Abstracts — 2019

Discourse Battlefield: Construction of pro- and anti- hijab discourses in the context of Bahrain

Al-Alawi, Wafa: University of Illinois at Urbana-Champaign

Social and religious discourse has been linked to the construction of multiple harmonious andconflicting group identities (De Fina 2012). With increasing interest in Muslim identities, a visible marker of Muslimness, the hijab, has appealed to recent scholarship. Yet, linguistic research has focused, to a great extent, on diasporic contexts. This study aims to look at the construction of group identities with regard to the topic of hijab in the context of Bahrain. Specifically, I look at how Bahrainis make strategic linguistic choices to discursively identify with three socioreligious group identities that I name: Hijab Proponents, Hijab Opponents and Hijab Moderates.

The data for this research come from 357 online survey responses from adult Bahrainis of different ages. Specifically, their open-ended responses to broad and specific questions about the hijab provide a rich foundation for a critical discourse analytic approach to trace their construction of individual and/or group identities (Bucholtz & Hall 2005). The responses appeared both in English and Arabic and were therefore analyzed manually to find how, through specific linguistic choices, individuals discursively identify with broader socioreligious group identities.

The results show that different group identities make different linguistic choices to report on identical themes. When asked to report on the issue of enforcing hijab on girls at the age of nine, Hijab Proponents used expressions such as: encouragement, persuasion and advice, while Hijab Opponents conversely used: brainwashing, coercion and pressure. Therefore, the individual responses reproduce the social discourses from which they are produced (Bourdieu 1982). This paper also provides an opportunity to speculate about how language research can promote marginalized identities who struggle in the presence of dominant socioreligious discourses in the context of Bahrain.

Morphological Awareness and Advanced Listening Comprehension in English as a Foreign Language 

Bian, Xu: Seattle Pacific University; Cai, Dianmei: Liaoning Normal University; Cai, Xiaojun: Zhongkai University of Agriculture and Engineering

The goal of the study was to investigate the relation between morphological awareness and listening comprehension with 152 advanced Chinese learners of English as a foreign language (EFL). Two facets of morphology, derivational and compound, were found to have different relationships with listening comprehension measured by a listening test from China’s Ministry of Education (including two tasks, the multiple-choice task after interviews and the gap-filling task following a mini lecture). The study provides empirical evidence that derivational awareness is significantly correlated with listening comprehension but also is an important factor for successful performance on the gap-filling listening task. In contrast, compound awareness was observed of weak correlation with listening comprehension. The role that the two kinds of morphological awareness play in listening comprehension has commonality with and difference from that in reading comprehension. 

Acoustic cues in the perception of contrastive vowel quantity in Norwegian

Block, Aleese: UC Davis

The present study aimed to explore the acoustic cues that are used in the perception ofcontrastive vowel quantity in Norwegian. Previous literature on Scandinavian vowel quantity has either focused on Swedish or been incomplete in terms of Norwegian. Behne et al (1996) states that in the production of Norwegian vowels, the only thing that is significantly conditioned by quantity is duration, not quality. The study did not, however, speak to the acoustic cues of perception. Comparable studies on Swedish—a language mutually intelligible with Norwegian—found that vowel quality was “likely not” used as a perceptual cue (Behne et al, 1997). Block (2018) showed that quantity affected both vowel duration and vowel quality; the study also indicated a significant role of secondary acoustic cues in quantity perception. This study sets out to look at the relationship between two key cues: vowel duration and vowel quality.

The perception study investigates how the two aforementioned acoustic cues are used by listeners. Stimuli are created from real quantity minimal pairs involving vowels [i], [a], and [u], matched for frequency, that are first manipulated for both vowel duration and the frequency of the first two formants to create continua that with the duration/formants of a short vowel at one end of the continua and that of a long vowel at the other. Listeners are presented with the word in an ambiguous sentence and asked to identify the token they hear. The percentage of long responses are recorded. The experiment is ongoing and preliminary results will be reported.

If results of this study show a significant interaction between duration and formant frequency, it would not only be a novel discovery, but would supplement the field’s understanding of Norwegian vowel quantity greatly.

 The Devoicing of Middle Chinese Voiced Initials in Wùcūn dialect 

Bo, Jiang: University of Washington

Wùcūn is a village in southern Anhui Province where Huī is spoken. The Middle Chinese(MC) voiced stop and affricate initials have all devoiced in the Wùcūn dialect. Those devoiced initials can be further divided into two categories in terms of whether they are aspirated or not. Based on a corpus of 558 syllables with devoiced obstruent initials, this article tries to explore the rule leading to this two-way distinction and discovers that MC stop initials and initials in level tone (píng tone) syllables tend to be aspirated in the modern Wùcūn dialect while MC affricate initials and initials in non-level tone (tone) syllables however does not. This unique feature distinguishes the Wùcūn dialect from other dialects in the surrounding areas, where MC entering tone (tone) are commonly aspirated. Two hypotheses are proposed in this article to explain this phenomenon: 1) low-registered entering tone might have already merged with low-registered departing tone (tone) before devoicing; 2) Mandarin in different time periods had exerted deep influences on the Wùcūn dialect. 

International Students in the University Spanish Classroom: An Examination of the Otherizing Processes of Hegemonic Whiteness

Canete-Blazquez, Raisa: Oregon State University

The presence of Whiteness in language education has been widely debated in education fields, with scholars such as Flores and Rosa (2015, 2017) arguing that raciolinguistic ideologies can help dismantle its perpetuation through pedagogical approaches. Some focus on the reproduction of hegemonic Whiteness in bilingual education, analyzing the experiences of Latinx students (Flores 2016). Others focus on foreign language education and the use of Whiteness and White supremacy to organize Spanish language classrooms and curricula in post-secondary educational contexts (Schwartz 2018). However, these perspectives have not adequately framed the experiences of international university students, who are particularly vulnerable to hierarchies that reflect power structures centered around language in departments of foreign languages (Valdés et al. 2003).

This paper provides the theoretical framework for my preliminary research, which addresses how Whiteness in university Spanish classes affects international students as “positioned in opposition to the idealized hegemonic White subject” (Flores, 2016, p. 15). The results of a pilot project and anecdotal findings from my experience as a Spanish teacher guide an ethnographic study in basic level university Spanish classes, with participant observation, interviews, and research fieldnotes as methods for data collection. These will help examine the otherizing processes of international students through pedagogical practices within the framework of hegemonic Whiteness (Flores 2016). I argue that the oversight of the different needs of international students hinders their learning experience and may lead to foreign language anxiety (Horwitz 2010), “embarrassment, low self-esteem, lack of desire to learn, and lack of academic preparation” (Valdés et al., 2003, p. 20), and it requires more attention. Therefore, this project will shed new light on the neglected issue of the experiences of international students in the Spanish classroom, as affected by hierarchies that “benefit those who most closely fit the ideals of hegemonic Whiteness” (Flores, 2016, p. 31).

