Monthly Archives: April 2018

Call for Papers: Annual Meeting on Phonology 2018

Call for Papers

Annual Meeting on Phonology 2018

We are seeking high-quality unpublished research in all areas of theoretical, experimental, and computational phonology for presentation at the 2018 Annual Meeting on Phonology (AMP 2018), to take place October 5-7, 2018 and hosted by the Linguistics Department at the University of California, San Diego. We welcome submissions that address issues in the phonologies of spoken and/or signed languages. This is the sixth installment of the Annual Meetings on Phonology, following the 2013 inaugural meeting at the University of Massachusetts Amherst, and subsequent meetings hosted by MIT, UBC/SFU, USC, and NYU.

This year’s conference features a workshop entitled “Methods in phonological data collection and analysis of underdocumented languages,” with associated tutorials and invited speakers. We are particularly interested in high-quality research submissions that address the workshop theme.

Submission Guidelines

We invite abstracts for either oral presentation (20 minutes, followed by 10 minutes of discussion) or poster presentation. Abstracts must be anonymous, so please be sure to eliminate any identifying information and metadata from the document. Length is limited to a maximum of two single-spaced pages (US Letter), figures and references included. Font size should be 12-point, with margins of at least one inch (2.54cm) left on all sides. Abstracts must be submitted in .pdf file format.

Submissions are limited to three per author, with at most one submission being single-authored.

The deadline for abstract submission is Friday, June 1, 11:59pm PDT (23:59 GMT-7).

Abstract submission link: https://easychair.org/conferences/?conf=amp2018

Invited Speakers

Methods workshop tutorials

  • Underdocumented language data corpus construction
    (Gabriela Caballero, UC San Diego)
  • Electroglottography for voice analysis
    (Marc Garellek, UC San Diego)
  • Investigating underdocumented tone systems
    (Bert Remijsen, University of Edinburgh)

SPOT tutorial

SPOT (https://people.ucsc.edu/~jbellik/research/spot/interface1.html) is a web application which performs automatic candidate generation and constraint evaluation over prosodic parses of sentences. It is designed for use in conjunction with OTWorkplace (https://sites.google.com/site/otworkplace/), a tool for manipulating OT tableaux, determining rankings, and calculating factorial typologies.

In association with the SPOT tutorial, we are hoping to gather both experienced and novice OTWorkplace users to share OTWorkplace uses and tools. If you are interested in attending an OTWorkplace workshop at AMP 2018, please fill out the form at https://tinyurl.com/otworkplace-amp2018 by the abstract deadline.

Contact

All questions about the conference should be emailed to phonology@ucsd.edu.

Publication

All presentations (in both the general and workshop sessions) are eligible for publication in the open-access on-line conference proceedings hosted by the Linguistic Society of America. Oral presentations will appear in the main Proceedings and poster presentations will appear in the Supplemental Proceedings.

Sponsors

Financial support for AMP 2018 is provided by the UC San Diego Division of Social Sciences. (Additional support from the National Science Foundation is pending for the Methods in phonological data collection and analysis of underdocumented languages workshop.)

Shaw & Kawahara (2018) – Assessing feature specification in surface phonological representations through simulation and classification of phonetic data

Assessing feature specification in surface phonological representations through simulation and classification of phonetic data
Jason Shaw, Shigeto Kawahara
direct link: http://ling.auf.net/lingbuzz/003878
February 2018
Many previous studies have argued that phonology may leave some phonetic dimensions unspecified in surface representations either because of deletion or lexical underspecification. Lacking a phonologically specified phonetic target, the phonetic signal in these cases can only be structured by phonetic interpolation between segments flanking the targetless element. However, natural variability in the phonetic signal presents a challenge for identifying instances of phonetic interpolation. Our approach to this challenge is to explicitly model phonetic variability providing a computational link between phonological hypotheses and phonetic data. To this end, we set up stochastic generators of competing phonological hypotheses and use them to compute, on a token-by-token basis, the likelihood that a phonetic signal is the consequence of phonetic interpolation, defined as a smooth interpolation between flanking segments, or, alternatively, that it is structured by a phonological target. The empirical material used to demonstrate the approach comes from Electromagnetic Articulography recordings of high vowel devoicing in Japanese. We use Discrete Cosine Transform to express tongue dorsum movement trajectories as a small number of frequency components (cosines differing in frequency and amplitude) that correspond to linguistically meaningful signal modulations, i.e., articulatory gestures. Our stochastic generators operate over this frequency space, generating tongue dorsum movements with realistic variation according to the presence or absence of a lingual articulatory gesture for the devoiced vowel. Finally, a Bayesian classifier trained on simulations of the targetless trajectory assigns posterior probabilities to the data. Results indicate that /u/ is optionally produced without a vowel height target in Tokyo Japanese and that the frequency of targetlessness varies across phonological environments.

Format: [ pdf ]
Reference: lingbuzz/003878
(please use that when you cite this article)
Published in: Phonology, to appear
keywords: phonetics, computational phonology, underspecification, articulatory target, ema, discrete cosine transform, bayesian classification, phonology

Yang (2018) – Three equations: A formalist perspective on language acquisition

Three equations: A formalist perspective on language acquisition
Charles Yang
direct link: http://ling.auf.net/lingbuzz/003872
February 2018
Language acquisition is a computational process by which language-specific experience is integrated into the learner’s initial stage of knowledge. To understand language acquisition thus requires precise statements about these components and their interplay, stepping beyond the philosophical and methodological disputes in the field such as the generative vs. usage-based approaches. I review three simple equations that have guided the study of child language acquisition: How learners form and select linguistic hypotheses, and how re- searchers assess the progress of language acquisition with rigor and clarity. Furthermore, I suggest that these equations are equally applicable to second language acquisition, yielding potentially important insights on the continuities and differences between child and adult language.

