The 34th Annual Northwest Linguistics Conference

Abstract Booklet

April 28-29, 2018 Simon Fraser University Harbour Centre Campus Vancouver, BC

Located on the traditional ancestral territories of Tsleil-Waututh, Sḵwx̱ wú7mesh (Squamish), and xʷməθkʷəy̓ əm (Musqueam) peoples

1 Northwest Linguistics Conference 34 Program

Saturday, April 28, 2018 HC 2270 HC 1510 8:30-9:30 Registration 9:30-10:00 Welcome Aaron Williams Sḵwxwú7mesẖ (Squamish) Nation 10:00-11:00 Plenary Talk I Phonetics and the Public Good Dr. Murray Munro (Professor, Simon Fraser University) Chair: Noortje de Weers 11:00-11:15 Coffee Break Session I: Phonetics Session I: I Chair: Yue Chen Chair: Quince Sholberg 11:15-11:45 Filipinos in Toronto are speak[ɪng] like this, Might should we consider this?: Double modal not talk[ɪn] like that inversion in Southern English Pocholo Umbal, Jean-Francois Juneau, Clara Sara Williamson Dubber, and Anthony Cheung (Simon Fraser University) (University of Toronto) 11:45-12:15 The entrainment of creaky voice in a Constructional Constraints in English Syntactic conversation corpus Amalgams: A Corpus-based Perspective Courtney Mansfield Rok Sim and Jong-Bok Kim (University of ) (Kyung Hee University) 12:15-1:45 Lunch (not provided) 1:45-2:45 Plenary Talk II Investigating the use of articulatory information in auditory speech perception: How listening is shaped by talking Dr. Henny Yeung (Assistant Professor, Simon Fraser University) Chair: Quince Sholberg 2:45-3:00 Coffee Break Session II: Language Acquisition I Session II: First Nations Languages Chair: Olga Vasileva Chair: Kelli Finney 3:00-3:30 The perception and production of intervocalic Pragmatics meets Prosodics: Nominal Lists in [θ] by Mandarin ESL learners Hul'q'umi'num' Salish Junyu Wu and Hua Lin Zachary Gilkison (University of Victoria) (Simon Fraser University) 3:30-4:00 Korean Learners' Individual Networks of Plain velars in the Sḵwx̱wú7mesh language Practice Ethan Pincott Kellianne Bennett (Simon Fraser University) (University of Washington) 4:00-4:30 Fluency in the EFL Chilean Classroom The (Predictable) Effect of Environment on Astrid Morrison Quality in ʔayʔaǰuθəm (Universidad Autonoma de Chile) Gloria Mellesmoen and Marianne Huijsmans (University of ) 5:00-7:00 Social (drinks and appetizers at Malone’s)

Sunday, April 29, 2018 HC 2270 HC 1510 9:30-10:00 Registration 10:00-11:00 Plenary Talk III Internally Headed Relatives: From Japanese to the Dene Languages Dr. Keir Moulton (Assistant Professor, Simon Fraser University) Chair: Anisa Dhanji 11:00-11:15 Coffee Break Session III: Syntax II Session III: Sociolinguistics Chair: Sara Williamson Chair: Zack Gilkison 11:15-11:45 Gender Identity and the Naturalness of First Nations Language Revitalization in Referential Singular They British Columbia: The case of the Tŝilhqot’in Trevor Block language in Yuneŝit’in (Simon Fraser University) Paula Laita Pallares (University of the Basque Country UPV/EHU) 11:45-12:15 Reflector: another potential interpretation of Language Contact in Amdo and Kham the Mandarin reflexive ziji Nathan Loggins Peng Han (University of Washington) (University of Calgary) 12:15-1:45 Lunch (not provided) Session IV: Language Acquisition II Chair: Dasha Gluhareva 1:45-2:15 Investigating Language Asymmetry and Handedness in Participants with Diverse Language Background Olga Vasileva (Simon Fraser University) 2:15-2:45 Instructed SLA of English articles and noun types by L1 Chinese international students Dakota Thomas-Wilhelm (University of Iowa/Universitat Autònoma de Barcelona) 2:45-3:15 Investigating Acquisition of L2 Articles: Evidence from Japanese-speaking and English-speaking learners of Standard Albandary Aldossari (Western University)

Plenary Talks *All plenary talks will be in HC 2270*

Saturday, April 28 Plenary #1: 10:00-11:00 am Phonetics and the Public Good Dr. Murray Munro (Professor, Simon Fraser University)

In its strict sense “Applied Linguistics” refers to a field of study that places value on the practical uses of linguistic research. Of course, one application of such work is in language teaching, which benefitted both in the early 20th century and again more recently from a close engagement with phonetics. In addition, the influence of the speech sciences in the public sphere has expanded in the past few decades, with consequences for accessibility, the arts, business, and even criminal prosecution. Using examples drawn from my own empirical work and that of other researchers I will discuss the ways in which phonetics can shed light on several current issues relating to education, public policy, and dishonest practices.

Plenary #2: 1:45-2:45pm Investigating the use of articulatory information in auditory speech perception: How listening is shaped by talking Dr. Henny Yeung (Assistant Professor, Simon Fraser University)

Phonetics and phonology assume similar representations in production and perception, yet we still understand relatively little about how cognitive and neural systems for speech perception and production are linked. Here I examine the role that sensorimotor information about production has in speech perception, and how this sensorimotor information is formatted. The first part of this talk will focus on production information as supplementary source of information that is useful for the resolution of ambiguous auditory speech, just like seeing a visual face aids with understanding speech. However, I will also show that visual and sensorimotor information tap into different cognitive representations of speech. The second part of this talk will explore the format of sensorimotor information, and results will indicate that low level articulatory cues are incorporated into auditory perception. Together, these findings have several implications for theories of gestural phonetics and phonology.

Sunday, April 29 Plenary #3: 10:00-11:00 am Internally Headed Relatives: From Japanese to the Dene Languages Dr. Keir Moulton (Assistant Professor, Simon Fraser University)

Relativization strategies in the world’s languages vary in a number of ways. One prominent source of variation is the location of the head noun of the relative. Languages like English place the head externally to the relative clause (EHRCs). In contrast, a range of (unrelated) languages place the head internal to the relative clause—so called internally headed relatives (IHRCs). A major question in the study of IHRCs is whether they have an underlying or abstract syntactic structure that is similar to EHRCs. Erlewine and Gould (2014) have recently argued for a unification of Japanese EHRCs and IHRCs; Platero 1974 and Grosu 2012 have claimed that IHRCs in the Dene language Navajo are also, at an abstract syntactic level, externally headed. In this talk I report joint work with Elizabeth Bogal-Allbritten that argues against Grosu’s claim about Navajo based on novel fieldwork elicitation; I then report joint work with Junko Shimoyama refuting Erlewine and Gould's analysis. I present an account of the variation in IHRCs that does not cast them as derivationally related to externally headed relatives. Filipinos in Toronto are speak[ɪng] like this, not talk[ɪn] like that Pocholo Umbal, Jean-Francois Juneau, Clara Dubber, and Anthony Cheung University of Toronto The variable realization of the English ending –ing (walk[ɪn] vs. walk[ɪng]), known also as the variable (ING), has been studied extensively within the variationist framework (e.g., Fisher 1958, Houston 1985, Labov 1966, Trudgill 1972, 1974). However, studies on the ethnic patterns of (ING) remain underwhelming (e.g., Drummond 2012, Walker 2012, 2015). Walker (2012), for instance, showed that Chinese speakers of in Toronto show low rates of [ɪn] use. In addition, Drummond (2012) found that for Polish immigrants in Manchester, identity played a role in their choice of (ING), such that those aligned with Polish identities favour [ɪŋk]. It is in this intersection of ethnicity and identity that this study is situated. This study explores the patterns of (ING) among Filipinos in Toronto, an important but understudied speech community. Focusing on the use of [ɪn], we examine the role of linguistic and social factors in the patterning of (ING), and also investigate the possible influence of ethnic orientation (perceived degree of ethnic group affiliation; Hoffman & Walker 2010:46) on the pattern of variability.

Spontaneous speech data from 8 second-generation and 1 first-generation Filipino immigrants (stratified by sex and age) were taken from Hoffman and Walker’s (2010) corpus of Toronto English. In ELAN (2017), 728 tokens of (ING) in word-final unstressed positions were extracted and coded impressionistically as either [ɪn] or [ɪng]. Previous studies have demonstrated that more verbal categories and two-syllable words favour the use of [ɪn] (Tagliamonte 2004); the same applies to males, members of the lower classes, and in casual speech (Fisher 1958, Trudgill 1972). (ING) may also be sensitive to age (Horvath 1985). As such, we also coded for the following factors: PART OF SPEECH, SYLLABLE, SEX and AGE. To operationalize ETHNIC ORIENTATION (EO), each speaker response to the Ethnic Orientation Questionnaire (Hoffman & Walker 2010) was assigned a score: 1 corresponded to low degree of ethnic affiliation; a 3 indicated high degree of participation; and 2 was in-between. Apart from computing each speaker’s overall mean EO score (EO_ALL), mean scores for questions relating only to social network (EO_SN), ethnic identity (EO_ETHNICID), language use (EO_LG), and language choice (EO_LGCH) were calculated. Following previous findings, we predicted the same patterns, and that those with higher EO scores will also favour the nonstandard form [ɪn].

We conducted a mixed-effect regression analysis in Rbrul (Johnson 2009) using the linguistic and social factors as fixed effects and SPEAKER as a random effect. In line with previous studies, our results revealed significant effects of SYLLABLE NUMBER, AGE, SEX, and EO_SN: words with more than two syllables, younger speakers and males in general favour the use of [ɪn]. PART OF SPEECH did not come out as significant. Furthermore, those with a more Filipino social network also favour [ɪn]. These results demonstrate that Filipinos also show similar linguistic and social conditioning as compared to other groups, but perhaps (ING) is driven more by social factors. Finally, this study provides evidence that differences in how speakers construct ethnic identities may affect the degree to which sociolinguistic variants are used.

References

Drummond, Rob. 2012. Aspects of identity in a second language: ING variation in the speech of Polish migrants living in Manchester, UK. Language Variation and Change 24.107-33.

