual factors (e.g., Simons et al. 2010, Abrusán 2011, Beaver et al. 2017).
Daniel Puthawala "Investigating Ellipsis and Parsing through RSVP"
Posters (2/13):
Björn Köhnlein and Yuhong Zhu "Contrastive foot structure, pitch accent, and stem allomorphy in Uspanteko"
We argue that a complex pattern of stem allomorphy in Uspanteko (Mayan) can be successfully analyzed within a morpheme-based model of morphology given two assumptions: i. underlying representations can contain metrical templates (e.g. Saba Kirchner 2013, Iosad 2016 for recent proposals); ii. pitch-accent contrasts in Uspanteko are a surface exponent of a difference between trochaic (falling tone) and iambic feet (level tone), as proposed in Authors (to appear). We claim that our analysis is more restrictive than an earlier account by Bennett & Henderson (2013), who (arbitrarily) divide relevant items into several nominal cophonologies. In analyzing non-concatenative exponence as an epiphenomenon of metrical affixation, our approach is in line with principles of Generalized Non-Linear Affixation (e.g. Bermúdez-Otero 2012, Trommer & Zimmermann 2014).
Yuhong Zhu "Are all lexical-specific changes "lexical diffusion"? Diachronic change of Suzhou Chinese coda /ŋ/ in register-sensitive Differing Readings"
This study aims to tease apart the interaction between two kinds of language change by focusing on the diachronic variation of coda /ŋ/ in the Suzhou dialect of Wu Chinese. I look at the interaction between standard dialect influence (from Mandarin Chinese) and regular Neogrammarian change within the speech community under the context of Differing Literary and Colloquial Readings, a register-based pronunciation variation. There are two main directions where /ŋ/ would change: it is either fronted to alveolar [n] due to contact with Mandarin in certain Literary forms, or it undergoes gradual neutralization and transfers the nasality to preceding vowels.
Jordan Needle "Towards an Embedding of HTLCG into LCGϕ"
Linear Categorial Grammar (LCG; Martin 2013, Worth 2016, Yasavul 2017) is a grammatical framework inspired by Oehrle (1994) and makes a sharp distinction between Curry (1961)’s notions of the phenogrammar and the tectogrammar. Making this distinction allows for an elegant analysis of direction-insensitive phenomena such as quantification, but is ill-equipped to handle direction-sensitive phenomena such as coordination. The goal of the present work is to remedy this situation by elaborating on the work done in Worth (2016), whereby the phenogrammar is augmented by subtyping on “phenominators"- functions that determine the “shape” of particular phenogrammatic terms. To show that this subtyping allows for better analyses of direction-sensitive syntactic phenomena, it is demonstrated that Hybrid Type-Logical Categorial Grammar (HTLCG; Kubota and Levine 2012) can be successfully embedded into the augmented LCG, dubbed LCGϕ (Linear Categorial Grammar augmented with phenominators).
Ellen Dossey, Zack Jones, and Cynthia G. Clopper "Interactions of social, lexical, and contextual factors on lexical processing"
Numerous factors, such as speaking style, regional dialect, lexical frequency, phonological neighborhood density, and semantic predictability contribute to variation in the speech signal. The goal of the current study was to examine the relative contributions of and interactions among these factors on the speed and accuracy of lexical processing. Cross-modal matching and intelligibility in noise tasks were conducted with visitors to a local science museum. Results revealed strong effects of regional dialect and semantic context on performance in both tasks, as well as complex interactions with the other factors. Differences across experiments reflect task-specific processing demands.
Michael White "Evaluation Order Effects in Dynamic Continuized CCG: From Negative Polarity Items to Balanced Punctuation"
Combinatory Categorial Grammar's (CCG; Steedman, 2000) flexible treatment of word order and constituency enable it to employ a compact lexicon, an important factor in its successful application to a range of NLP problems. However, its word order flexibility can be problematic for linguistic phenomena where linear order plays a key role. In this paper, we show that the enhanced control over evaluation order afforded by Continuized CCG (Barker & Shan, 2014) makes it possible to not only implement an improved analysis of negative polarity items in Dynamic Continuized CCG (White et al., 2017) but also to develop an accurate treatment of balanced punctuation.