Blunts, Brownies, and Cartridges: An Analysis of Twitter Data Sentiment towards Different Methods of Cannabis Consumption 

Chatusripitak, Michael; Namgung Sanghun; Rivero, Jasmine: San Fransico State University (all authors)

This preliminary research is investigating sentiment of Twitter data towards three different methods of cannabis consumption: smoking, edibles, and vaping. Our goal is to determine whether or not there is a statistically significant difference in sentiments toward these methods. 

Data will be collected from publicly available tweets from the Twitter Application Programming Interface (API). Sentiment will be analyzed using the Python programming language, specifically utilizing Naïve Based Classifiers and the Natural Language Toolkit (NLTK) library. 

Our research has been inspired by work done comparing the sentiments of Twitter users towards different tobacco products (traditional cigarettes, hookah, and e-cigarettes) done by Myslin, Zhu, Chapman, & Conway, 2013. There has been little research into any similar disparity between the different means of cannabis consumption. However, recent publications have investigated individual means and how they are viewed on a national scale. For instance, consumption of edibles can be measured by Twitter users’ public posts (Daniulaityte et al. 2017), entirely separate from the tweets about other forms of cannabis consumption (Daniulaityte et al. 2015, Lamy et al. 2016). 

We believe that our research could serve use to researchers working in public health as well as those in working in the cannabis industry. For public health use, our findings could help decisions in regards to which methods of cannabis consumption should be researched most based on what methods of consumption are most popular. For the marijuana industry, data relating to which methods of marijuana consumption have the most positive sentiment attached to them can be used by businesses in product development and marketing. 

Differences in cross-generational prosodic alignment toward device and human voices

Cohn, Michelle: UC Davis

Alignment toward interlocutors’ speech is a robust behavior reflecting how perceptive people areto voices. Alignment patterns also provide a window into the mechanisms underlying speech production, such as the representational and/or social factors that condition systematic phonetic variation patterns. Speech production varies across individuals from different generations for a variety of social, experiential, cognitive, and physiological reasons. Furthermore, people across generations have different experiences with technology, specifically, voice user interfaces (VUI), like Apple’s Siri. The prevalence of VUIs is a recent phenomenon (e.g., Siri was released in 2011). We consider whether speaker age is a factor in prosodic alignment toward digital device voices.

In this study, older ( x̄ = 61 years old) and younger ( x̄ = 22 years old) adults shadowed productions of words by two human and two device (Siri) voices, varying in gender (male and female). We find that alignment toward device and human voices varies across generations. For vowel duration, younger participants showed the largest degree of alignment toward the male human voice [Age Group*Model Talker, p<0.05] and also aligned, but to a lesser extent, to then male device and human female voice. For f0 variation, younger participants diverged from both the device and human male voices [Age Group*Model Talker, p<0.05]. Younger shadowers did not converge toward the female device voice in duration or pitch. Older participants aligned with the female device voice the most for vowel duration and both male and female device voices for f0 variation.

We discuss how the characteristics of device voices, and human interlocutors, influence the extent to which we apply human social rules of speech communication to personified artificially intelligent systems. As speech becomes a more dominant mode of interfacing with technology, understanding how VUI systems influence human language patterns will be more important.

Best Practices for Teaching L2 Reading

Dykstra, Shelley; Alins, Diego: UC Davis (both authors)

Recent scholarship in second language teaching has called attention to the importance and benefits of incorporating reading, specifically graded readers (i.e., books that are adapted to the proficiency level of the L2 learners), in the L2 classroom (Rodrigo, 2012, 2013; Martinez, 2017). Previous literature has explored levels of anxiety when reading in a second language (Sellers, 2000; Yamashita, 2007) but because incorporating reading in the early stages of L2 learning is not common practice in L2 classrooms, little research has been conducted on the best pedagogical practices needed for students to learn the target language. Seeking to contribute to this growing area of scholarship, in this study, we have integrated graded readers in the first-year Spanish program at a Northern California University and carried out classroom observations of the teachers involved. These observations seek to better understand first year Spanish teachers’ pedagogical practices when incorporating graded readers and to devise methodology that (1) reduces students’ anxiety towards L2 reading, (2) fosters collaboration among peers (and with the instructor) in order to solve reading-related activities and (3) encourages linguistic discussions based on the reading materials. Drawing on classroom observations and teacher questionnaires, the presentation will focus on best teaching practices such as content based teaching, pre-reading and postreading scaffolding, and the role the amount of human interaction plays in reducing anxiety in foreign language classrooms when incorporating reading at the early stages of L2 learning. Identification of best practices will give programs across the country a better understanding of how to prepare teachers to effectively incorporate reading activities in beginner university language courses.

Perceptions of Gendered Language: A Comparison between Native English Speakers and Persian ESL Speakers 

Ebadi, Anita: California State University, Northridge

Native and non-native speakers of most languages have perceptions of gendered language even though not all languages include gender. For instance, while many studies have looked at native speakers’ perceptions of gendered language, little research has looked at non-native speakers’ perceptions of gendered language. My paper specifically observes Persian non-native English speakers’ perceptions of gendered language in contrast to native English speakers. In this study, respondents evaluated written scenarios containing gendered language by hypothesizing the sex of the individuals who had verbally produced the scenarios. Then, I interviewed the respondents so they could report the reasons behind their responses. I conclude that native speakers of English and non-native speakers of English differ in their perceptions of gendered language, with non-native speakers relying more on the gendered language present in the scenarios in order to identify the sex of the individual producing them. 

TED Talk Corpora For L2 College Students: Pedagogical Tools For Spoken Discourse

Evans, Jennifer

A more globalized world is leading to a demand for newer approaches to language teaching. Thus, lexical chunks or bundles, forming the underbelly of oral discourse, may be analyzed via corpus concordance software and serve as a useful pedagogical tool for teachers and a practical learning tool for students in the L2 classroom. This paper illustrates the manner in which authentic language patterns and features can be taught through a hypothetical L2 English lesson based upon a corpus of 200 TED Talks with the language element of focus being lexical bundles with prepositions. The pushed use of such lexicon may serve to enhance the oral discourse of intermediate-level L2 college students as they aim to achieve communicative competence in their academic as well as professional lives.

Perceptual adaptation to Amazon’s Alexa and human voices: asymmetries in learning and generalization of a novel accent across real and AI talkers

Ferenc Segedin, Bruno: UC Davis

Perceptual adaptation to Amazon’s Alexa and human voices: asymmetries in learning and generalization of a novel accent across real and AI talkers Voice-activated digital devices, like Amazon’s Alexa, are a new type of interlocutor in our speech communities. And, similar to human voices, they have idiosyncratic speech patterns that require perceptual fine-tuning. One unexplored area is how perceptual adaptation to a novel accent in speech produced by a digital device compares to adaptation to human voices. Furthermore, prior work has found evidence that when people adapt to a novel accent in one talker, generalization of that learning can apply to new voices. Hence, we also tested whether perceptual adaptation to an accented device voice generalizes to a novel human voice, and vice versa.

