Faculty & Researcher Works
Permanent URI for this collectionhttps://hdl.handle.net/10125/49522
Browse
Recent Submissions
Item type: Item , Data transparency and citation in the journal Gesture(John Benjamins Publishing Company, 2019) Gawne, Lauren; Krajcik, Chelsea; Andreassen, Helene N.; Berez-Kroeker, Andrea L.; Kelly, Barbara F.Item type: Item , Color figures from the Open Handbook of Linguistic Data Management(2021-01-18) Berez-Kroeker, Andrea L.; McDonnell, Bradley; Koller, Eve; Collister, Lauren B.; Akama, HiroyukiThese images are from the chapter by Hiroyuki Akama, "Managing, sharing and reusing fMRI data in computational neurolinguistics" in the Open Handbook of Linguistic Data Management, edited by Andrea L. Berez-Kroeker, Bradley McDonnell, Eve Koller and Lauren B. Collister. Cambrige, MA: MIT Press Open.Item type: Item , Item type: Item , Stimuli Items Used in the Study "Revisiting Tagalog Word Order Preferences"(2019-08-07) Bondoc, Ivan Paul; Schafer, Amy J.This study investigates linear word order tendencies in single-sentence contexts for four Tagalog voice alternations. The work hypothesizes two grammatical constraints that influence production of declaratives in Tagalog: an agent-first constraint, where the agent preferentially occupies the first post-verbal argument position; and a pivot-second constraint, where the pivot preferentially occupies the second argument position. We test these proposed constraints in a sentence continuation task. Native Tagalog adults demonstrated word order patterns aligning with these constraints in the patient, benefactive, and instrument voice. The two constraints resolved in the agent voice to exhibit two similarly preferred patterns. The study findings extend support for a probabilistic account where probability distributions, influenced by interacting grammatical constraints, determine Tagalog word order preferences.Item type: Item , Topological Relations in Pohnpeian(Walter de Gruyter GmbH, Berlin/Boston, 2017) Rentz, BradleyThis article explores a more nuanced understanding of topological relations in the Pohnpeian language (Austronesian). The BowPed Toolkit (Bowerman and Pederson 1992. Topological relations picture series. In Space stimuli kit 1.2: November 1992, 51, Nijmegen: Max Planck Institute for Psycholinguistics. http://fieldmanuals. mpi.nl/volumes/1992/bowped/.) is employed as an elicitation tool with five Pohnpeian speakers. Evolutionary classification tree modeling is used as a discovery tool to find patterns in the data. The results show that the two prepositions in Pohnpeian, nan and ni, should be redefined in terms of topological relations as ‘containment’ and ‘attachment’ respectively. Likewise the meaning of some prepositional nouns are further revised.Item type: Item , Data Citation in Linguistics: Looking forward to new standards(2018) Gawne, Lauren; Berez-Kroeker, Andrea L.; Andreassen, Helene N.Item type: Item , Data Transparency and Citation in Gesture(2018) Gawne, Lauren; Krajcik, Chelsea; Andreassen, Helene N.; Berez-Kroeker, Andrea L.; Kelly, Barbara F.Gesture Studies has a strong history of research that spans multiple fields, but there is still not a robust culture of valuing reproducibility.Item type: Item , Agent versus non-Agent motions influence language production: Word order and perspective in a VOS language(Cognitive Science Society, 2018) Sato, Manami; Niikuni, Keiyu; Schafer, Amy J.; Koizumi, MasatoshiIs language production isolated from our experiences of physical events, or can physical motion affect the conceptual saliency of the components of a to-be-described event, in ways that affect its linguistic description? This study examined the influence of physical motion on the interpretation and description of simple transitive events. More specifically, we investigated whether engagement in non-speech physical actions affects the relative location of verbs versus arguments in sentence production, and the relative location and prominence of Agents, by testing native speakers of Truku, a language that allows flexibility in each of these options and presents under-studied typological patterns.Item type: Item , Data presented in Camp & Schafer (2018, LabPhon)(2018) Camp, Amber B.; Schafer, Amy J.Item type: Item , Appendices for "Survey of Current Reproducibility Practices in Linguistics Journals, 2003-2012."(2017) Berez-Kroeker, Andrea L.; Gawne, Lauren; Kelly, Barbara F.; Heston, TylerItem type: Item , Supplemental files for "Survey of Current Reproducibility Practices in Linguistics Journals, 2003-2012."