5 ELAN [computer software]. 2017. Nijmegen: Max Planck Institute for Psycholinguistics. Retrieved from https://tla.mpi.nl/tools/tla-tools/elan/. Fisher, John L. 1958. Social influence on the choice of a linguistic variant. Word 14.47-56. Hoffman, Michol F. and James A. Walker. 2010. Ethnolects in the city: Ethnic orientation and linguistic variation in Toronto English. Language Variation and Change 22.37-67. Horvath, Barbara M. 1985. Variation in Australian English. Cambridge: Cambridge University Press. Houston, Ann Celeste. 1985. Continuity and change in English : The variable (ING). Unpublished PhD Dissertation. University of Pennsylvania. Johnson, Daniel Ezra. 2009. Getting off the GoldVarb standard: Introducing Rbrul for mixed- effects variable rule analysis. Language and Linguistics Compass 3.359-83. Labov, William. 1966. The social stratification of English in New York City. Washington, DC: Center for Applied Linguistics. Tagliamonte, Sali A. 2004. Someth[n]’s goi[g] on! Variable (ING) at ground zero. Language Variation in Europe: Papers from the Second International Conference on Language Variation in Europe, ICLaVE 2, ed. by Britt-Louise Gunnarsson, Lena Bergström, Gerd Eklund, Staffan Fidell, Lise H. Hansen, Angela Karstadt, Bengt Nordberg, Eva Sundergren and Mats Thelander, 390-403. Uppsala: Uppsala University Department of Scandinavian Languages. Trudgill, Peter. 1972. Sex, covert prestige, and linguistic change in urban British English. Language and Society 1.179-195. Trudgill, Peter. 1974. The social differentiation of English in Norwich. Cambridge: Cambridge University Press. Walker, James A. 2012. Velar nasals in Toronto English. Paper presented at New Ways of Analyzing Variation (NWAV) 41, Indiana University. Walker, James A. 2015. Canadian English: A sociolinguistic perspective. New York: Routledge.

6 The entrainment of creaky voice in a conversational corpus Courtney Mansfield University of Washington

Speakers naturally adapt their speaking style to that of their interlocutor to promote pro-ficient communication and to maintain a positive social identity. This process is described as linguistic entrainment or accommodation in the literature and is often grounded in Com-munication Accommodation Theory (Giles, 2007), which suggests that speakers converge and diverge to their partner’s speech features to signal social identity and affect. Lexical items (Brennan, 1996) and phonetic features such as intensity have been shown to entrain in cooperative conversation while other features such as pitch are idiosyncratic in speakers in some contexts (Levitan and Hirschberg, 2011). The entrainment of has not been examined and the question of whether speak- ers adapt this feature towards their interlocutor remains unanswered. Creaky voice has been shown to index social meaning in some populations (Henton, 1986) and signify traits such as ‘intimacy’ or ‘friendliness’ in perceptual studies (Gobl and Chasaide, 2003). I hypothesize that creaky voice will entrain in a cooperative dialog, supported by its status as a social marker. Previous literature suggests that creak is utilized more often in female speakers (Abdelli-Beruh, Wolk, and Slavin, 2014), and further studies suggest that women entrain to a higher degree than men in conversation (Namy, Nygaard, and Sauerteig, 2002). For these reasons, I hypothesize that women will both use proportionally higher amounts of creak and adapt to their partners more than men. Entrainment studies require a significant amount of data, but manual annotations are resource-intensive. While automatic approaches to creak classification allow for the explo- ration of larger datasets, phonation classifiers are relatively new and have not been tested widely on a diverse range of datasets. Therefore, I approach the problem using a combination of manual annotations of phonation from Panfili (2016) and automated creak classifications from the state-of-the-art Recursive Neural Network of Drugman, Kane, and Gobl (2012). I use the ATAROS corpus, which includes Pacific Northwest speakers engaged in cooperative tasks designed to elicit various levels of engagement and stance-taking (Freeman et al., 2014). Posterier probabilities of creak are automatically extracted over a set of 16 dyads, where a subset of 11 dyads contain both automatic and human-annotated creak measures. Two di- mensions of entrainment are considered, following the methodologies of Levitan and Hirschberg (2011); these are referred to as proximity and convergence measures. Proximity is a measure of the similarity of speech patterns of a speaker to an interlocuter across an entire conver-sation. Proximity is expressed as the difference between the negative absolute difference of the speaker and interlocuter’s creak rate versus the negative absolute difference of the mean difference between speaker and all other participants. Convergence is indicated through a comparison of the mean absolute difference of the speaker and interlocuter’s creak rate at an early and late window in the conversation. Dyads are grouped by speaker and interlocuter gender, and four groups are examined: male speakers with male partners, male speakers with female partners, female speakers with female partners, and female speakers with male partners. Men showed significant proximity with female partners (p<0.05) and male partners (p<0.01) on both system and human-annotated creak rates. Women showed significant proximity with male partners on the human-annotated subset only (p<0.05). Convergence across the conversation (measured over the first and last 160-second segment of conversation) did not show significant effects

7 and was extremely variable across dyads. Additionally, in the case of male speakers, partner gender was found to be significantly correlated with creak rate (p<0.01). Men with female partners were on average creakier than men with male partners, with average creak rates of 18.7% and 6.8% respectively. Overall average creak rates were higher for women than men, at 22.7% and 14% respectively. This study demonstrates linguistic entrainment of creak as shown by the significant prox- imity of speakers across most groups. Interestingly, my findings diverge from previous lit- erature which suggests that women entrain at higher rates than men (Namy, Nygaard, and Sauerteig, 2002). While convergence or divergence of speakers was not significant at the group level, the timeline for convergence is not well understood, and it is possible that the duration of the creak rate measured does not capture the reality of speakers who may accom- modate very rapidly or over a longer window of time. While main effects for entrainment were found, proximity and convergence varied across individuals. Understanding the variation of entrainment and the linguistic features involved in the entrainment process has important implications on both understanding natural conversation and designing automatic dialog systems, where a one-size-fits-all approach to modeling voice quality or speaker entrainment would not capture the subtleties of social context.

References

Abdelli-Beruh, Nassima B, Lesley Wolk, and Dianne Slavin (2014). “Prevalence of vocal fry in young adult male American English speakers”. In: Journal of Voice 28.2, pp. 185–190. Brennan, Susan E (1996). “Lexical entrainment in spontaneous dialog”. In: Proceedings of ISSD 96, pp. 41–44. Drugman, Thomas, John Kane, and Christer Gobl (2012). “Resonator-based creaky voice detection”. In: Thirteenth Annual Conference of the International Speech Communication Association. Freeman, V. et al. (2014). ATAROS Technical Report 1: Corpus collection and initial task validation. Tech. rep. Linguistics Phonetics Lab, University of Washington, Department of Linguistics. Giles, Howard (2007). Communication accommodation theory. Wiley Online Library. Gobl, Christer and N´ıAilbhe Chasaide (2003). “The role of voice quality in communicating emotion, mood and attitude”. In: Speech communication 40.1, pp. 189–212. Henton, Caroline G (1986). “Creak as a sociophonetic marker”. In: The Journal of the Acoustical Society of America 80.S1, S50–S50. Levitan, Rivka and Julia Hirschberg (2011). “Measuring acoustic-prosodic entrainment with respect to multiple levels and dimensions”. In: Twelfth Annual Conference of the Inter- national Speech Communication Association. Namy, Laura L, Lynne C Nygaard, and Denise Sauerteig (2002). “Gender differences in vocal accommodation: The role of perception”. In: Journal of Language and Social Psychology 21.4, pp. 422–432. Panfili, L. (2016). “The Physiological Underpinnings of Vowel Height and Creaky Voice”. In: The Acoustical Society of America. Salt Lake City, UT.

8 Might should we consider this?: Double modal inversion in Southern United States English Sara Williamson Simon Fraser University

Introduction. Southern United States English (SUSE) is among the dialects of English that permit a double modal construction, e.g. might could or might should. While elicitation studies have tested double modal inversion in questions in SUSE (Di Paolo et al., 1979; Di Paolo, 1986; Hasty, 2012), the experimental results are mixed. Di Paolo et al. (1979) found joint double modal inversion, as in (1a), to be acceptable in Texas English, whereas Di Paolo (1986) found it to be marginal as compared to inversion of the second modal, as in (1c). Hasty (2012) suggested that only second-modal inversion is possible in Tennessee English. These studies used differing meth- ods in different varieties of SUSE, impeding a cross-study comparison. This work addresses two questions: (i) is joint inversion of double modals acceptable in SUSE? And (ii) are there regional differences in inversion patterns within SUSE? Results from an experiment conducted with speak- ers of Kentucky, Tennessee, and Texas Englishes show that joint inversion is preferred in some varieties of SUSE. However, at least in Tennessee, second-modal inversion is also viable. The experiment. 108 native English speakers whose local dialects permitted double modals participated in a web-based acceptability judgment task. 44 of the participants were born and raised, at least to age 12, in Kentucky; 27 were born and raised in Tennessee; and 37 were born and raised in Texas. The task tested the relative acceptability of three levels of INVERTED MODAL (joint inversion (1a), first-modal inversion (1b), and second-modal inversion (1c)) in each of the three levels of REGION for the common double modals might could and might should (Hasty, 2011). All target items (32 test item sets, 32 fillers) were given as informal dialogues to encourage participants to give accurate judgments for non-standard constructions (Henry, 2005). (1) Billy is at the park with his mother. He says to her: a. “Might could I play on the monkey bars?” b. “Might I could play on the monkey bars?” c. “Could I might play on the monkey bars?”

Raw ratings (1–7) were z-score transformed prior to analysis. A main effect of INVERTED MODAL and an interaction between INVERTED MODAL and REGION (p < .01) were found (Fig. 1). Joint inversion was rated significantly higher than first-modal inversion overall (mean −.47 vs. −.62, p < .001), as well as in Kentucky (−.39 vs. −.55, p < .001), Tennessee (−.55 vs. −.65, p < .05), and Texas (−.51 vs. −.68, p < .001). Since Di Paolo (1986) and Hasty (2012) consistently found first-modal inversion to be unacceptable in SUSE, these results suggest that joint inversion, by con- trast, is at least marginally acceptable in SUSE. Joint inversion was also rated significantly higher than second-modal inversion in Kentucky (−.39 vs. −.65, p < .001) and in Texas (−.51 vs. −.67, p < .001). However, no significant difference between these conditions was found in Tennessee (−.55 vs. −.59, p = .44), suggesting that both forms of inversion are acceptable in this region. Implications. These results provide a means of evaluating theoretical approaches to the SUSE double modal construction. Previous analyses predict either the availability of joint inversion (e.g. Di Paolo, 1989) or the availability of second-modal inversion (e.g. Battistella, 1995; Close, 2004; Hasty, 2012), but cannot readily account for variable patterns. Yet these results show that while

9 joint inversion is at least marginally acceptable in some varieties of SUSE, second-modal inversion is also acceptable, at least in Tennessee English. Therefore, an adequate theoretical analysis of the SUSE double modal construction must not only allow joint inversion in questions, but must also account for regional variations in the accepted patterns of inversion.