First, we exposed listeners to words with a shifted phoneme realization in either a device or human voice (betweensubjects design) (vowelnasality shifted, i.e., CVC words as [CṼC]; CVN as [CṼ̃N]). Later, participants were tested on whether they shifted their identification of CṼ syllables as ending with either an oral or nasal coda in the shifted talker. Higher identification of nasal vowels as CVC words is evidence of adaptation. Additionally, we tested whether listeners applied the shift to a novel voice not heard in exposure, which was opposite “humanness” category of the shifted talker.

A significant interaction between Talker and Humanness of the shifted talker (p<0.01) reveals that listeners learned the shift in device voices (shifted=65% CVC responses vs. unshifted=47%) but not in human voices (shifted=25%, unshifted=27%). This suggests that a novel accent is more acceptable in nonhumans. Furthermore, listeners only exposed to the shifted accent in the human voice, showed generalization to the device voices in testing (57%). Yet, the shift in device voices did not generalize to human voices. We discuss these asymmetries in perceptual adaptation and generalization in elucidating the impact of voice-AI on human speech communities.

Linguistics of Phantasmagoria: Experience and Symbolism in Baudelaire’s “À une Passante”

Hamilton, Cherie: University of California, Riverside

The burgeoning cityscape of Paris during the mid-19th century forced a new trajectory uponintellectual discourse as artistic communities and understandings of urbanity evolved. This paper, part of an ongoing project on the cultural linguistics of Charles Baudelaire, focuses on the symbolist poem “À une Passante” featured in Les Fleurs du Mal, Baudelaire’s 1857 volume of poetry critiquing modern beauty in an era of fetishized commodity and artistic reproduction. Originating in the revolutionary performances of Étienne-Gaspard Robert in the early 19th century, the term “phantasmagoria” referred to his use of projection to create a series of flashing images that seemingly floated in front of the audience. The term was later used by Walter Benjamin in his Arcades Project to describe the commodity culture’s rule over the Parisian metropolis by mid-century. Baudelaire’s use of ellipsis, tenses, and verb clustering establishes his encounter with the passante in a linguistic framework that itself is phantasmagoric in nature, conveying a sensory overload and confusing notions of individuality. Through an emphasis on the poem’s precise points of linguistic metamorphosis and the relationship of parole to langue, this work explores an overshadowed area of transformation that captures the situational aspects of tenses and moods in language by establishing language as deeply entrenched in, and connected to, speakers’ cultural perceptions.

Evidentials as a valency changing operation and what Evidentials can reveal about Morphological theory

Hardy, Thomas: UC Davis

An evidential is a marker of information source as part of a closed grammatical system; pertaining to evidence (Aikhenvald, 2018: 41). All languages have a way to express the information source for a statement, in contrast to this lexical ability is the morphological ability to mark the information source grammatically (Aikhenvald, 2004). Evidentials are not restricted to the semantic level. Evidentials are morphologically scattered and can occur attached to any type of lexeme; however, they cause the event-changing operation of experiencer backgrounding. This paper deeply explores evidentials in Korean, in order to give a concrete example of the complete effects that evidentiality has on a language. This complexity allows general statements to be made about morphological theories and what evidentials can contribute to them. Additionally, the pragmatic effects of evidentials and a possible measurement strategy are presented.

Fake News: Generic Masculine in German

Harris, Sarah: UC Berkeley

 The generic masculine (GM), i.e., the use of masculine terms to refer to humans, is prescribed in both English and German as a gender-neutral stylistic device and is recommended in situations where gender is irrelevant or unknown. GM is asymmetrically referential and therefore ambiguous: While men are always referenced, non-male persons must determine whether the +male term is used inclusively (generically), or exclusively (only referring to men). Research in both languages indicates that GM is interpreted overwhelmingly as referring to men, which can result in significant legal and economic consequences for women and non-binary individuals. While GM faces a decline in English, German’s gender system complicates discussions of sexism in language, including whether language practices should change, and what those changes might be. Absent from this discussion is an analysis of the effects of GM in the media, specifically as it relates to visibility and perception of participants and their gender. My research examines German print articles about unidentified persons and argues that the use of GM in media can engender misleading interpretations of reality, which are biased toward male participants. The ambiguity of GM renders it inadequate for news reports, and its use should be discouraged. In circumstances where its avoidance is not feasible, I argue that media reports should indicate to the reader when +male terms are being used generically to correct for a male bias. 

Getting to know you: /s/ and interlocutor familiarity in social interaction !

Keshav, Maya: UC Santa Barbara

People pronounce /s/ in ways that vary systematically with gender. This is not caused by differences in anatomy: instead, /s/ varies with gender categories that are socially constructed in intersection with class and age (Stuart-Smith 2007). Even among people sharing a broad gender identity, productions of /s/ vary systematically with sexuality and specific gender presentation (Zimman 2017). Nevertheless, few have examined how individuals vary /s/ as they interact with different people.

In this project, I examine the speech of one transgender male speaker in four twenty minute interactions, each with one interlocutor. The interlocutors were all strangers to the speaker, all similar ages, and included a transgender male, a transgender female, a cisgender male, and a cisgender female. In each session, the pair of speakers was alone in a room, recording themselves with a Zoom recorder. I analyze the trans male speaker’s productions of /s/ in each interaction, measuring the center of gravity. I also examine if /s/ changes throughout each interaction, as the speakers grew more familiar.

This model of gender in social interaction offers a useful alternative to concepts of “safe space” for gender non-conforming people. Gratton (2016) showed that non-binary speakers vary speech depending on their environment, since they have different levels of need to present as explicitly gender non-conforming in different places. However, I suggest that social interaction is just as important as physical environment. People perform gender differently depending on their needs and desires in particular interactions; and this may also change moment by moment as an interaction progresses. While this project focuses on transgender people’s performance of gender, everyone similarly performs salient facets of identity in every social interaction. !!

Korean EFL learners’ development of bi-clausal wh-questions

Kim, Jung-Tae:Incheon National University in South Korea

The present study examined the interlanguage forms of English bi-clausal wh-question produced by Korean EFL learners and see how those forms change as the learner’s English proficiency increases. Specifically, this study asked if the change can be explained by the ‘derivational theory of complexity’ which claims that language develops from derivationally simpler structures (i.e. structures involving less syntactic operations) to more complex ones (i.e., structures involving more syntactic operations). One hundred seven adult Korean learners of English in three proficiency groups took part in a production task designed to elicit English bi-clausal wh-questions. The results showed that Korean EFL learners produced a variety of alternative structures to English bi-clausal wh-questions, including wh-scope marking, silent scope marking, wh-scope marking with wh-in-situ, L1 clause order transfer, and Wh-about-wh constructions. The appearance of these alternatives indicates that Korean learners have relied on derivationally less complex interlanguage forms in order to compensate for their lack of L2 competence to execute the highly complex derivational process needed to form bi-clausal wh-questions. Comparison of three proficiency groups supported the derivational theory of complexity. In general, more proficient L2 learners produced greater number of long distance (LD) wh-movement structures while less proficient learners produced larger number of structures that are derivationally less complex, such as silent scope marking construction and scope marking with wh-in-situ construction. The general interlanguage order of the three most common production types in our data (silent scope marking construction (e.g, do you think where Mary had dinner?) -> wh-scope marking construction (e.g, what do you think where Mary had dinner?) -> LD wh-movement construction (e.g, where do you think Mary had dinner?)) is consistent with the order of degree of derivational complexity. It is argued that derivational complexity plays a role in the developmental process of L2 interlanguage for bi-clausal wh-questions.