Zhiguo Xie "A Degrees-as-Kinds Analysis of Differential Verbal Comparatives in Chinese and Its Theoretical Implications"
It has been heatedly debated in recent semantic literature whether degrees should be modeled as simple semantic primitives, ontologically complex entities, or even both. In this talk, we add to this thread of scholarly inquiry by re-examining the Differential V erbal Comparative (DVC) construction in Chinese. DVC sentences exhibit two (seemingly) peculiar properties: (i) differential phrases are obligatory to appear in them, and (ii) DPs can serve as differential phrases in them. X. Li (2015) claimed that analyses based on degrees as points would fall short of the DVC construction. Rather, she argued that it is amenable to a degreeless semantic analysis that involves mapping between two sets, instead of comparing the cardinalities of two sets. We report fresh observations to demonstrate that differential DPs in DVC sentences do not manifest the full range of properties of individual-denoting DPs in non-comparative contexts. We submit to the idea that a degree is the entity correlate of a property formed on the basis of a measure, akin to the (Chierchia-style) representation of kinds (Scontras 2017). This new kind of degree, coupled with a set-theoretic difference-based semantics for comparatives, correctly predicts the behaviors of DVC sentences. Our analysis of the DVC construction, therefore, suggests that it can receive a degree-based account, and provides support to the degrees-as-kinds representation by extending its empirical scope. In addition, by combining degrees-as-kinds with a difference function-based semantics, our analysis provides a possible means to circumvent compositionality problems faced by previous degrees-as-kinds analyses.
Kosuke Kuroda "On the Optionality of Coherence Relation Marking: A Comparison between English and Japanese"
In this poster presentation, I deal with the difference between English and Japanese in the optionality of coherence relation marking; specifically, causality marking (e.g., ‘because’ in ‘George is dishonest because he’s a politician.’; ‘therefore’ in ‘George is a politician, and therefore he’s dishonest.’). English is a language which can omit causality marking while Japanese cannot. I will argue that this difference is attributable to the following hypothesis: English is a language in which the kind of information directly available to the speaker and that indirectly available through inference, hearsay, or report can be conveyed without any markers indicating how the information is obtained; Japanese is a language in which the kind of information directly available to the speaker is conveyed without any markers, but that indirectly available through inference, hearsay, or report is obligatorily conveyed with markers indicating that the information is so obtained (cf. Ikarashi (2015); Hirose (2013, 2016)). This hypothesis is independently motivated by those phenomena in English and Japanese that involve the distinction between directly available and indirectly available information.
Satomi Takemoto "Causative constructions of Unaccusative Verbs in Japanese"
In this presentation, I will take up the Japanese causative construction. It has been observed in previous studies that the Japanese causative construction formed from an unergative verb and a transitive verb is grammatical, while the causative formed from an unaccusative verb is non-grammatical (cf. Shibatani(1976); Ritter and Rosen(1993)). This is because the unaccusatives have the corresponding transitive (lexical causative) verbs (cf. Miyagawa(1984); "blocking"). However, I will point out that the unaccusative verbs can create causative expressions; in the case of Cause subjects, not Causer subject. Also I will propose that this difference is influenced by the existence of the vBECOME head (cf. Harley(2008)).
Haruka Shimura "The Emotional Implicature in the Present Perfect Progressive"
This presentation focuses on the feature of the Present Perfect Progressive with emotional coloring. According to Leech (2004), the Present Perfect Progressive has the Perfect features combined with those of the Progressive. Both the Present Perfect and the Present Perfect Progressive can be used to describe the same situations. The previous studies indicate that compared with the Present Perfect, the Present Perfect Progressive describes the completed situations with emotional coloring (e.g. irritation, surprising, and discomfort (Onions 1929, Jespersen 1931)). In this presentation, I will point out this specific function of the Present Perfect Progressive, and systematically analyze it by the semantic approach.
Acknowledgements
We acknowledge support from the Campus in Campus Initiative between the Ohio State University and the University of Tsukuba, the Institute for Comparative Research in Human and Social Sciences, and the Language and Communication Science Laboratory at the University of Tsukuba, and the Department of Linguistics at OSU.