Figure 1: Mean (raw) acceptability rating of test conditions

Selected References Battistella, Edwin. 1995. The syntax of the double modal construction. Linguistica Atlantica 17:19–44. Close, Joanne. 2004. English auxiliaries: A syntactic study of contraction and variation. Doctoral Dissertation, University of York. Di Paolo, Marianna. 1986. A study of double modals in Texas English. Doctoral Dissertation, University of Texas at Austin. Di Paolo, Marianna. 1989. Double modals as single lexical items. American Speech 64:195–224. Di Paolo, Marianna, Charles McClenon, and Kenneth Ranson. 1979. A survey of double modals in Texas. Texas Linguistic Forum 13:40–49. Hasty, J. Daniel. 2011. I might not would say that: A sociolinguistic investigation of double modal acceptance. University of Pennsylvania Working Papers in Linguistics 17:91–98. Hasty, J. Daniel. 2012. This might could help us better understand syntactic variation: The double modal construction in Tennessee English. Doctoral Dissertation, Michigan State University. Henry, Alison. 2005. Non-standard dialects and linguistic data. Lingua 115:1599–1617.

10 Constructional Constraints in English Syntactic Amalgams: A Corpus-based Perspective Rok Sim and Jong-Bok Kim Kyung Hee University

The so-called cleft-syntactic amalgamation construction (Cleft-Syn-Amal) involves two independent clauses (main clause (MC) and interrupting amalgam clause (AC)), while sharing a constituent named the ‘content kernel’.   (1) [ He gives a speech in [ I think it was Florida.]] MC AC O O

The ‘content kernel’ Florida in (1) functions as both the object of the preposition in as well as the predicate of the verb was. The construction can occur in various syntactic positions, as illustrated by the COCA (Corpus of Contemporary American) examples:

(2) a. Verb object: This president has answered [I think it’s like 4,200 questions]. (COCA: 1992 SPOK)

b. Prepositional object: On [I believe it was Thursday morning], less than ninety-six hours after the event I actually saw a building going back up. (COCA: 2011 SPOK)

c. Predicational: And the ingredients are, [I thought it was Cachaca], but no it’s Cachaca. (COCA: 2015 SPOK)

d. Adjunct: It almost sank [I believe it was Monday or Tuesday of this week]. (COCA: 1990 SPOK)

For example, the ‘content kernel’ 4,200 questions in (2a) functions as both the object of the verb answered as well as predicate of the verb is. Meanwhile, the content kernel in the other examples function as the object of a preposition, a predicative complement, and an adjunct. The construction displays many idiosyncracies challenging any form-function matching account. In terms of meaning, (2a) means that the president has answered ‘x’, and the value of ‘x’ is, at least to the speaker’s knowledge, 4,200 questions, and also accompanies a ‘hedge’ reading (Lakoff 1974, Kluck 2009). Most of the existing analyses adopt Lakoff’s (1974) deletion-based account (Kluck 2011b). For example, (1) is derived from the combination of two clauses, ellipsis of the empty element, and then elide the clause:

(3) [He gives a speech in [Florida]] [I think it was Florida that He gives a speech in.] (NP-ellipsis and clausal-ellipsis)

Such a deletion-based analysis leaves many peculiar properties of the construction unaccounted for (Tsub- omoto and Whitman 2000, Grosu 2006). For example, the Lakoff-style analysis requires the mandatory application of the ellipsis as seen from He gives a speech in I think it was Florida (*that he gives a speech in.), and needs to limit the type of possible subject as well as introducing verb. In addressing the unusual mapping between form and function in the construction, this paper first tries to investigate the authentic uses of the construction, by performing an extensive investigation of the corpus COCA. Some of the corpus findings we have found indicate that the dominant subject of the AC is the pronoun I. The verb type introducing the AC is also limited to non-factive verbs like think, believe, guess to allow a hedge reading. Based on this kind of corpus search, we sketch a base-generation approach, couched upon the framework of Construction Grammar (CxG) where constructions, pairings of form with meaning, are the basic units of language and linked as form or inheritance hierarchies (network) (see Goldberg 2006

11 and references therein). Our supposition is that English employs a special clausal-level construction where the content kernel is in-situ (e.g., Thursday morning in (2b), Cachaca in (2c), and Monday or Tuesday of this week in (2d)) and introduced by a non-factive embedding verb. The present analysis factors out generalizations of Cleft-Syn-Amal constructions. This factorization allows us to capture generalizations as well as idiosyncracies about English Cleft-Syn-Amal construction in a systematic way.

Selected References

Goldberg, Adele. 2006. Constructions at work: Constructionist approaches in context. New York: Oxford University Press. Grosu, Alexander. 2006. An amalgam and its puzzles. In Hans-Martin Gartner, Sigrid Beck, Regine Eckhardt, Renate Musan, and Barbara Stiebels (eds.), Between 40 and 60 puzzles for Krifka. Kluck, Marlies. 2009. Intertwined clauses, interacting propositions: A note on the interpretive aspects of sentence amalgamation. Conference of the Student Organisation of Linguistics in Europe (ConSOLE) XVI : 77-101. Kluck, Marlies. 2011b. Sentence amalgamation. Ph.D dissertation, University of Groningen. LOT Disser- tation Series 285. (Landelijke Onderzoekschool Taalwetenschap, the Netherlands National Graduate School of Linguistics.) Lakoff, George. 1974. Syntactic amalgams. Berkeley Studies in Syntax and Semantics 1(9): 1-24. Tsubomoto, Atsuro and John Whitman. 2000. A type of head-in-situ construction in English. Linguistic Inquiry 31, 176-182.

12 The perception and production of intervocalic [θ] by Mandarin ESL learners Junyu Wu and Hua Lin University of Victoria [email protected] [email protected] English voiceless interdental [θ] in words such as thank and method is among the most difficult for ESL learners to acquire. This has been demonstrated in many studies with participants from a variety of first language backgrounds including German, Dutch, French, Polish, Arabic, Persian, Chinese, Korean, and Japanese (Brannen, 2011; Hanulíková & Weber, 2010; Rau, Chang & Tarone, 2009; Gonet and Pietroń, 2010; Schmidt, 1987; Dubois & Horvath, 1999). Even highly proficient ESL speakers have been found to regularly substitute [θ] with other sounds which are, most often, [s], [t], or [f] (Brannen, 2011; Hanulíková & Weber, 2010). Preference for one or another of the three substitutions seems to vary from one L1 to another and sometimes, from one speaker to another of the same L1. ESL learners of L1 Mandarin, for instance, have been found to strongly favor [s] as a substitute in word-initial or word-final position (Rau, Chang & Tarone, 2009), while L1 speakers of Hong Kong English opt for [f] as a substitute (Peust, 1996). One gap in the literature on L1 Mandarin learners is [θ] in word-medial, intervocalic position--all previous studies have focused on the onset or coda position. This research aims to find out if the intervocalic environment plays a role in the substitution by examining the production and perception of English voiceless interdental in word-medial intervocalic position by L1 Mandarin speakers. An experiment was conducted. Three adult Mandarin ESL learners were recruited. Production and perception data were collected via an identification type of perception task and two word-reading production tasks, respectively. The production task recorded the L2 participants reading six real words with [θ] in word-medial position (e.g., method [ˈmɛθəd] and nothing [ˈnʌθɪŋ]) both in isolation and in a carrier sentence. The perception task has 24 words including the above-mentioned six real words and 18 pseudo words constructed with the real words but having the voiceless interdental replaced by [s], [f], or [ð] (e.g., [ˈmɛsəd] or [ˈnʌfɪŋ]). The pseudo words are meant to be distractors. As well we are curious to know if the participants can tell that [θ] is not in these non-words The results show that the production accuracy rate at 91% is high. Four substitutions are found among the 9% errors: [s], [ð], [θs], [fθ]. The latter two substitutions are surprises since to our knowledge they have never been reported in the literature on this matter. To be sure that our

13 auditory judgement is correct, a measurement of the center of gravity in five locations of each of the surprises is taken, which confirms the auditory impression. The overall perception accuracy rate at 96% is also high, and the errors constitute two substitutions [f] and [ð]. The vowel environment also seems to play a role in the perception. A prior low-mid and a following high- (sample word: nothing) and a prior high-back vowel and a following high-mid front vowel (sample word: toothache), for instance, yield different results; the accuracy rates are 33% and 67%, respectively. It was also interesting to see that perception of the three distractors all incorrectly contain [θ]. Finally, our study reveals no clear relationship between the L2 production and perception of the intervocalic [θ].

Reference Brannen, Kathleen J. 2011. The perception and production of interdental in second language acquisition. McGill University Press. Dubois, S., & Horvath, B. 1999. Let’s tink about dat: inter-dental in Cajun English. Language Variation and Change, 10, 245–261. Hanulikova,Adriana & Weber,Andrea. 2010. Production of English interdental fricatives by Dutch, German, and English speakers. Proceedings of the 6th International Symposium on the Acquisition of Second Language Speech, New Sounds 2010, Poland, 1-3 May 2010. pp. 173-178. Gonet, Wiktor & Pietron, Grzegorz. 2010. English interdental fricatives in the speech of Polish learners of English. A 2003-2006 KBN Project “Nauczanie fonetyki jezyka angielskiego w Polsce– diagnoza i terapia”. Peust, C. 1996. Sum: th-substitution. The Linguist List 7.1108. Available from http://linguistlist.org/issues/7/7-1108.html Rau, D. Victoria, Chang, Hui-Huan Ann, and Tarone, Elaine E. 2009. Think or sink: Chinese learners’ acquisition of the English voiceless interdental fricative. Language Learning 59(3): 581-621. Schmidt, R. W. 1987. Sociolinguistic variation and language transfer in phonology. In G. Ioup & S. H. Weinberger (Eds.), Interlanguage phonology: The acquisition of a second language sound system. (pp. 365–377). Cambridge, MA: Newbury House.