(2017) Berez-Kroeker, Andrea L.; Gawne, Lauren; Kelly, Barbara F.; Heston, TylerItem type: Item , Norming Data on the Study "Influences of Grammatical Constraints on Tagalog Linear Word Order Preferences"(2018-10-18) Bondoc, Ivan Paul; Schafer, Amy J.This study investigates how grammatical constraints in a probabilistic architecture drive linear word order tendencies in four Tagalog voice alternations in single-sentence contexts. This study hypothesizes two constraints influential to the generation of a linguistic message in Tagalog: an agent-first constraint, where the agent preferentially occupies the first post-verbal argument position; and a pivot-second constraint, where the pivot preferentially occupies the second argument position. In data from a sentence continuation task, native Tagalog adults demonstrated word order patterns aligning with the constraints in the patient, benefactive, and instrument voice. These two constraints resolved in the agent voice to exhibit two similarly preferred patterns. The study findings concur with an account where probability distributions, influenced by interacting grammatical constraints, determine Tagalog word order preferences.Item type: Item , Tsova-Tush "intensive" consonants(2018-06-21) Hauk, BrynTsova-Tush (ISO 639-3: bbl), also known as Batsbi or Bats, is a severely endangered Northeast Caucasian language spoken in the village of Zemo Alvani, Georgia. Tsova-Tush stops have a phonemic three-way contrast between aspirated, ejective, and voiced; additionally, another feature (represented here orthographically by <ː>) distinguishes stops at two places of articulation (dental /tʰː, t'ː/, uvular /qʰː, q'ː/). Some previous descriptions label this distinguishing feature "intensive," but provide no articulatory or acoustic description of "intensiveness" while rejecting the possibility that these could be geminates. This raises the following question, the focus of the present study: what are the acoustic properties of the so-called "intensive" consonants in Tsova-Tush and their non-intensive counterparts? In this paper, multiple acoustic measurements of non-intensive and intensive stops are compared; specifically, total duration, closure duration, and VOT of the stops; intensity during the burst and during the post-burst interval; duration of the preceding vowel; and f0 and H1*-H2* of the following vowel. This study found that intensive consonants had a longer total duration and a longer closure duration than non-intensives, while these stops did not differ from each other on any other measure tested. Therefore, it is concluded that Tsova-Tush "intensive" stops are in fact geminates.Item type: Item , Minimal Checklist for the Preservation of Digital Language Documentation Materials(2018) DELAMAN, The Digital Endangered Languages and Music Archive NetworkThis checklist has been developed by DELAMAN to serve as a guide to the minimal level of digital data preservation that is generally accepted by the professional standards of language documentation. It should be noted that going above and beyond this baseline level of preservation is desirable, encouraged and even compulsory by granting agencies, particular archives, and/or professional expectations.Item type: Item , Diachronically stable, lexically specific variation: The phonological representation of secondary /æ/-lengthening(2017-06-12) Kettig, ThomasSecondary /æ/-lengthening, also known as the bad-lad split (bad as [bæːd] and lad as [læd]), is noted in several 20th century impressionistic descriptions of Southern Standard British English (SSBE) [4, 8, 12]. This paper reports results from an acoustic phonetic analysis of the duration of monosyllabic /æ/ (trap) words in SSBE. The apparent stability of this lengthening over the past century indicates that such sub-phonemic lexical variation is not necessarily indicative of a change still in progress. While its sub-phonemic nature is best accounted for with a usage-based phonological framework in which detailed phonetic properties are stored in lexical representations, the appearance of lengthening in higher-frequency words otherwise predicted to undergo reduction points to a more complex interaction of vowel duration and frequency than previously reported [7, 10]. Twenty-one native SSBE-speaking students at the University of Cambridge, aged 18–24, were recorded reading sentences embedded with 73 monosyllabic words containing the stressed vowel /æ/ (token n=1,777). Duration and F1/F2 were measured in Praat [1], with modal, breathy, and preaspirated sections coded separately and vowel duration defined as the modal plus breathy portions. A linear mixed effects model predicting duration with voicing and manner and place of articulation of coda consonants as well as word frequency entered as fixed effects along with by-word and by-speaker random intercepts and their interactions was run. Effects of phonological environment