 Rhotic Variation in Heritage Persian

Kline, Tyler: UC Davis

Western Persian has been described as having at least three allophonic variants of /r/: in intervocalic positions it surfaces as a flap [ɾ], in word-final positions it devoices as [r̥], and it surfaces as [r] elsewhere. Additionally, some accounts include an alveolar approximant [ɹ]. Interestingly, scholars have recently observed that Persian heritage speakers will substitute [ɾ] in intervocalic position with [ɹ]. The presence of [ɹ] in intervocalic position as a variant of /r/ in heritage speakers can address questions about how language contact can lead to variation and change in bilingual populations. 

Models of language transfer can help explain when rhotic variants are substituted by another unexpected variant. One of these models is the Perceptual Assimilation Model (PAM), which refers to instances when two sounds from different languages merge as one due to behaving in categorically similar ways. With this in mind, I argue that Persian heritage speakers acquire an English-based allophonic framework for certain consonantal classes before rhotics, which are not acquired until later in the acquisition process. The phonological nature of the sounds that fall into the English rhotic class will interact with Persian rhotics. Given that English [ɾ] is an alveolar plosive allophone, if rhotics are acquired later then the bilingual speaker treats Persian and English rhotics as belonging to the same category and will essentially merge them, with approximants being substituted for intervocalic flaps. 

Focusing on these emerging patterns, while based on models like PAM, can help explain variation patterns across communities and make predictions about the specific types of sound changes that might occur in cases of bilingualism. Examining this case of Persian rhotic substitution may lead us to a better understanding of how languages in contact affect bilingual phonologies and how such interactions and their effects spark sound change. 

Intergenetic synchronic polysemy of instrumental ‘take’ serial verbs and tendencies: cases of Kwa, Gur, and Volta-Niger

Lee, Taegyeong: University of New Mexico

Serial verb constructions (henceforth SVCs) have been widely attested in Oceania (Crowley 2002), Southeast Asia (Bisang 1995), and West Africa. Among are Western African languages where SVCs have long been studied, tracking back to Riis (1845) and until recently Aikhenvald (2018). In this ongoing study, I present the grammaticalization of instrumental ‘take’ SVCs in Twi studied by Lord (1993:ch.5), and interpret this grammaticalization within a functional view grounded in Cognitive Grammar and subjectification tendencies (Traugott 1989). I respectively argue that (i) a less profiled subevent of taking in SVCs undergo grammaticalization, while a more profiled subevent of the following action does not, despite the fact that the ‘take’ verb can semantically predetermine semantic components of the action verb and that the ‘take’ has grammaticalized from a complex figure construal of coordinate clause constructions; (ii) synchronic polysemy of ‘take’ (take > possess > use) are subject to the unidirectional path where propositional statements come to be expressive for pragmatic inferences.

The implicational hierarchies in Kwa languages, of which Twi is a member, (Shluinsky 2016:4.3) of the synchronic polysemy (instrumental > manner > material, or instrumental > comitative) are also discussed within the framework of subjectification tendencies. I argue that (i) instrumental meaning is extended to the meaning of manner due to our tendency to extend physical domains to perceptional/cognitive domains, and that (ii) instrumental meaning is separately extended to comitative meaning due to our tendency to extend physical events to interpersonal domains. The ongoing research is conducting a typology study on variations of these hierarchical implications in some Volta-Niger and Gur languages genetically and geographically affiliated with Kwa, and in correlating their different hierarchies with the degrees of their morphosyntactic characteristics and semantic features.


Lemos, Keyla: Federal University of Ceara (Brazil)/University of California, Irvine

The access to computers, smartphones, and other electronic devices may have produced technology experts; however, it is arguable if those experts in technology are knowledgeable when it comes to digital literacy. While there is a positive effect regarding the access to information, the consequences of this shift, i.e., from reading on paper to reading on a screen, are still unknown. For Coiro (2011), one of the differences concerning online reading is the fact that the traditional reading and thinking strategies are not sufficient for fostering successful navigation and comprehension of online texts. According to Castek and Coiro (2015), online reading requires complex skills that are not included in most literacy curricula. Baron (2015) believes that digital reading presents more distractions; therefore, print materials are more suitable for reading complex texts, like academic books, for example. This quantitative study aimed at evaluating comprehension and recall of information in academic and journalistic texts read in different media (print, desktop, tablet and smartphone) in order to assess the influence of the media and the text genre on reading comprehension and retention. Students (n = 37) majoring in TESOL participated in the research. Main findings corroborate the assumption that there are no significant differences in the levels of comprehension and retention of information between reading using a digital format or a printed one. There was an indication that the text genre more than the media influenced the results of comprehension and retention, since readers performed better on tests using journalistic texts. Thus, the results of this study imply that there should not be concerns regarding the use of digital media for teaching and learning as levels of comprehension are not affected by the format of the text.

Personal Reference and Equivocation: Circumventing Gender Indexicality in Political Interviews

Licata, Gabriella: UC Berkeley

Political discourse involving interlocution with female politicians reveals an underlying sexist substrate, wherein female politicians are indexed in gendered frames, i.e., Mother, that are interlaced into public space frames, i.e., Mother/Politician, uncovering a semiotic process whereby gender ideologies are created, disseminated, and maintained. Female politicians find themselves discussing personal matters and indexing gendered identities within the public forum, especially in conversations where context necessitates a response (i.e., an interview). According to Bavelas et al. (1988), politicians equivocate when the response choices are disagreeable to some party but nonetheless they must fulfill their second-pair part. This study takes a qualitative approach to analyzing how equivocation is disseminated in personal reference forms by Virginia Raggi (Five Star Movement, or M5S)—current mayor of Rome—in a pre-election interview (March 17, 2016) on an Italian political talk show. Findings indicate that when the interviewer attempts to situate Raggi in personal-space frames, wherein Raggi could easily index gendered identities, she consistently prevaricates, avoiding use of first-person references and personal matters by alluding instead to plural-person frames that highlight membership of a group (i.e., Working Parents), allowing her to maintain positive face within the frame of Politician and not Working Mother or Female Politician, identities that the interviewer constantly underscores with the intention of undermining her capabilities as a politician. Media discourse presents a forum that can present overt and covert sexism via linguistic choice and footing, discriminating female politicians as distinct from their male counterparts and conflating their private- and public-space identities. However, as demonstrated in Raggi’s speech acts, the use of personal reference is one mechanism that female politicians can manipulate to avoid having to index gender and enter into personal-space topics—sexist interview techniques used to delegitimize their potential in public roles—and thus maintain positive face in order to keep their rightful place in politics.