14 Korean Learners’ Individual Networks of Practice Kellianne Bennett University of Washington

Background. Social networks in second language acquisition give an overall pictures of the learner in their many various social contexts. Therefore, investigating learners’ social networks has the potential to reveal crucial information about the learner, including how they sustain the motivation to learn, who they elect to socialize with and use the language with, how they choose to learn, how they formulate their identity as a language learner, and so on. Research on social networks in second language acquisition has mostly focused on the effectiveness of study abroad programs (e.g., Isabelli-García, 2006) and on which types of networks facilitate language learning or literacy skills (e.g., Kurata, 2004; Ferenz, 2005). Recently, Zappa-Hollman (2007) and Zappa-Hollman and Duff (2015) developed the concept of an individual network of practice (INoP), which is based in the theoretical concept of a community of practice (CoP) (Lave and Wenger, 1991) but maps all of the social connections of one learner at the center of the network, as opposed to mapping the network connections within a set group of people. The research focused on the second language socialization (e.g., Duff, 2012) of Mexican students studying at an English-speaking university in (second language context). By diagramming the social connections from the learner’s perspective and supplementing the diagram with qualitative interviewing, a complete picture of the learner’s social context emerges. Current study. The goal of the current study is to expand on the concept of INoPs. First, it introduces INoPs into the language socialization context of foreign and heritage languages (Korean; KFL and KHL). Second, material resources such as websites and textbooks are considered and mapped along with the social contacts of the learner in the INoP. Third, the ties between the learner and their social contacts and material resources are quantified; the learner reported estimates of both time spent with each social contact or material resource as well as the percentage of the target language the learner used with each social contact or material resource. Research questions. (1) How many social contacts and material resources are there in KFL/KHL learners’ INoPs? What is the nature of the learner’s ties to these resources? (Native or nonnative interlocutors, number of in-class vs. out-of-class contacts, etc.) (2) How do the social contacts and material resources in a learner’s INoP promote the KFL/KHL socialization of the learner? (3) Which social contacts or material resources do learners report as more or less helpful, either for overall language proficiency or for sustaining motivation to study Korean? What are the characteristics of those social contacts or material resources that make them more or less helpful from the learner’s perspective? Methodology and participants. 6 KFL and 7 KHL learners (all but one currently enrolled in 2nd-4th year Korean classes at the University of Washington) responded to an initial online survey and reported the social contacts and material resources in their INoPs. The five respondents with the most overall nodes in their INoPs were selected for hour-long semi-structured qualitative interviews (audio recorded and transcribed). The researcher diagrammed each interviewee’s INoP from their survey responses and discussed it with the learner during the interview. Based on the interview, the diagrams were revised and further qualitative comments on the language socialization of the learner were gathered. Results. The 13 survey respondents reported an average of 4 contacts in in-class networks and spent 1.78 hrs/wk (total) with this group. A similar number of average contacts, 3.62, were reported as out-of -class contacts but respondents spent much more time with their out-of-class

15 network contacts, an average of 4.22 hrs/wk. The natures of the learners’ overall INoP varied; groups of friends, language exchange partners, family, roommates, and coworkers all appeared in respondents’ networks. Respondents reported an average of 2.85 material resources used to study Korean or engage in Korean culture, and each respondent spent on average 8.74 hrs/wk using their material resources, which included, e.g., podcasts, news, books, TV, music, websites, etc. Of the five respondents selected for interviews, one’ respondent’s diagrammed INoP is shown below (social contacts only). Extended results, including the learners’ comments on how helpful or motivating particular contacts or groups of contacts were, are not included here due to space constraints. This study of KFL/KHL INoPs provides insight into social networks in the foreign and heritage language learning contexts by combining quantitative and qualitative results.

Selected references. Duff, P. (2012). Second language socialization. In A. Duranti, E. Ochs, & B. Schieffelin (Eds.), Handbook of language socialization (pp. 564-586). Malden, MA: Wiley-Blackwell. Ferenz, O. (2005). EFL writers’ social networks: Impact on advanced academic literacy development. Journal of English for Academic Purposes, 4, 339-51. Isabelli-García, C. (2006). Study abroad social networks, motivation, and attitudes: Implications for second language acquisition. In M.A. Dufon & E. Churchill (Eds.), Language learners in study abroad contexts (pp. 231-258). Cleveland, OH: Multilingual Matters. Kurata, N. (2004). Communication networks of Japanese language learners in their home country. Journal of Asian Pacific Communication, 14 (1), 153-179. Lave, J., and Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge, England: Cambridge University Press. Zappa-Hollman, S. (2007b). The academic literacy socialization of Mexican exchange students at a Canadian university. Unpublished doctoral dissertation, University of British Columbia, Vancouver, Canada. Zappa-Hollman, S. & Duff, P. (2015). Academic English socialization through individual networks of practice. TESOL Quarterly, 49, 2: 333-368.

16 Fluency in the EFL Chilean Classrooms Astrid Morrison Universidad Autonoma de Chile Speech fluency in second language (L2) learning has for long been considered an important factor in developing knowledge and skills in an L2 and a key construct in assessing L2 ability, as such, fluency is regarded as one of the main goals for L2 learning. When defining the concept, fluency is usually associated with the idea of “flow” and “smoothness of speech”. Fillmore (1979) provided one of the earlier definitions of fluency, describing it as “the ability to talk at length, with few pauses, the ability to fill time with talk” (Fillmore, 1979 p. 93). Given the importance of fluency in communicating with speakers of different languages in the globalised world, studying it has gained currency from research and pedagogy perspectives, and it is also highlighted in language benchmarks and policy documents, including the national curricula in different countries and the Common European Framework of Reference for Languages (CEFR) (Council of Europe,2001). However, despite this attention, fluency is neither clearly represented nor promoted in L2 teaching curricula, material or classroom practices. Recent research in this area suggests that fluency is not promoted in textbooks (Rossiter et al., 2010), not clearly known by L2 teachers, and not actively promoted in L2 classrooms (Tavakoli & Hunter, 2017). This presentation reports a research conducted as part of an MA dissertation, which aimed at investigating both teachers’ and textbooks’ role in the promotion of fluency in the EFL classrooms. On one hand, it aimed at exploring the extent to which the oral activities from EFL textbooks, used in state and semi-private schools in Chile, focus on promoting fluency. On the other hand, the study was interested in investigating Chilean teachers’ perceptions, both in terms of what fluency is, and to what extent they promote it in their classrooms. While there are a few studies investigating fluency (e.g. Rossiter et al., 2010; Tavakoli & Hunter, 2017), to the best of my knowledge there are no studies that have exanimated fluency from a multi-practice perspective. This study is therefore the first to examine fluency from two interrelated but different aspects of a lessons: teachers and textbooks. Data for the textbook analysis of the study was collected from two EFL textbooks used in state and semi-private schools in Chile. The oral activities from both textbooks were analysed using a framework for evaluating fluency-focused activities, which combined both Rossiter et al. (2010) and Tavakoli & Hunter’s (2017) frameworks. Using this framework, all speaking activities from both textbooks were analysed by the

17 researcher. To ensure reliability of the analysis, 20% of the activities were second rated by a more experienced researcher. In order to investigate teachers’ understanding of the concept and classroom practices in promoting it, self-reported questionnaire used in Tavakoli & Hunter (2017) was adapted. The new questionnaire was then used to collect data from sixty Chilean EFL teachers, investigating their understanding of fluency, fluency-promoting activities and the extent to which they feel confident about promoting it in their classrooms. The findings suggest that both teachers and textbooks prioritise general-speaking to fluency-focused activities, with little attention to and focus on promoting fluency. The results also highlight a misinterpretation of fluency-focused activities among teachers, with many considering fluency activities equal to general speaking ones. Even though teachers report high levels of confidence in their knowledge and skills for promoting fluency in classroom, they appear to have limited knowledge of what fluency is, or what activities can be used to promote it. These findings could have significant implications for classroom teaching both in terms of suitability of material and preparedness of teachers of this context. The results clearly suggest that despite the significant role fluency plays in L2 learning, teaching practices, whether textbooks or teachers, are not up to the challenge or cannot prepare learners for the fluent communication that is needed from them outside class. Finally, the lack of connection between research and teaching practice discovered in the Chilean context agreed with similar studies conducted in other countries (Rossiter et al., 2010; Tavakoli & Hunter, 2017), which suggests that this might be a global issue among language teachers.

18 References:

Council of Europe. (2001). Common European Framework of Reference for Languages: Learning, Teaching, Assessment. Council of Europe.

Fillmore, C. J. (1979). On fluency. In C. J. Fillmore, D. Kempler, & W. S. Y. Wang (Eds.), Individual differences in language ability and language behavior (pp. 85-102). New York: Academic Press.

Kormos, J., & Dénes, M. (2004). Exploring measures and perceptions of fluency in the speech of second language learners. System, 32(2), 145-164. http://dx.doi.org/10.1016/j.system.2004.01.001

Rossiter, M., Derwing, T., Manimtim, L., & Thomson, R. (2010). Oral Fluency: The Neglected Component in the Communicative Language Classroom. Canadian Modern Language Review, 66(4), 583-606. http://dx.doi.org/10.3138/cmlr.66.4.583

Tavakoli, P., & Hunter, A. (2017). Is fluency being ‘neglected’ in the classroom? Teacher understanding of fluency and related classroom practices. Language Teaching Research, http://dx.doi.org/10.1177/1362168817708462

19 Pragmatics meets Prosodics: Nominal Lists in Hul’q’umi’num’ Salish Zachary Gilkison Simon Fraser University (ISO code: hur) is a Salish language, of the Central Salish branch, spoken as a first language by around 40 elders in southwestern British Columbia, along the shores of the Salish sea. Data for this research are from the Island dialect — Hul’q’umi’num’— drawn from our text corpus of transcriptions and translations of recordings of over twenty different speakers and dating back to 1962. This is part of a larger project studying the pragmatics and prosodics of performance of stories, with the goal of understanding the devices Elders use to convey information and add interest to their narratives. In particular, this paper focuses on the use of lists of NPs as a vehicle for expressing and categorizing nominals based on 100 examples gleaned from the corpus. Lists in argument position are quite rare in Hul’q’umi’num’ but lists appearing outside of the clausal syntax, for example as an elaboration in sentence-final position (1) or sandwiched between the clause and a summation (2) or addition comment (3), are common. (1) m;˚ø.stem s÷i÷®t;∫-s— sm;y;ƒ, køewe÷;c, spe÷;ƒ. all.what food-3POS deer elk bear They have everything to eat—deer, elk, bear. (ST)

(2) m;˚ø.÷;¬.stem— s˙u:m, ÷ap;l, pes— m;˚ø.stem ni÷ ÷;Σ all.what— berries, apples, pears— all.what AUX LNK ÷a¬;≈t;t. gather-PROG Everything—berries, apples, pears—we picked them all. (EW)

(3)... hay ÷;Σ yaƒ ÷;¬ ÷;Σ s÷i÷®t;∫-s— very LNK always QLF LNK N.eating-3POS ... there’s really always food—

©; s÷a≈øa÷, s√;¬a÷;m ÷i÷ ©; swe:m— ÷;Σ ÷;¥-s. DT butter.clam cockles CONJ DT horse.clam LNK good-3POS butter clams, cockles, and the horseclams—they’re good. (ST)

What do lists sound like? Pitch —not pauses— is what sets off lists from the clauses to which they are attached. List intonation (we will play some examples) is quite distinctive. Pitch is reset for each NP; this contrasts with normal clausal pitch, which shows declination throughout, including in questions. The length of the pauses before an item reflects the cognitive load of remembering it.

What’s on the list? Lists are infrequently closed, where the included NPs denote the set, but usually lists are open. Open and closed lists differ prosodically (cf. Couper-Kuhlen (1986), Selting (2007), according to the pitch at the end of each NP, especially the upturn on the last NP on an open list. Including an NP as an explicit entry on a list conveys two concepts: the NP is intended to be a member of a set and it is also representative or an exemplar of a larger set, categorizing the . Lists play an important semantic role in a language like Hul’q’umi’num’ that lacks higher-level generic terms (‘animals’, ‘mammal’, ‘seafood’, ‘fruit’).

20 Sometimes a speaker comes back to a list and adds more items if they feel that more examples are needed to convey the scope of the concept underlying the list. Even lists with closed intonation can allow additional entries added afterwards, i.e. presuppositions are easily cancellable (cf. Matthewson (2006) and Davis, Matthewson, & Shank (2004)).