on length were observed as expected; results also indicated significant lengthening of higher-frequency words (p=0.009). The coefficients of the by-word random intercepts offer a window into how ‘unexpectedly’ long or short individual /æ/ words are once the phonological features of the post-tonic consonant(s) are accounted for; words singled out a century ago as being ‘long’ (e.g. bag, that, bad, jam, sad) still top the ranking (Figure 1). The secondary /æ/-lengthening reported here bears a striking resemblance to the reconstructed initial stages of primary /æ/-lengthening (also known as the trap-bath split), a diachronic process that began in Southern England around the 17th century as allophonic length variation in reflexes of the Middle English /æ/ vowel [12]. While Lexical Phonology has been used to analyze primary /æ/-lengthening, such a model predicts that allophonic contrasts, processed post-lexically, should never be involved in lexically selective change [6, 9]. On the other hand, usage-based theories in which people store detailed phonetic properties of individual words allow persistent biases in pronunciation to accumulate into small but consistent sub-phonemic differences [2, 11]. However, while Exemplar Theory has been used to account for vowel reduction in higher-frequency words, the present data indicates that the opposing process of lengthening has in fact occurred some high-frequency words [3, 5]. It appears that just as primary /æ/-lengthening represents an ossified, half completed sound change, secondary /æ/-lengthening seems to have undergone the same lengthening process but lexically ossified at the sub-phonemic level. Phonemic contrastiveness does not seem to be a prerequisite for the stable maintenance of a lexically specified split, even when this directly counteracts an overarching articulatory process. Given this active resistance to the process of phonetic reduction, it is proposed that this in fact represents stable transmission and storage of lexically specified sub-phonemic allophony. In a usage-based framework, it would be predicted that such variation should affect more frequent rather than less frequent words, since a learner would have to encounter such words often enough to consistently store the duration differences in the word memories accessed in speech production. Figure 1. Words investigated, plotted by amount of lengthening unaccounted for by fixed effects. [1] Boersma, P. & D. Weenink. 2015. Praat: doing phonetics by computer [Computer program]. Version 5.4.07. http://www.praat.org/. [2] Bybee, J. 2000. The phonology of the lexicon: Evidence from lexical diffusion. In M. Barlow & S. Kemmer (Eds.), Usage-based models of language. Stanford: CSLI, pp. 65–85. [3] Bybee, J. 2001. Phonology and language use. Cambridge: Cambridge University Press. [4] Cruttenden, A. 2001. Gimson’s Pronunciation of English, Sixth Edition. Oxford: Oxford University Press. [5] Gahl, S. 2007. “Thyme” and “Time” are not homophones. Word durations in spontaneous speech. Language 84(3): 474–496. [6] Harris, J. 1989. Towards a lexical analysis of sound change in progress. Journal of Linguistics 25(1): 35–56. [7] Hay, J., J. Pierrehumbert, A. Walker, & P. LaShell. 2015. Tracking word frequency effects through 130 years of sound change. Cognition 139: 83–91. [8] Jones, D. J. 1918. An outline of English phonetics. Leipzig: Teubner. [9] Kiparsky, P. 1988. Phonological change, in F. J. Newmeyer (Ed.), Linguistics: The Cambridge Survey. Volume I, Linguistic Theory: Foundations. Cambridge: Cambridge University Press, pp. 363–415. [10] Phillips, B. 2006. Word frequency and lexical diffusion. New York: Palgrave Macmillan. [11] Pierrehumbert, J. 2001. Exemplar dynamics: Word frequency, lenition, and contrast. In J. Bybee & P. Hopper (Eds.), Frequency effects and the emergence of lexical structure. Amsterdam: John Benjamins, pp. 137–157. [12] Wells, J. C. 1982. Accents of English (Vols. 1-3). Cambridge: Cambridge University Press.Item type: Item , Toward a Typology of Intonation Unit Cues(2011) Mulder, Jean; Berez, Andrea L.Item type: Item , The Use of Archives in Endangered Language Preservation: The State of the Art(2015) Henke, Ryan; Berez, Andrea L.; Holton, Gary; McDonnell, Brad; Rouvier, RuthItem type: Item , Data Citation in Linguistic Typology: Working Towards a Data Citation Standard in Linguistics(2017) Gawne, Lauren; Berez-Kroeker, Andrea L.; Andreassen, Helene N.; Okura, EveItem type: Item , Making Pacific Language Materials Discoverable(2017) Berez-Kroeker, Andrea L.; Kleiber, Eleanor; Yarbrough, Danielle; Chopey, Mike; Shelby, RyanItem type: Item , Reproducible Research in linguistics: The past and future of “data work” in our field(2017) Berez-Kroeker, Andrea L.