Examining Musical Skills and Phonological Ability in Second Language Acquisition

Lindblad, Mayu: UC Davis

Past studies on the association between music and language have shown three directional relationships: music to language, tone language to music, and bidirectional relationships. These studies emphasized that musicians gain refined phonological skills through musical training and such skills may be transferred to second language acquisition. To understand the mechanism of musical training and its effects on both perceptual and productive phonological ability in languages, the current study investigated the relationships between musical aptitude and second language skills. Seventy native-English-speakers learning Mandarin Chinese in middle and high school were recruited. To measure Mandarin Chinese perceptual skills, the participants listened to 16 syllables, 20 two-syllable words, and five sentences and identified lexical tones. To measure Mandarin Chinese productive skills, the participants read aloud another set of syllables, words, and sentences in Mandarin Chinese. Additionally, their recorded Chinese pronunciations were rated by two native Mandarin speakers. Music skills were measured by a music aptitude test and survey data. Based on their music skills, the participants were grouped into high and low music groups. Data are currently being analyzed and I hypothesize that the high musical aptitude group will outperform the low musical aptitude group in the overall accuracy of both Mandarin Chinese listening and pronunciation. However, I hypothesize there will be no differences in lexical tone types in perception and production based on results from the past studies. Pedagogical implications in applying music to second language education, particularly Mandarin, will be discussed.

Fostering Global Competence in Learning Japanese Language

Nagai, Ayako: University of California, Irvine

This paper discusses the significance of cultivating global competence and introduces a thematic unit of social justice incorporated in beginning-level Japanese courses as well as research projects implemented in an advanced-level Japanese course. In order to become globally competent, students need to investigate the world, recognize perspectives, communicate ideas, and take actions (Mansilla & Jackson 2011). In the beginning-level courses, the unit, which focuses on unfavorable labor conditions known as “black arbeit” in particular, was designed, incorporating the basic tenets of World-Readiness Standards, Core Practices, and Backward Design with Integrated Performance Assessment (IPA). It starts with social justice take-away understandings, essential questions, and summative assessment tasks in three modes of communications, followed by language objectives and formative assessment tasks in sequence. Five-step daily lesson plans are also shared. Furthermore, sample students’ works are showcased as their achievements. In the research projects of an advanced-level course, students did research on a social issue of their interests and presented it in class in order to propose solutions for the issue. The topics which students selected include animators’ unfavorable job situations, kids’ eating alone, convenience stores in saturated state, and sluggish publishing industry. Formative assessments were applied in the process of projects where the teacher provided constant feedback and students altered the course of their project accordingly. At the end of each presentation, students discussed the solutions with their classmates. This paper emphasizes the interactive structure of becoming globally competent, which rotates from investigation, recognition, communication to actions. It also helps concretely visualize how the basic tenets of global competence were applied in designing a thematic unit in beginning-level courses as well as implementing research projects in an advanced-level course. 

Finnish language attrition in an English language environment, lexical processing tasks

Nana Lehtinen: Department of Psychology and Speech-Language Pathology, University of Turku, Turku, Finland; Kati Renvall: Department of Psychology and; Speech-Language Pathology, University of Turku, Turku, Finland; Marja-Liisa Helasvuo: Department of Finnish Language and Finno-Ugric Languages, University of Turku, Turku, Finland

This research project focuses on changes in the native language (L1) of late bilinguals, namely L1 attrition. L1 attrition can be described as gradual, non-pathological loss of language by healthy individuals. It is typically seen in immigrant settings where L1 falls into disuse or is used alongside an environmental one (L2). First markers of language attrition are often perceived on lexical level and described as a smaller or a less accessible lexicon. Here the focus is on language attrition of Finnish immigrants in an English speaking environment, specifically on lexical level processing skills. The study is a part of a PhD project, where data collection follows international guidelines for language attrition studies, namely the Language Attrition Test Battery. For the purposes of this presentation we will analyze five Verbal Fluency Tasks (VFT) from the data and address the following questions: Can markers of language attrition be detected in L1 spoken by first generation immigrants in lexical retrieval tasks? Are there differences between performances in L1 and L2 in the experimental group and how do these relate to performance in L1?

Research design is an experimental case-control group study (experimental group n=41, control group n=45), in which experimental group consists of Finnish immigrants in Northern California, USA and control group consists of monolingual Finnish speakers living in Finland. Five VFT tasks (two semantic categories and three phonological categories) in L1 for both groups will be analyzed for group differences using applicable statistical methods. In addition, VFT tasks in L1 and L2 for the experimental group will be analyzed using applicable statistical methods to assess language balance and processing skills in both languages as these might have an influence on the performance in L1. Analyses are ongoing and results will be shared in the presentation.

 Hong Kong’s language policy change: its impact on social identity and culture 

Po, Byron: UC Davis

Since 1956, the PRC’s official language policy has always been focusing on the “unification of Chinese”, the promotion of Putonghua (Mandarin), its official romanization system of pinyin and simplified Chinese characters, the removal of illiteracy (Bolton & Lam 2006). It is inevitable that the PRC government would converge the majority language from Cantonese to Mandarin in Hong Kong after its transfer of sovereignty because of the ideology of unification. In 1998, a year after the handover, Putonghua became a mandatory subject in all schools. Five years later, Hong Kong government started to promote “trilingual and biliteracy” policy, which aimed to increase the competitiveness of students. In 2008, regardless of the lack evidence on how Mandarin could be a better teaching medium than Cantonese, the government heavily subsidized the primary and secondary schools that are willing to replace Cantonese to Mandarin in their Chinese class. Throughout this series of language policy changes, the status of Cantonese declines overtime and native Cantonese speakers afraid that one day, Cantonese would be completely replaced by Mandarin and their identity would change from “people from Hong Kong” to “Chinese”. The main focus of this paper would be how the language policy shift shatters the identity of Hong Kong, affects the local culture and the ideology behind the series of pro-Mandarin policies. Beyond that, the suppression of non-Mandarin language in China would be discussed throughout examples from Guangzhou, which is the biggest-Cantonese speaking region in the world, and example from Xinjiang, which is the home of most of the Uyghur speakers (Dwyer 2005). The paper will conclude with a discussion of current language policy on various languages in Chinese and Hong Kong (Lam 2005), as there are some parallel characteristics and it may foretell the future of Cantonese speakers’ culture and social identity. 