What’s the order of the NPs on the list? The order of the list reflects the semantic salience of the item, i.e. as an NP that pops into the mind of the speaker it is added to the list. Thus, lists give us a look into the cognitive and cultural world of the speaker. The best exemplars—most common items or favorite items— come before more unusual items. In (1-3) deer, berries, and butter clams are “keystone” species; bear, pears, and horseclams are treasured rarities.

Are lists like other supplemental information? We see some similarity with supplemental appositives (Potts 2004), i.e. COMMA intonation and Hul’q’umi’num’ list intonation. Appositives are seen as supplemental updates that are anaphorically connected to a head noun (Martin 2016). The sentence lacks some information without the appositive, but is truth-conditionally complete. In our cases, the argument NP 'priming' the list is usually broad, e.g. m;˚ø÷ ;¬ stem ‘everything’ or s÷i÷®t;∫-s ‘what is eaten’, so cannot be evaluated without a list.

Why are lists so common in Hul’q’umi’num’? Hul’q’umi’num’ is claimed to have a low referential density (Gerdts and Hukari 2008), that is NP arguments are often omitted. Expressing NPs in lists is an efficient way of conveying nominal information while keeping the load off of the argument structure. A description of a language will need to address not only clausal syntax but also have NPs in non-arguments are constructed and how they relate to the clausal syntax. Prosodic information allows the speaker to toggle in between the clause syntax and list. We see that lists are not “extra” information, but convey important insights into Hul’q’umi’num’ categorization and cognition. Carefully placed and constructed lists are an important way that Elders convey world view.

Couper-Kuhlen, E. (1986). An introduction to English prosody. Tübingen; London: Max Niemeyer : Edward Arnold. Davis, H., Matthewson, L., & Shank, S. (2004). Clefts vs. nominal predicates in two Salish languages. In Studes in Salish Linguistics in Honor of M. Dale Kinkade. Missoula, Montana: University of Montana Press. Gerdts, D. B. & Hukari, T. E. (2008). The expression of noun phrases in Halkomelem texts. Anthropological Lingusitics 50 (3/4), 1–41. Martin, S. (2016). Supplemental update. Semantics and Pragmatics, 9(0), 5-1–61. Matthewson, L. (2006). Presuppositions and Cross-Linguistic Variation. In Proceedings of NELS 26. Potts, C. (2004). The Logic of Conventional Implicatures. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199273829.001.0001 Selting, M. (2007). Lists as embedded structures and the prosody of list construction as an interactional resource. Journal of Pragmatics, 39(3), 483–526. https://doi.org/10.1016/j.pragma.2006.07.008

21 Plain velars in the Sḵwx̱ wú7mesh language Ethan Pincott Simon Fraser University The Sḵwx̱ wú7mesh language is a Central Salish language spoken from Burrard Inlet, north through Howe Sound, and up the Squamish and Cheakamus River valleys. Like other Salish languages, Sḵwx ̱ wú7mesh has a large inventory, including four series of dorsal phonemes (both velars and uvulars contrasting plain versus labialized). However, the plain k– series is significantly rarer than the rest (Kuipers, 1967), which is surprising considering that plain velars are some of the most frequent sounds in the world’s languages, and that they are unmarked relative to other dorsals (De Lacy, 2006). The scarcity of the k–series is due to the fact that earlier *k k’ x have fronted to palato-alveolar č č’ š in Sḵwx ̱ wú7mesh and other Central Salish languages (Galloway, 1988), leaving a gap in the phonological system. In fact, this same sound change has occurred in the neighbouring Wakashan and Chimakuan language families as well, indicating that the shift has diffused between related and unrelated languages across the entire Salish Sea area.

Despite their marginal status, plain velars do occur in a handful of words in Sḵwx̱ wú7mesh. This situation is true throughout the Central Salish branch (Galloway, 1988), which raises the issue of where these plain velars come from. Montler (1997) discusses the origins of plain k in the Straits languages, finding that all cases are loans from English or . Suttles (2004) identifies most Musqueam words with k k’ as loans as well, with a couple being “baby-talk” substitutes for the uvulars q q’. The purpose of this research is to determine the origins of the plain velars in the Sḵwx ̱ wú7mesh language, and compare the occurrence of the k–series in Sḵwx ̱ wú7mesh with other Central Salish languages.

Around forty roots occur with plain velars in Sḵwx̱ wú7mesh, far more than in Musqueam and the Straits languages. Of these, around half are clear borrowings from a European language or Chinook Jargon. In the case of words that have a French origin, it is difficult to tell whether they were borrowed via Chinook Jargon, or whether they were borrowed directly from French. In contrast, borrowings from English are relatively rare. Borrowing from another indigenous language accounts for a few more of the forms, but it is not always clear which language is the source.

22 Some occurrences of the k–series in Sḵwx̱ wú7mesh have cognates with labio-velars or uvulars in other languages. The semantics of some of these words suggest that they may be children’s language substitutions for the more difficult to pronounce sounds. Others suggest sound symbolic shifts of q > k, perhaps having a diminutive function. The most difficult to explain category of plain velars are those which appear to be archaisms which failed to undergo the *k > č shift. These roots have cognates in other Central Salish languages with the expected fronted reflexes of the Proto-Salish k–series, which raises interesting questions about the chronology of this shift in the evolution of the Central Salish branch. Finally, there are several words which do not fall neatly into any of these categories and remain unexplained. Further research, especially examining non-Salish languages for related forms, may be able to account for these roots in Sḵwx̱ wú7mesh. Whatever their origin, the occurrence of plain velars in Sḵwx̱ wú7mesh speaks to a complex history, both phonological and sociolinguistic, that remains largely unexplored.

References

De Lacy, P. (2006). Markedness: Reduction and preservation in phonology (Vol. 112). Cambridge University Press.

Galloway, B. D. (1988). Some Proto-Central Salish sound correspondences. In In honor of Mary Haas, from the Haas Festival Conference on Native American Linguistics (pp. 293-343).

Kuipers, A. H. (1967). The : Grammar, texts, dictionary (Vol. 73). Walter de Gruyter GmbH & Co KG.

Montler, T. (1997). On the Origins of š in the Straits . International journal of American linguistics, 63(3), 289-301.

Suttles, W. P. (2004). Musqueam reference grammar (Vol. 2). UBC Press.

23 The (Predictable) Effect of Environment on Vowel Quality in ʔayʔaǰuθəm Gloria Mellesmoen and Marianne Huijsmans University of British Columbia

A practical question in the documentation of endangered languages is whether the orthography privileges transparency in surface forms, aiding pronunciation, or phoneme forms, aiding the identification of . The mapping between underlying and surface forms, and its predictability in either direction, is central to this discussion. In many Salish languages, the are represented phonemically in the orthography and pronunciation is highly predictable from adjacent consonants (e.g. Kinkade 1967, Gibson 1973, Mattina 1973, Kuipers 1974, 1989, Thompson & Thompson 1992, van Eijk 1985; Bessell 1998a,b; Bird & Leonard 2009, Nolan 2017). Similar conditioning effects have been reported in ʔayʔaǰuθəm, a Central Salish language, such that the surface forms of the four underlying vowels are heavily influenced by adjacent consonants (e.g. Blake, 2000; Watanabe, 2003; Huijsmans, Mellesmoen, & Urbanczyk, 2018). Previous descriptions suggest that adjacent back (uvular) or palatal consonants are associated with lowered or raised vowels, respectively. This shown in (1), where the preceding and following uvular cause the high front vowel /i/ to retract to [ɛ] in (1a) and the low back vowel /a/ fronts to [ɛ] following a palatal, obscuring the underlying distinction between /a/ and /i/.

(1) a. /qiχ/ [qɛχ] ‘younger sibling’ b. /č̓ agat/ [č̓ ɛgat] ‘to help someone’

The effect of the back and palatal consonants is currently reflected in the orthographic representation, which matches the broad phonetic transcription in (1). Though the literature suggests that vowel quality is sensitive to adjacent consonants, the only acoustic description of vowels in ʔayʔaǰuθəm is Blake and Shahin (2008), which compares the vowels in stressed and unstressed positions, without accounting for adjacent consonants. The degree and directionality of conditioning effects has not been quantified. We aim to confirm the previously reported conditioning effects of consonants on vowel quality and measure these effects at different places of articulation for each of the four phonemic vowels (/a,ə,i,u/). We used a total of 489 tokens elicited from two fluent speakers of ʔayʔaǰuθəm. For each vowel, we considered the effect of preceding and following palatal and . These are compared to the same vowel in a neutral environment (where the preceding and following consonants were neither palatals or uvulars). This yielded a total of five conditions: back-vowel-neutral (97 tokens), neutral-vowel-back (94 tokens), palatal-vowel-neutral (105 tokens), neutral-vowel-palatal (89 tokens), and neutral-vowel- neutral (104 tokens). F1 and F2 measurements were taken at six points over the duration of the vowel. We find the anticipated effects of back and palatal consonants on vowel quality in both directions, with palatal consonants having a greater effect on following vowels and a greater effect on /a/ and /ə/ than the high vowels. For all vowels, there is raising and fronting next to palatal consonants and lowering and backing next to uvular consonants.

24 We conclude, based on the acoustic evidence, that surface forms in ʔayʔaǰuθəm are largely predictable from environment, as suggested by previous impressionistic work. This phonetic description of the effect of consonant place of articulation on vowel quality can be used in future for discussions about representing vowels orthographically and has a practical application for identifying underlying morphological forms.

References Bessell, N. 1998a. Phonetic aspects of retraction in Interior Salish. In E. Czaykowska-Higgins & M. D. Kinkade (eds.), Salish Languages and Linguistics: Theoretical and Descriptive Perspectives. Berlin: de Gruyter. Bessell, N. 1998b. Local and non-local consonant-vowel interaction in Interior Salish. Phonology 15: 1–40. Bird, S., & Leonard, J. 2009. Universality of articulatory conflict resolution: Evidence from Salish languages. Northwest Journal of Linguistics, 3(2), 1–29. Kinkade, M. Dale 1967. Uvular-pharyngeal resonants in Interior Salish. IJAL 33. 228–234. Kuipers, A. H. 1974. The . The Hague: Mouton. Kuipers, A. H. 1989. A report on Shuswap with a Squamish lexical appendix. Paris: Peeters/SELAF. Gibson, J. 1973. Shuswap grammatical structure. University of Hawaii Working Papers in Linguistics 5(5). Huijsmans, M., Mellesmoen, G., & Urbancyzk, S. (2018). To copy or not to copy, that is the question: Regarding consonant-vowel interactions in ʔayʔaǰuθəm imperfective . In Matthewson L., Guntly, E., & Rochemont, M. (Eds.), Wa7 xweysás i nqwal’utteníha i ucwalmícwa: He loves the people’s languages: Essays in honour of Henry Davis (659–674). Vancouver, BC:UBC Working Paper in Linguistics. Mattina, Anthony. 1973. Colville grammatical structure. University of Hawaii Working Papers in Linguistics 5(4). Nolan, T. 2017. A phonetic investigation of vowel variation in Lekwungen. MA thesis, University of Victoria. Thompson, L. C. & M. T. Thompson. 1992. The . University of Montana Occasional Papers in Linguistics 8. van Eijk, J. 1985. The Language. Amsterdam: University of Amsterdam.

25 Gender-Identity and the Naturalness of Referential Singular They Trevor Block Simon Fraser University Research Question. There are two questions investigated in this study. First, I look at whether referential singular they requires that its antecedent is genderless (e.g. someone, cyclist), or if it may also refer to antecedents with expected gender (e.g. mechanic, secretary). Second, I investigate whether there are differences between non-binary and cisgender people when judging the naturalness of referential singular they, given the previously mentioned contexts. Background. Singular they is commonly used to refer to genderless, generic singular antecedents, while he/ she is used to refer to gendered singular antecedents (Bodine, 1975; MacKay, 1980). Gender- neutral they has become more prominent in recent years, frequently used by individuals who identify as non-binary. Although attitude studies towards gender-neutral language have been conducted (Sarrasin et al, 2012), no studies have looked at naturalness ratings across various genders for singular they. A study conducted via paper questionnaire by Doherty and Conklin (2017) shows that when them refers to low gender-expectancy antecedents, as in (1), speakers find those particular constructions acceptable and rate them similarly to constructions with him/her. However, when them refers to high gender-expectancy antecedents, as in (2), speakers find those constructions highly unacceptable, rating them worse than gender-mismatch cases (e.g. mechanic… she), suggesting that them cannot refer to antecedents with expected or known gender.

(1) …he hit a cyclisti and knocked themi straight off the bike. (2) He saw a mechanici there and asked themi warily if there were any problems. The current study investigates singular they instead of singular them for two reasons. First, the third person singular pronoun is used more commonly as a subject than as an object in speech and writing. Second, they will always be pronounced and heard as /ðeɪ/ in speech. In contrast, the first consonant of them may be dropped in fast speech, and the following vowel may be neutralized to sound similar to him, resulting in /əm/. The latter possibility may cause lower ratings for singular them since speakers may frequently misinterpret them as him in fast speech, influencing their judgment in written context. In addition, the current study prevents participants from re-reading sentences while making judgments, and also prevents participants from re-rating older sentences to make their judgments consistent. Experimental items were also modified to include only one antecedent. Current Study. 30 non-binary and 50 cisgender participants were recruited. All recruited non-binary participants use they/them/their as their preferred pronouns, instead of he/him/his, she/her/hers, or others. Participants were instructed to fill out a pre-experiment demographic survey. Participants then read sentences in a word-by-word self-paced reading paradigm. Following each sentence, participants were asked to answer a yes-no comprehension question, or give a naturalness rating about the previous sentence on a scale of 1 (unnatural) to 7 (natural). Following the 60 trials, participants completed a post-experiment demographic survey.

26 In addition to test sentences based on Doherty and Conklin’s (2017) experiment, sentences with proper names were included and paired with either the gendered pronouns he/she, or gender- neutral they, as shown in (3).

(3) Nataliei was birdwatching alone in the park. After seeing an exotic bird, shei/theyi decided to write a journal entry. Results. Non-binary participants tended to rate all sentence types higher than cisgender participants. For both gender groups, sentences with they were rated as the most unnatural compared to he/she matches and mismatches. For both gender groups, gender-mismatch (high-marked) cases were rated as more natural than sentences with they. Non-binary participants rated sentences with proper names and they higher than cisgender participants. Judgments within gender groups are consistent whether they refers to low gender-expected antecedents or high gender-expected antecedents.

Example Cisgender Non-binary Control mechanic – he 6.09 6.33 High-Marked mechanic – she 5.90 6.27 High-They mechanic – they 5.52 5.93 Low-Marked cyclist – he 6.15 6.32 Low-They cyclist – they 5.55 5.95 Name-Match Sarah - she 6.03 6.28 Name-They Sarah - they 4.65 5.64 Table 1. Naturalness Ratings (1-7) Implications. The results suggest that they may refer to antecedents with or without expected gender, but that cisgender participants find it less natural than non-binary participants. The differences are likely due to usage and familiarity, since people who identify with the pronouns they/them/their are more likely to encounter and use them on a daily basis. Usage and familiarity also explains the case where they refers to proper names, suggesting that non-binary participants have less restrictions in their usage of referential singular they than cisgender participants. The results also suggest that, theoretically, singular they does not consider gender agreement with its antecedent. A self-paced reading study is currently in the works to investigate any differences in online processing. Selected References. Bodine, A., (1975). Androcentrism in prescriptive grammar: Singular ‘they’, sex-indefinite ‘he’, and ‘he or she’. Language in Society, 4(02), 129-146. Doherty, A. & Conklin, K., (2017). How gender-expectancy affects the processing of “them”. The Quarterly Journal of Experimental Psychology, 70(4), 718-735. MacKay, D. G. (1980). On the goals, principles, and procedures for prescriptive grammar: Singular they. Language in society, 9(3), 349-367. Sarrasin, O., Gabriel, U., & Gygax, P. (2012). Sexism and attitudes toward gender-neutral language. Swiss Journal of Psychology.

27 Reflector: another potential interpretation of the Mandarin reflexive ziji Peng(Benjamin)Han University of Calgary Introduction The Mandarin reflexive ziji can refer to antecedents in local or higher clauses (Huang and Tang, 1991; Pan, 2001), as demonstrated in sentence (1). Not only that, if sentence (1) is situated in a text narrating one’s thoughts, ziji can also refer to the reflector, i.e. the agent of thinking (Li, 1991). Ziji of this use is temporarily termed self-reflective ziji. (1) Johni shuo Billj chang piping zijii/j. John say Bill often criticize self ‘Johni said that Billj often criticized {himi, himselfj}.’ Self-reflective ziji is different from local or long-distance binding in that its antecedent is not necessarily (overtly) represented in the sentence containing ziji. Ziji’s reference to a reflector (especially an extra-sentential one) serves as another potential interpretation, complicating the studies of ziji as a long-distance anaphor. Careful studies of self-reflective ziji are thus necessary to gain a comprehensive understanding of ziji and tease apart its different uses. The current study explores the distribution of self-reflective ziji through a corpus study, which indicates two types of self-reflective ziji with distinct properties and licensing conditions. Moreover, an attempt is made to account for self-reflective uses of ziji as logophoricity. Background Self-reflective ziji requires an account that allows both intra-sentential and extra- sentential reflectors. Zijis referring to intra-sentential antecedents are mostly instances of long- distance anaphors embedded in clausal arguments of thinking verbs; these instances are explained by logophoricity (Huang and Liu, 2001; Anand, 2006). Actually, logophoricity can explain ziji’s extra-sentential coreference as well. For instance, Anand and Hsieh (2005) propose that logophoric ziji is bound by a covert referentially denoting element: perspective center, representing the psychological perspective from which a sentence is articulated. The value of a perspective center can be discourse-dependent, thus allowing extra-sentential coreference. Charnavel (2017) has a similar account, claiming that there is a syntactic projection high in the left periphery, headed by an empty operator OPLOG, with a clause as complement and a silent pronoun (pro) as specifier. Ziji is locally bound by pro, and discourse coreference ensures that pro is anteceded by a corresponding perspective center. Two types of perspective centers are distinguished, attitude holders and empathy loci, corresponding to intellectual and emotional perspective centers respectively. For a text narrating a reflector’s thoughts, the reflector is the intellectual perspective center. It seems promising to take a logophoric approach to self-reflective ziji. Main Research In order to learn about the linguistic facts of self-reflective ziji, I collect all sentence-initial zijis in the corpus of Center for Chinese Linguistics of Peking University; the corpus covers 509,913,489 Chinese characters. Sentence-initial zijis can help exclude interference from intra-sentential binding, as they are generally unbound for lacking c-commanding elements. In the corpus, 590 tokens of sentence-initial ziji are found to refer to reflectors, with 141 (23.9%) tokens anteceded by 1st-person speakers/writers, 1 (0.17%) token by 2nd-person pronouns and 448 (75.93%) tokens by 3rd-person referents. Among all 1st-person oriented tokens, 61 tokens occur in monologues or texts produced for the purpose of self-reflection. The quoted sentences in (2) exemplify reflective monologues. Ziji in reflective monologues or texts refers to reflectors, which coincide with speakers/writers, forming ziji’s 1st-person reference. (2) John kaishi zi-yan-zi-yu, “mingtian yao kaoshi. ziji neng kaoguo ma?” John begin self-talk-self-say tomorrow will exam self can pass Q ‘John began to talk to himself, “there will be an exam tomorrow. can I pass it?”. ’ Self-reflective ziji also occurs in narratives of one’s thoughts. Actually, the text in (2) can be rephrased as (3), with the protagonist John’s monologue presented as a narrative of his thought. It seems the third sentence in (3) is an indirect quotation or report of John’s monologue in (2).

28 (3) John fanchou-le. mingtian yao kaoshi. ziji neng kaoguo ma? John worry-PERF tomorrow will exam self can pass Q ‘John became worried. There would be an exam tomorrow. Can I pass it?’ Most of the self-reflective zijis appear in narratives of protagonists’ thoughts, including all 2nd and 3rd-person oriented tokens, and 80 first-person oriented tokens with narrators coinciding with protagonists. Narratives of thoughts are embedded in discourses where narrators consciously relate protagonists’ innate thoughts. That is, co-texts with overt realization of the protagonists (i.e. reflectors) are necessary, e.g. John in the first sentence of (3), although co-texts may not explicitly indicate the protagonists’ entering thinking status. As for self-reflective ziji in reflective monologues or texts, ziji invariably refers to 1st-person speakers/writers (also reflectors), and no co-text is necessary. Self-reflective zijis in (2) and (3) are characterized as context-dependent and discourse-dependent respectively. Following Charnavel (2017), self-reflective ziji is bound by a covert perspective center pro. In context-dependent cases, speakers/writers are permanently available perspective centers (i.e. reflectors) co-referring with pro. For discourse-dependent cases, narratives of thoughts are related from the perspective of protagonists, and pro is coreferential with the protagonists. Conclusions Based on the corpus data, self-reflective ziji comes in two types. Context- dependent ones occur when people write texts or talk to themselves as a means of self-reflection, whereas discourse-dependent ones occur when narrators relate protagonists’ thoughts from the protagonists’ perspectives. The two types differ in their purposes (communicative v.s. reflective) and the necessity of co-texts. Logophoricity can well capture the properties of each type. It is anticipated that ziji in discourse-dependent cases are more subject to potential ambiguity and rules governing discourse coherence. This study enriches the knowledge on self-reflective ziji and alerts future research to the interference brought by ziji’s potential to be interpreted as a reflector.