 The Contribution of the Dialogic Discourse Analysis to Understand Gender Diversity in Brazilian Academic Context 

Porto, Ludmila: Universidade Estadual da Paraíba

This paper aims to understand how gender diversity is interpreted by students who attend to Portuguese, Spanish, and English Languages and Literature Undergraduate Courses in the Department of Letters and Arts at the Universidade Estadual da Paraíba, Brazil. As those students intend to become teachers after graduation, we believe that they shall be ready to deal with diversity since their formation period. Thus, first of all, we contextualize gender studies at the present time (Butler 2016; Connell & Pearse 2015), in Brazil, in the field of languages and literature studies, in order to think about how Brazilian people deal with gender issues in academic contexts and out of them. Then, based on Dialogic Discourse Analysis (Bakhtin 2003; Brait 2006), we try to understand how gender diversity emerges from the speeches of those students who answered a questionnaire and a semi-structured interview about the subject. Despite the importance the field of gender studies has reached in the last years in the United States, the discussions on gender as a social and cultural construction in Brazil are not socially well satisfactory yet. That’s why the preliminary results of the analysis of the data demonstrate that those students are not able to deal with gender diversity in their own lives, which can be considered a problem for their future professional formation, especially considering current Brazilian political scenery where intolerance overcrosses private and public boundaries. In conclusion, the results of the research point to the need for more inclusive teacher training regarding gender diversity in Brazil. 

Linguistic policies and indigenous languages in the present-day Colombian territory during the late colonial period

Ramirez Restrepo, Maria del Pilar: UC Santa Barbara

The communication problems between Christians and Indians when Europeans crossed the Atlantic Ocean has been recurrently interpreted as superfluous. It has been taken for granted that mutual understanding was immediate, or else insurmountable, as the only valid forms of communication were the harquebus and the cross. However, this problem may be more complicated. The irruption of Europeans in the continent not only involved their interaction with people whose customs were unknown to an Iberian of the sixteenth century but also with linguistic structures that did not correspond to known morphosyntactic systems. Two hundred years after the arrival of the Iberians to the New Kingdom of Granada, which was the territory of present-day Colombia, many natives, especially those who lived far from civil and ecclesiastical authorities, had not been incorporated into the colonial system. They did not speak Castilian, they did not dress like Spaniards, and they did not have their customs. Although the Spanish Crown had established policies of Hispanicization throughout the colonial period, many of them had not been fulfilled or had only partially developed. The goal of this paper is to argue that linguistic policies application depended in no small extent on local contexts. Analyzing manuscripts and bibliography of both linguistics and history, I will argue that the goal of Christian missionaries to learn the indigenous languages and write indigenous languages grammars and dictionaries during the eighteenth century was not to offer treaties for the linguistic and ethnological knowledge, as many scholars have affirmed. Missionaries used diverse local strategies (such as choosing an intermediate language that did not correspond to the most spoken language among natives) to convert them to Christianity, so then natives could serve as vassals for the Spanish Bourbon Empire.

Interpretation of Political Cartoons on Mexican Online Newspapers: A Pragmatic Analysis

Reyes-González, Ezequías: Universidad Autónoma de Baja California; Saldívar, Rafael: Universidad Autónoma de Baja California

The purpose of this research is to describe the interpretation process of political cartoons (PCs) that combine both image and text from the standpoint of two well-known theories namely the Cooperative Principle -CP- (Grice, 1975) and the Relevance Theory -RT- (Sperber & Wilson, 1986). The research is qualitative. Using the criterion sampling, a corpus of 19 cartoons depicting presidential candidates for the 2018 general election in Mexico was collected. All the cartoons were published on 4 Mexican online newspapers. A pragmatic analysis was done. Each cartoon was analyzed twice; once within the frame of the CP and then the RT. The role of humor was also taken into account in both approaches. Although it is commonly accepted that intentional violation of Grice’s maxims is necessary to achieve humorous effect, it was found that humor is not always the ultimate goal of PCs. In fact, the findings of the research show that cartoonists usually flout conversational maxims with some further communicative intention. Hence, the CP holds in the interaction between the cartoonist and the reader as the artist’s message is embedded in the cartoon and the reader derives implicatures. On the other hand, relevantists claim that interpretation can be achieved without needing any underlying principle in force. Within the RT frame, the reader will interpret a cartoon in terms of a variation of the mutual parallel adjustment. Analysis shows that both visual and verbal explicatures are commonly enriched pragmatically for the retrieval of implicated premises and derivation of implicatures as long as there is enough contextual information. Although humor researchers seem to favor a more cognitive approach, it will be suggested that Grice’s CP may still be valid when it comes to interpretation of political cartoons.

Doralzuelan or Venezuelan in Doral? A case study of Venezuelan immigrants in the south of Florida

Romero Pino, Blanca: Arizona State University

The steady influx of Venezuelan immigrants to the United States has resulted in the creation of a close-knit community of these immigrants in the city of Doral, Florida, now nicknamed Doralzuela given the strong imprint Venezuelan have left in this city. This study aimed at gaining understanding on how the process of immigration and settlement in the context has affected Venezuelan immigrants’ identity, their perception and use of English and Spanish in daily interactions, and how, or if, their bonds with the home country has affected their incorporation to the host society. The study followed a qualitative design. Eight semi-structured interviews were conducted and analyzed following Riessman’s (2008) notion of dialogic narrative analysis. Two of the emergent themes of this analysis was the negotiation of identity through language and Doralzuela, the new Venezuela. These themes were discussed and multiple and distinct views on each theme were identified.

Thai-English Code-Mixing in Thai Reality Television Shows: The Face Thailand and The Face Men Thailand

Ruanglertsilp, Ekkarat: Arizona State University

This research explores the linguistic phenomena of Thai-English code-mixing on Thai modeling competition reality shows: The Face Thailand season two and The Face Men Thailand season one. These two shows were selected because of the higher frequency of code-mixing incidences by bilingual individuals compared to other Thai reality shows. This research investigated the relationship of gender and role of the bilingual cast members with the incidences of code-mixing. This is to gain insights into whether the male cast members code-mixed more frequently than their female counterpart, and whether the cast members who have the roles of mentor code-mixed more frequently than their contestant counterpart. Recorded data from each of the reality television shows was collected from the first five episodes, standardized, and analyzed with statistical independent sample T-test.

The results demonstrated that female cast has a higher number of code-mixing incidents than their male counterparts. However, statistics showed this number is statistically insignificant. Regarding roles, the results revealed that mentors code-mixed more frequently than contestants and the number is statistically significant.

During the presentation, the author presents results which both conform to and challenge existing relevant literature in the field (e.g. Trudgill, 1972). Reasons which explain why the bilingual cast members employed Thai-English code-mixing in conversations are also discussed.

Most research on mixed language varieties have been conducted on naturally occurring data, but not on recent semi-scripted data (reality show dialogue). The semi-scripted data is rare in Thai-English code-mixing literature and worth exploring. Furthermore, researchers (e.g., Rosenthal 1995 and Hill 2005) claimed that the art forms of scripted television shows mirror humans’ realities. This research contributes to the gap in the literature of Thai-English codemixing in Thai mass media, especially reality TV shows, and further the understanding of how the attitude of English language is perceived in Thailand’s modern society.