References

Anand, Pranav. 2006. De de se. Doctoral Dissertation, Massachusetts Institute of Technology. Anand, Pranav, and Feng-fan Hsieh. 2005. LDR in Perspective. In WCCFL 24. Charnavel, Isabelle. 2017. Logophorcity and locality: the view from French anaphors. URL Http://ling.auf.net/lingbuzz/002683. Huang, C.-T. James, and C.-S. Luther Liu. 2001. Logophoricity, Attitudes and ziji at the interface. In Long-Distance Reflexives, ed. Peter Cole, Gabriella Hermon, and C.-T. James Huang, volume Syntax and Semantics 33, 141–195. New York: Academic Press. Huang, C.-T. James, and C.C. Jane Tang. 1991. The Local Nature of the Long-Distance Reflexive in Chinese. In Long-distance anaphora, ed. Jan Koster and Eric Reuland, 263–282. New York: Cambridge University Press. Li, Naicong. 1991. Perspective-taking in Mandarin discourse. Doctoral Dissertation, State University of New York at Buffalo. Pan, Haihua. 2001. Why the Blocking Effect? In Long-Distance Reflexives, ed. Peter Cole, Gabriella Hermon, and C.-T. James Huang, volume Syntax and Semantics 33, 279–316. New York: Academic Press.

29 First Nations Language Revitalization in British Columbia: The case of the Tŝilhqot’in language in Yuneŝit’in Paula Laita Pallares University of the Basque Country UPV/EHU This presentation will share the main results obtained from the research on revitalization of the Tŝilhqot’in language in the Yuneŝit’in community. Tŝilhqot’in is a Dene language (Athapascan family) located in the interior of British Columbia. This province is the home of 34 indigenous languages, 61 dialects, and 8 language families (FPCC 2014). Like most of those languages, Tŝilhqot’in is in danger of disappearing since the number of speakers continues to decrease every year. This is because of a combination of historical and current socio-economic reasons that stem from European contact in the early 1800s. Colonization, wars, epidemics, the historical policies of assimilation carried out by the Canadian Government and the Residential School System (where children were removed from their homes and forbidden to speak their mother tongue) have all contributed to the decline of the language. The linguistic vitality of Tŝilhqot’in is one of the highest in the province. This First Nation language presents a large number of speakers: 19.9% of the total population are fluent; around 866 out of 4,352 people (FPCC 2014). However, Tŝilhqot’in can still be classified as Stage 7 ‘Shifting’ on the Graded Intergenerational Disruption Scale (Fishman 1991) as the language is rarely learned now as the mother tongue by children. Language revitalization through re-establishing home transmission may still be possible since grandparents and some of the parents learned Tŝilhqot’in as their first language. Traditionally an oral language, the first Tŝilhqot’in documentation efforts started in the early 60s when the alphabet was created. A decade later, the orthography and grammar were developed and recordings, transcriptions and translations of traditional knowledge and stories started to be produced, together with language teaching materials and other resources. In the last decade, many Tŝilhqot’in language revitalization efforts have been taken in Yuneŝit’in, one of the six Tŝilhqot’in communities. In 2015, Yuneŝit’in Government together with other two Tŝilhqot’in communities, Tl’esqox and Xeni Gwet’in First Nation, developed a Strategic Language Revitalization Plan and since then community priorities towards language revitalization have been addressed in the form of language immersion and culture camps, language immersion preschool programs, language immersion mentorship programs and development of new language resources. This research aims to advance the ongoing Tŝilhqot’in language revitalization work in Yuneŝit’in. It is a collaborative project run by Yuneŝit’in Government and myself and the main purpose is to explore community needs and priorities on Tŝilhqot’in language revitalization in Yuneŝit’in by gathering and sharing community perspectives on language knowledge and usage, language teaching/learning strategies and language resources. The research follows indigenous methodologies to ensure the work is not only respectful and culturally responsive, but also based on approaches and processes that fit with indigenous cultures, worldviews and ways of being (Absolon and Willett 2005, Graveline 2000, King 2013, Kovach 2010, Sinclair 2003, Tuhiwai Smith 1999). Participatory and community-based research principles are also applied (Strand et al. 2003). This work follows a mixed method approach, as both qualitative and quantitative data have been collected and analyzed (Cresswell 2013); however, the qualitative analysis prevails in this study in order to gain deep understanding of community perceptions and underlying reasons of the research topic (Mayan 2009). Aspects of the grounded theory methodology are also present, owing to the inductive nature of the research process (Glaser and Strauss 1967). Data collection methods are participant observation, semi-structured

30 conversation, sharing circle and document analysis of meeting minutes, language materials and other related documents. To ensure a representation of the community perspectives, all Yuneŝit’in families and different generations have been engaged in the project. Tŝilhqot’in Language Committee members, language teachers and experts, Yuneŝit’in Chief and Council members and Band office staff have also participated because of the influence of their roles in the community. Results will represent Yuneŝit’in community perspectives on the following: Tŝilhqot’in language knowledge and usage; main reasons and consequences of the language loss; importance of keeping the language alive; strategies to promote language use; priorities, challenges and strategies to teach/learn/acquire the language; and development of language resources.

References Absolon, K. and Willett, C. (2005) “Putting ourselves forward: Location in Aboriginal Research”, in Brown, L. and Strega, S. (2005) Research as Resistance Critical, Indigenous, and Anti-Oppressive Approaches. Toronto: Canadian Scholars’ Press Cook, E.D. (2013) A Tsilhqút’in Gammar. Vancouver: UBC Press Creswell, J.W. (2013) Research Design Qualitative, Quantitative, and Mixed Methods Approaches. Thousand Oaks, CA: SAGE Publications Fishman, J. (1991) Reversing the language shift. Clevendon, UK: Multilingual Matters First Peoples’ Cultural Council (FPCC) (2014) Report on the Status of B.C. First Nation Languages, Brentwood Bay, BC: First Peoples’ Cultural Council, retrieved from: http://www.fpcc.ca/files/PDF/Language/FPCC-LanguageReport-141016-WEB.pdf Glaser, B. G. and Strauss, A. L. (1967) The discovery of grounded theory: Strategies for qualitative research. Chicago, IL: Aldine Graveline, F. J. (2000) “Circle as Methodology: Enacting an Aboriginal Paradigm”, in Qualitative Studies in Education, 200, Vol. 13, No. 4, pp. 361-367. Manitoba, CA: Brandon University King, J. (2013) “Sisters in Spirit Research Framework: Reflecting on Methodology and Process”, in Jerry P. White, Peters, J., Beavon, D. and Dinsdale, P. (Eds.) (2013) Aboriginal Policy Research, Volume 10. Voting, Governance, and Research Methodology. Thompson Educational Publishing, Inc., pp. 269-285 Kovach, M. (2010) “Conversational Method in Indigenous Research”, in First Peoples Child & Family Review, Volume 5, Number 1, pp.40-48 Mayan, M. J. (2009) Essentials on Qualitative Inquiry. Walnut Creek, CA: Left Coast Inc. Sinclair, R. P. (2003) “Indigenous Research in social work: the challenge of operationalizing worldview” in Native Social Work Journal. Articulating Aboriginal Paradigms: Implications for Aboriginal Social Work Practice. Sudbury, ON: Lauretian University Strand, K., Cutforth, N., Stocker, R., Marullo, S. and Donohue, P. (2003) Community-Based Research in Higher Education: Principles and Practices. San Francisco, CA: Jossey-Bass Tuhiwai Smith, L. (1999) Decolonizing Methodologies, Research and Indigenous Peoples. London, UK: Zed Books.

31 Contact Languages of West China: The Relationship Between History and Linguistic Theory Nathan Loggins University of Washington

China’s centuries-long incursion into ethnic regions along its historic borders with cultural Tibet has left many language varieties as emblems of cultural contact between the inward bound Han ethnic group and local groups speaking Tibeto-Burman or Mongolic languages. Specifically, languages spoken in Qinghai and southern Gansu provinces and the western half of Sichuan show the results of language contact between the various groups: typically an SVO language, with largely analytic morphology, the Chinese spoken around the Qinghai capital of Xining exhibits SOV word order and post-positional case markers; the languages Wutun, spoken on the Qinghai/Gansu border, and Daohua, spoken in central Sichuan, are also SOV, post- positional case-marking, with ergative alignment, but also a primarily Sinitic lexicon. The origins of these languages are murky, and explanations for the language mixing have followed from the grammar alone or have been left unexplored. While Wutun and Daohua are considered by Janhunen, et al. (2008) and Atshogs (2004), respectively, to be mixed languages, the Xining dialect has been claimed by Dede (1999, 2007) to be restructured Chinese, the result of language shift, possibly among local Monguor people, to the language of the encroaching Chinese state. However, nominal control of the region by China before the mid-20th century does not necessarily entail a motivation for language shift, nor is it obvious how the newly arrived Han Chinese would have sustained contact with local Tibetans to lead to the mixed languages of Wutun and Daohua. Chinese control of much of the region was scant at best prior to the 1950s, and the state presence was largely in the form of military campaigns directed at central Tibet. Dede’s position on Xining seems to be largely motivated by grammatical data alone, assuming an inevitable adoption of Chinese culture by local peoples, the resulting contact-induced changes following from substratal interference of the shifting population as explained programmatically by Thomason and Kaufman (1992). The following paper examines the historical record to consider what sort of contact situation may have been plausible, given the means of Chinese-speakers’ arrival in the region, and how it compares with the settings of other languages whose development has been largely influenced by contact-induced change. It considers the possibilities of education systems and the areal monastic complex as factors. However, more promising explanations lie in the areas of trade and intermarriage, two phenomena common to the region. Such factors are more plausible means for the kind of intimate contact needed for language contact to have occurred. Employing a comparative methodology, the remainder of the study is an attempt to contextualize the languages of the region among different types of contact-heavy languages discussed in the literature. Comparisons are drawn between the settings of the languages in question, and mixed languages such as , spoken along the northern U.S.-Canadian

32 border, where French hunters and trappers married local Cree women, and Mednyj Aleut, spoken in the Copper Islands stretching between and Russian Siberia, the result of Russian Men and local Aleutian women intermarrying, both situations similar to that of central western China. Such languages, as analyzed by Bakker and Muysken (1994) and Vellupillai (2015) bear a stronger resemblance to Wutun and Daohua, and to some extent the Xining dialect, than either do to more traditional creoles, such as those spoken in the Carribbean or the South Pacific. Languages arising from trade settings tend to have simpler grammatical structures, as described by McWhorter (2001, 2007), and gravitate towards the least marked features available in the community (see Mufwene 1991, 2008). Other cases of presumed language shift tend to leave similarly simplified grammars in place, as explained in McWhorter (2007) and Trudgill (2011). Such does not seem to be an accurate description of the languages analyzed here, lending credit to the interpretation of language mixing as the product of bilingual, multi-ethnic communities formed through intermarriage, a process referred to as language-intertwining, rather than language shift or creolization via trade settings.

阿错意西微萨 [Atshogs, Yeshes Vodgsal = Yīxīwēisà Ācuò], Dǎohuà yánjiū 倒话研究 [Research on Daohua], Běijīng:Mínzú 民族出版社, 2004. Bakker, Peter and Muysken, P.C. "Mixed languages and language intertwining." In Arends et al. (1994) 41-52. Dede, K. R. S. (1999). Language contact, variation and change: the locative in Xining, Qinghai (Doctoral dissertation, University of Washington). Dede, K. (2007). The deep end of the feature pool: syntactic hybridization in Chinese dialects. Journal of Chinese Linguistics, 35(1), 58. Janhunen, Juha & Marja Peltomaa & Erika Sandman & Xiawu Dongzhou (2008). Wutun. Languages of the World / Materials, vol. 466. Muenchen: LINCOM Europa. McWhorter, John. "The world’s simplest grammars are creole grammars." Linguistic typology 5.2/3 (2001): 125-166. McWhorter, John. Language interrupted: Signs of non-native acquisition in standard language grammars. Oxford University Press, 2007. Mufwene, Salikoko S. "Pidgins, creoles, typology, and markedness." Development and structures of creole languages (1991): 123-143.Thomason and Kaufman Mufwene, Salikoko S. Language evolution: Contact, competition and change. Bloomsbury Publishing, 2008. Thomason, Sarah Grey, and Terrence Kaufman. Language contact, creolization, and genetic linguistics. Univ of California Press, 1992. Trudgill, Peter. Sociolinguistic typology: Social determinants of linguistic complexity. Oxford University Press, 2011. Velupillai, Viveka. Pidgins, Creoles and mixed languages: an introduction. Vol. 48. John benjamins publishing Company, 2015.

33 Investigating Language Asymmetry and Handedness in Participants with Diverse Language Background Olga Vasileva Simon Fraser University Right-handedness and language are considered unique humans abilities. Despite the fact, that other animals demonstrate various communicative abilities and motor preferences, no other species demonstrates such profound population-wide linguistic and manual asymmetries (Fitch & Braccini, 2013). As both handedness and language are unique to humans and are strongly lateralized, various researchers hypothesised these abilities are interdependent in individual development and possibly evolution (e.g. Corballis, 2003). Consequently, research investigating handedness and language relationship became a field of studies in its own right.

Despite the fact that handedness – language relationship have attracted researchers attention for a long time, the exact nature of this relationship is far from being understood with some researchers (e.g. Fagard, 2013) expressing serious concerns regarding our level of understanding of these relations. Although handedness and language asymmetry are presumably related in both evolutionary and individual development, more research is necessary to test this assumption.

One of the reasons research in handedness and language results in inconsistent findings and does not allow drawing firm conclusions is that studies addressing handedness or language asymmetry in adults frequently rely on quite homogeneous samples. That is, participants in such studies tend to be monolinguals in a given language, and most often this language is English. This situation raises the question, to which extent can such studies findings be extrapolated to humans in general, and how informative are they for understanding language-handedness relationship.

An integrative approach to the problem of language and handedness in humans requires more comprehensive approach to research. Specifically, it is necessary to examine handedness – language relations in more linguistically diverse samples. That is, participants with varied language background (depending on the type of language), and bi/multilinguals in addition to monolinguals.

The current submission discusses a project attempting to implement such integrative approach. The purpose of the project is to investigate handedness – language relations in a sample of participants with diverse linguistic background. Specifically, language asymmetry and handedness are investigated in participants having proficiency in various languages, with a number of participants being bilinguals.

34 According to two recent meta-analyses bilinguals exhibit a more balanced asymmetry in language processing with the right hemisphere being significantly involved in language processing (Vaid & Hull, 2006; Vaid & Hull, 2007). Moreover, the results indicate that overall monolinguals (MG) and late bilinguals (LB) (individuals acquiring a second language after the age of six) demonstrate a more left-lateralized pattern of language processing, while the early bilinguals (EB) (individuals acquiring a second language before the age of six) are less lateralized in that respect. Consequently, since BL differ from ML in language asymmetry, and since there is a presumed connection between handedness and language asymmetry in the adults (Cochet & Vauclair, 2012), it might be expected that bilinguals also differ from monolinguals in handedness.

No previous research has addressed the question of handedness – language asymmetry in relation to bilingualism. Additionally, while many previous studies have employed indirect measures of language asymmetry (e.g. asymmetry of a communicative gesture), in the present study participants are taking a dichotic listening test to determine their actual linguistic asymmetry profile (Hugdal, 2003). If language asymmetry and handedness is associated in the adults, we can expect congruent relationship between these parameters in each of the groups (EB, LB, ML) and significant differences between groups.

Finally, in order to capture the relationship between handedness and language more effectively, the study employs additional measures allowing separating these relationships from potential co-founding parameters, such as general motor asymmetry (footedness) and cognitive factors. The research team is currently working on data collection and coding; findings are planned to be reported at the conference.

References Cochet, H., & Vauclair, J. (2012). Hand preferences in human adults: Non-communicative actions versus communicative gestures. cortex, 48(8), 1017-1026. Corballis, M. C. (2003). From hand to mouth: The origins of language. Princeton University Press. Fagard, J. (2013). Early development of hand preference and language lateralization: are they linked, and if so, how?. Developmental psychobiology, 55(6), 596-607. Fitch, W., & Braccini, S. N. (2013). Primate laterality and the biology and evolution of human handedness: a review and synthesis. Annals of the New York Academy of Sciences, 1288(1), 70-85. Hugdahl, K. (2003). Dichotic listening: an experimental tool in clinical neuropsychology. In Experimental methods in neuropsychology (pp. 29-46). Springer US. Hull, R., & Vaid, J. (2007). Bilingual language lateralization: A meta-analytic tale of two hemispheres. Neuropsychologia, 45(9), 1987-2008. Vaid, J., & Hull, R. (2006). Laterality and language experience. Laterality, 11(5), 436-464

35 Instructed SLA of English articles and noun types by L1 Chinese international students Dakota Thomas-Wilhelm University of Iowa/Universitat Autònoma de Barcelona

Abstract: In the field of SLA, much research has been done on either the acquisition/recognition of English countable/uncountable noun distinctions (Choo & Slabakova, 2014; Choi & Ionin, 2017) or English articles (Ionin, Ko, & Wexler, 2004), little research has been done on how the presence or absence of articles might relate to the acquisition of English noun types. This project seeks to fill this gap by looking at the acquisition of English countable/uncountable noun distinctions and articles by Chinese ESL learners. Under particular investigation is how learners of an "article-free" language (e.g., Chinese) acquire English articles in combination with different noun types. Using two different data collection tasks, a self-paced reading task (SPRT) and an acceptability judgment task (AJT), the study investigates the acquisition of English article and noun type distinctions after explicit instruction using linguistically-informed teaching materials (Lopez, 2017; Snape & Yusa, 2013), which are materials that have been designed to use metalinguistic explanations and L1 comparisons to teach linguistic concepts. Analyses of the results show that there may be an effect of instruction on explicit knowledge, but not implicit knowledge. Implications of the study may provide a new teaching pedagogy ESL grammar.

References: Cho, J. & Slabakova, R. (2014). Interpreting definiteness in a second language without articles: The case of l2 Russian. Second Language Research, 30(2), 159-190. Choi, S. H. & Ionin, T. (2017). Acquisition and processing of mass nouns in L2-English by L2 learners from generalized classifier languages: Evidence for the role of atomicity. In M. LaMendola & J. Scott (eds.), Proceedings of the 41st annual Boston University Conference on Language Development, 154-167. Somerville, MA: Cascadilla Press. Ionin, T., Ko, H., & Wexler, K. (2004). Article semantics in L2 acquisition: The Role of Specificity. Language Acquisition, 12(1), 3–69. Lopez, E. (2017). Teaching the English article system: Definiteness and specificity in linguistically- informed instruction. Language Teaching Research, 00(0), 1-18. doi: 10.1177/1362168817739649 Snape, N., & Yusa, N. (2013). Explicit article instruction in definiteness, specificity, genericity and perception. In M. Whong, K. Gil, & H. Marsden (Eds.), Universal Grammar and the second language classroom (pp. 161–183). Dordrecht: Springer.

36 Investigating Acquisition of L2 Articles: Evidence from Japanese-speaking and English-speaking learners of Standard Arabic Albandary Aldossari Western University

The present study examines the Fluctuation Hypothesis of Ionin, Ko and Wexler (2004) and the role of transfer in L2 learners of Arabic whose first language is Japanese, a language that lacks articles and English, a language that encodes definiteness and indefiniteness. According to Ionin et al, L2 learners fluctuate between definiteness and specify when the L1 does not have articles. However, transfer from an L1 with articles will override fluctuation. As Arabic is a language with a marker for definiteness, but not for indefiniteness, this may create difficulties for L2 learners. The Arabic definite article is -al-, which is a bound , but there is no explicit indefinite article (Jaensch & Ghisseh, 2009). The Arabic definite article is a and it has to be attached to the noun it modifies. For example, the sentence /katabtu al-dars/, which means 'I wrote the lesson' is definite. In contrast, /katabtu dars/ means 'I wrote a lesson', with no explicit marker for indefiniteness. I will report an empirical study that examined the acquisition of Arabic articles by learners of English, a language that encodes definiteness and indefiniteness, and Japanese that lacks articles. This study tries to answer three questions concerning L2-Arabic article use: (1) does the absence of L1 articles lead to fluctuation in the acquisition of Arabic; (2) does knowledge of English overcome fluctuation in interpretation of Arabic definite contrast; and (3) does the lack of indefinite articles in Arabic complicate the acquisition process. There are two groups of participants: the first group consists of 10 English native speakers of Arabic at the advanced level who are living in Saudi Arabia; the second group consists of 10 Japanese native speakers learning Arabic at the advanced level living in Japan. There is also a control group of 10 native Arabic speakers. There are two experimental tasks: a preference task in which the learner reads a short scenario and then decides between two sentences, one with an article (definite interpretation), and one without (indefinite interpretation). The second is a sentence completion task in which participants are asked to complete a sentence in Arabic with definite or indefinite interpretation. Distractors will include numbers and demonstratives. This study will contribute to our understanding of the acquisition of articles in general and the acquisition of Arabic as an L2, a language that has not been studied sufficiently leading to inadequate and limited knowledge (but see Aldeeky, 2014). This study is on-going and preliminary results will be presented.

References Aldeeky, M. (2014). The non-native speakers errors in the use of the Arabic definite article. Studies in Literature and Language 9, (3), 40-46. Jaensch, C., & Sarko, G. (2009). Sources of fluctuation in article choice in English and German by Syrian Arabic and Japanese native speakers. EUROSLA Yearbook, 9(1), 33-55. Ionin, T., Ko, H., & Wexler, K. (2004). Article semantics in L2 acquisition: The role of specificity. Language Acquisition, 12(1), 3-69.

37