Human vocal alignment to voiceAI is mediated by acoustic expressiveness

Sarian, Melina: UC Davis

When humans converse, they tend to align their acousticphoneticspeech patterns. Degree of vocal alignment is also thought to convey empathy between interlocutors. But, what if the interlocutor is not a real human? Are humans’ acoustic-emotional alignment patterns driven by our desire to become close to a sentient individual, or are they an automatic, learned response to interlocutors who display cues of humanity? To what extent humans align with a voice-activated digital device’s displays of emotional expressiveness, is an understudied question. This study investigates alignment of expressive speech toward voice activated digital devices, specifically, Amazon’s Alexa system.

First, 10 participants completed a shadowing task of words produced by a voiceAI system, Amazon’s Alexa, containing either regular and expressive acoustic-phonetic properties. Expressive productions contained longer vowel durations, larger pitch contour ranges, and lower pitch relative to regular productions. Prior to shadowing, participants produced each item to serve as baseline productions. Following the shadowing experiment, 35 independent raters assessed overall degree of alignment of the shadowers’ items to the model talkers’ recordings, relative to baseline productions.

Results for the device voice reveal a main effect of Condition, with greater imitation of expressive†productions (p<0.001). In other words, people align to expressive productions in device voices to a greater extent than regularly produced items. Also, there was greater imitation of expressive tokens for male speakers (Imitator Gender*Condition, p<0.001). Data collection testing alignment toward human expressive and regular productions is currently underway for comparison.

We find that humans align robustly with artificially intelligent voices when they convey hyper-expressiveness. We discuss these findings for speech production models, as well as applications for the field of affective computing. This work presents a novel tests for our scientific understanding of human conversational dynamics and the social role of voiceAI in human speech communities.


Simon, Megan: San José State University

Mam is a Mayan language spoken by 480,000 people in the Guatemalan highlands. Previousresearch has noted considerable regional dialect divergence and reduced intelligibility between variants, but very little research has been conducted on the degree of intelligibility between them (see England 1983:610, Godfrey & Collins 1987, B’aayil et al. 2000).

Understanding the linguistic landscape of Mam is increasingly relevant to the immigration court system in the United States. In 2015, Mam was the ninth most common language in immigration court (US DOJ Executive Office for Immigration Review 2017). Given the existing research, it can be reasonably hypothesized that an interpreter from one region may not be able to adequately interpret for a client from another region. This paper addresses the question by (1) estimating the degree of intelligibility between various regional dialects and (2) directly investigating whether Mam interpreters in the US perceive dialectal variation to be a challenge in their work.

In the first section, a computational approach is applied to existing word list data from B’aayil et al. (2000). Phonetic distances between variants are calculated using the Levenshtein distance algorithm, a measure which is strongly correlated with the results of comprehension tests in other languages (see Gooskens 2006, Yang & Castro 2008, Yang 2012). Social factors, such as contact, attitude, and bias also play a strong role in intelligibility between dialects. Therefore, the second part of the study consists of sociocultural interviews with Mam speakers in the US and Guatemala. Additionally, I interview Mam interpreters and immigration lawyers to ask directly how differences between regional dialects factor into the interpretation process.

This investigation tests the implementation of a relatively recent computational tool in linguistic fieldwork, Levenshtein distance, combined with traditional methods in sociolinguistics, to address questions that are directly relevant to the needs of indigenous language communities.

Links between Code-switching and Task-switching in Bilingual Preschoolers

Sung, Priscilla & Akhtar, Nameera: University of California, Santa Cruz (both authors)

Several studies indicate an early bilingual advantage in various components of executive functioning (EF), but others do not. One potential explanation of these inconsistencies is that only some types of bilingual experiences enhance specific aspects of EF. 

Code-switching (CS), defined broadly as shifting between different languages, is one element of bilingual experience that may strengthen task-switching (TS) abilities, one component of EF that involves shifting between different rule sets. Recently, Green and Abutalebi (2013) proposed that engaging in CS with different monolingual speakers exercises cognitive abilities that are most likely to translate into TS benefits. The current, ongoing study is centered on this research question: Is there a link between bilingual preschoolers’ TS performance and specific CS practices at home and school? 

We examined a group of young emerging bilinguals: 3- and 4-year-olds (current n = 16) attending a Mandarin-English bilingual preschool. We measured children’s TS with five progressively difficult tasks. We measured their CS practices at school via a teacher survey, and their CS practices at home via a 7-day parent diary study. 

Preliminary analyses show that children who engaged more frequently in CS practices when switching conversational partners at school performed significantly better on the most difficult TS task (2(1, N = 16) = 8.91, p < .01). Children who engaged more frequently in these CS practices at home also tended to perform better than those who engaged less frequently in these practices, but this difference did not reach statistical significance (2(1, N = 16) = 4.40, p < .11). This suggests that early experiences with CS that involve switching conversational partners are linked to enhanced TS performance. These findings add important nuance to our understanding of the relationship between bilingualism and EF by highlighting one bilingual experience that may contribute to preschoolers’ developing TS abilities. 

 A Sociophonetic Analysis of the /a/~ /æ/ Vowel Merger in Trinidadian English 

Tikasingh, Genevieve: UC Davis

Trinidad and Tobago is the southernmost country in the Eastern Caribbean and represents a linguistic site in which both a mesolectal English Creole and a local standard variety of English exist. Recently, research on Trinidadian phonology has proposed an ongoing merger between the mesolect and the local standard (Youssef and James, 2008). Similar research has documented a number of ongoing vowel mergers, including that between the two variants of the TRAP vowel, /a/ and /æ/, of which /a/ has typically been considered more characteristic of the mesolect and /æ/ more closely associated with the local standard variety. This study investigates the proposed TRAP vowel merger in the speech of local radio show hosts, interviewees, and one call-in speaker in Trinidad. Both the proposed /a/~/æ/ merger and the potential for /a/ to be raised to /a̝/ are considered. The specific research questions are: Are /a/ and /æ/ merging in normalized vowel space? Is /a/ being raised to /a̝/? Eighteen tokens of the TRAP vowel for each of 12 speakers from two radio stations in Trinidad were analyzed using Praat and the DARLA suite of semi-automated vowel alignment and extraction tools. The resulting vowel extraction data from this study are compared with the vowel systems of 36 speakers of AAVE, Gullah, and Caribbean Anglophone Creoles from the 1840s–1990s documented in Thomas (2001). Whereas /æ/ in the speech of African Americans in Thomas’ chapter occupies a distinct vowel space that is higher and fronter, the data within the current study show /æ/ to be realized lower and farther back in the mouth, and therefore closer to /a/, than has typically been considered characteristic of /æ/. Results from the current study therefore suggest that a merger between /a/ and /æ/ may in fact be occurring in Trinidadian radio speech. 

“When Heroin was a Hero and Solutions created Problems”: The discourse of opioid policy enactments

Torres, Peter: UC Davis

In recent years, there has been great interest in the discussion of chronic pain management (Matthias et al. 2014; Henry et al. 2016), yet not much attention has been devoted to the discourse of opioid policy enactments at the local level. This is particularly important since there are no objective clinical tests in measuring chronic pain (Sullivan and Ferrell, 2005), which means treatment decisions are often based solely on the patients’ verbal manifestations of suffering (Burgess et al., 2008). In this study, I look into opioid policies and their enactment in a particular community of (medical) practice by analyzing the resulting discourse employed in four different medical consultations.

I illustrate how the language of opioid policies shifted its focus across time—from aggressively addressing the crisis of pain undertreatment through promotion of narcotic-use to eventually solving the resulting crisis of pain overtreatment/drug overdose by advocating for alternative treatments. I then use the language socialization framework (Heath, 1983; Schecter and Bayley, 2004) to illustrate how policies affect language-use in the medical space, from the perspectives of physicians who are often villainized in this narrative. I propose that physicians face a cycle of language (dis)socialization when adhering to the constantly-changing policies.

Lastly, I conduct discourse analysis to (1) illustrate the ways physicians socialize their patients into alternative pain treatment, (2) describe how patients react to interventions proposed by physicians, (3) and demonstrate the discourse strategies physicians employ as a response to patient disagreements. Setting shared goals is vital in reducing inappropriate opioid prescribing (Henry et al. 2017). However, morally-fraught discussions surrounding opioids are often dominated by fear and mistrust (Merrill et al. 2002). This study points out specific discourse practices that could be employed in designing useful interventions concerning opioid treatment, especially by physicians who are constantly re-socializing themselves to new approaches in addressing disputes.

Grammaticalization of Yǒu + VP Construction in Early Southern Min Literature

Tsai, Ya-Ting: San José State University

Yǒu 有is a verb that indicates possession or existence and typically only takes noun phrases as objects. Yǒu preceding a verb phrase (VP) was previously considered a deviation existing in southern Chinese varieties such as Cantonese and Southern Min. However, this construction occurs extensively in both Taiwan and China nowadays. In this construction, yǒu has developed a modal-verb-like function, which indicates a speaker’s attitude of confirmation or suspicion towards the referring event. Scholars agree that Taiwan Mandarin borrowed the usage through long-term contact with Taiwanese, a branch of Southern Min, then this usage spread to Mainland China through Taiwan’s pop culture. Nevertheless, its origin remains unclear. Some scholars argue that this usage once existed in Old Mandarin roughly 2600 years ago, but was lost in Modern Mandarin while it was preserved in Southern Min. Southern Chinese varieties are known for preserving many Early Mandarin constructions because of geographical isolation. Such a hypothesis is reasonable, but it has not been fully investigated. To explore how the function of yǒu changes over time and its origin, a diachronic analysis of historical Southern Min is necessary. This paper analyzes six early Southern Min literary works from 1365 to 1911, and examines the different complements of yǒu. It finds that the yǒu stative verb construction significantly outnumbers the yǒu dynamic verb construction, perhaps because stative verbs are semantically similar to nouns in terms of time stability. Thus, the yǒu + VP construction may have first originated with yǒu stative verbs later accepting dynamic verbs as the modality function evolved. As a result, almost any type of verb is able to occur after yǒu in Modern Taiwanese and Mandarin.

Tracking the Formant Motion of Vowels: The Effects of Consonants

Wheeler, Madison: UC Davis

The purpose of this study is to examine consonants affecting vowel formant motion. The aggregate function of formant measurement at the beginning, middle, and end of each vowel were taken from a database of recorded utterances from 20 California-English speakers. A total of 3,600 /bVd/, /dVd/, /bVb/, or /dVb/ utterances were spoken by UC Davis students and faculty. Participants were asked to read and produce real and synthetic words in seclusion from a words list. From these recordings the vowels were segmented, and their times normalized so that the mean of their F1, F2, and F3 values could be measured. The movement of these vowels due to their consonant environment can be described and modeled as a line or parabola. Results include the definite movement of vowels as dependent on their environment as portrayed through the figures below: [Unfortunately the figures could not be displayed on this site.]

A Syntactic Analysis of the Floating Quantifier Construction and a Related Construction in Mandarin Chinese

YuHua, Zheng: National Taiwan Normal University

This investigation aims to explore a set of quantifiers in Mandarin, which has great resemblance with the Floating Quantifier all in English, as shown in (1). First, this set of quantifiers can float in sentences like the FQ all. Secondly, FQ all and this set of quantifiers can refer to their associated nominal phrases even though they are not adjacent. Finally, like FQ all, this set of quantifiers can have the similar interpretation to their non-floating counterpart (see ex.2). Given these similarities between this set of numeral NPs and the FQ all, this set of quantifiers can be deemed as the FQ in Mandarin. 

(1) a. na yi-dui fuqi (liang-ge ren) yinggai (liang-ge ren) 

that one -CL husband.and.wife two-CL person may two-CL person 

hui (liang-ge ren) yiqiq wenjudian caigou 

will two-CL person together stationery shop 

‘The husband and his wife may go shopping in the stationery together.’ 

b. (All) The carpet will (all) have (all) been (all) being dusted for two hours.

(2) a. na yi-dui fuqi (liang-ge ren) a ! yinggai hui yiqiq 

that one -CL husband.and.wife two-CL person EXCL may will together 

wenjudian caigou 

stationery shop 

‘The husband and his wife may go shopping in the stationery together.’ 

b. All the carpet will have been being dusted for two hours. 

This study proposes that the FQ in Mandarin, which is called Adverbial quantifier (AVQ) should be accounted for by Adverbial Analysis (Bobaljik 1995, Doetjes 1997, Fitzpatrick 2006) instead of Stranding Analysis (Sportiche 1988, Cirillo 2009), and the non-floating quantifier in (2), which is named Adnominal quantifier (ADQ) syntactically behaves like appositive, which means it adjoins a nominal phrase. Specifically, the sentence with an AVQ and the sentence with an ADQ do not have a derivational relationship. There are some arguments for this proposal. First, under Adverbial Analysis, some unexpected sentences can be rules out, as shown in (3). 

(3) *Na-dui fuqii lai-le [ ti liang-ge ren] 

that-CL husband.wife come two-CL person 

Lit:‘That husband and his wife come.’ 

Under Stranding Analysis, this sentence is possible, but under Adverbial Analysis, it is impossible for an adverb to occur in the end of a sentence in Mandarin.

Furthermore, the sentence with an ADQ can only have exhaustive reading but the sentence with an AVQ can have both exhaustive and partitive interpretations, as shown in (4). That means the ADQ and the AVQ underlyingly not the same. 

(4)a. Na-xie xuesheng yiding san-ge ren nenggou yiqi 

that-CL student definitely three-CL person be.able.able together 

zai zhe-chang bisai dejiang 

in this-CL game win.a.prize 

‘Three of these students (The three students) are able to win a prize in this game.’ 

b. Na-xie xuesheng san-ge ren a! yiding nenggou 

that-CL student three-CL person EXCL!definitely be.able.able 

yiqi zai zhe-chang bisai dejiang 

together in this-CL game win.a.prize 

‘The three students are able to win a prize in this game.’ 

The purpose of this study is to thoroughly discuss the AVQ and the ADQ, and provide a proper account.