Format: [ pdf ]
Reference: lingbuzz/003872
(please use that when you cite this article)
Published in: Linguistic Approaches to Bilingualism (Target article)
keywords: language acquisition, l1, l2, computational linguistics, corpus linguistics, productivity, usage-based, morphology, syntax, phonology

Tyler (2018) – Simplifying Match Word: Evidence from English functional categories

Simplifying Match Word: Evidence from English functional categories
Matthew Tyler
direct link: http://ling.auf.net/lingbuzz/003869
February 2018
In work on the syntax-prosody interface, there is a prevalent idea that while lexical categories are preferentially mapped to prosodic words, no such pressure exists for functional categories (Selkirk 1984, 1996, 2011, Truckenbrodt 1999, Elfner 2012). In Match Theory (Selkirk 2011), in which syntax-prosody isomorphism is enforced by a series of violable constraints, this supposed pressure is built into the system with the claim that the Match Word constraint in some sense ‘ignores’ functional categories. I argue that this idea is misguided, and that Match Word does not discriminate between lexical and functional heads. The pervasive phonological reduction of function words, rather than being a consequence of Match Theory, is instead ascribed to idiosyncratic prosodic properties of function words specified in the lexicon. In particular, I adopt the model of prosodic subcategorization frames (Inkelas 1989, Zec 2005). This approach explains particular interactions that would be unexpected if Match Word were genuinely indifferent to functional categories, and fits in with a large body of evidence suggesting that functional elements can behave in prosodically idiosyncratic ways (Nespor & Vogel 1986, Inkelas & Zec 1990, Zec 2005, Bennett et al. to appear). The evidence comes from the behavior of several classes of English function categories: prepositions, auxiliaries, determiners, weak object pronouns and contracted negation ‘-n’t’.

Format: [ pdf ]
Reference: lingbuzz/003869
(please use that when you cite this article)
Published in: submitted ms.
keywords: english, function words, functional categories, prosody, syntax-prosody interface, match theory, syntax, phonology
previous versions: v1 [February 2018]

Traoré & Féry (2018) – Nominal classes and phonological agreement in Frò?ò (Tagbana)

Nominal classes and phonological agreement in Frò?ò (Tagbana)
Yranahan Traoré, Caroline Féry
direct link: http://ling.auf.net/lingbuzz/003863
February 2018
Besides total vowel harmony and nasal harmony, Frò?ò, a Gur language spoken in Côte d’Ivoire, has a pervasive consonantal agreement in its nominal domain: the morphologically ‘dependent morphemes’ (in terms of concord, Corbett 1991) or ‘associate morphemes’ (Hockett 1958), of nominal heads, agree in their articulator and [±continuant] features, a special case of alliterative concord, because the head noun is playing no role in the alliteration. Besides the consonantal agreement features, these free associate morphemes have their own vocalic features and their own [±voice] feature. The paper starts with a review of nominal classes in Frò?ò and the morphological structure of simple nouns. It is proposed that Frò?ò has a partly non-concatenative morphology and that a standard Distributed Morphology analysis cannot fully account for the phonology of the resulting morphemes. In the second part, the role of phonology is investigated. It is proposed that well-formed morphemes are the result of partial or deficient phonological specifications in need of repair. The phonological approach is couched in an optimality-theoretic framework (see Saba Kirchner 2010 and Bye & Svenonius 2012 among others for similar proposals). Part of the final specification of these morphemes are due to phonological repairs elicited by markedness and faithfulness constraints, and the filling in of features due to vowel and nasal harmony, as well as consonant epenthesis.

Format: [ pdf ]
Reference: lingbuzz/003863
(please use that when you cite this article)
Published in: none so far
keywords: nominal classes, alliterative concord, tagbana, morphology, phonology

Whang (2018) – Recoverability-driven coarticulation: Acoustic evidence from Japanese high vowel devoicing

Recoverability-driven coarticulation: Acoustic evidence from Japanese high vowel devoicing
James Whang
direct link: http://ling.auf.net/lingbuzz/003856
February 2018
High vowel devoicing in Japanese, where /i, u/ in a C1-V-C2 sequence devoice when both C1 and C2 are voiceless, has been studied extensively, but factors that contribute to the devoiced vowels’ likelihood of complete deletion is still debated. This study examines the effects of phonotactic predictability on the deletion of devoiced vowels. Native Tokyo Japanese speakers (N=22) were recorded in a sound-attenuated booth reading sentences containing lexical stimuli. C1 of the stimuli were /k, ?/, after which either high vowel can occur, and /?, ?, s, ç/, after which only one of the two is likely to occur. C2 was always a stop. C1 duration and center of gravity (COG), the amplitude weighted mean of frequencies present in a signal, were measured. Duration results show that devoicing lengthens only non-fricatives, while it has either no effect or a shortening effect on fricatives. COG results show that coarticulatory effects of devoiced vowels are evident in /k, ?/ but not in /?, ?, s, ç/. Devoiced high vowels, therefore, seem to be more likely to delete when the vowel is phonotactically predictable than when it is unpredictable.

Format: [ pdf ]
Reference: lingbuzz/003856
(please use that when you cite this article)
Published in: Journal of the Acoustical Society of America
keywords: recoverability, predictability, devoicing, reduction, japanese, phonology
previous versions: v1 [February 2018]

Tokizaki (2018) – Externalization, stress and word order

Externalization, stress and word order
Hisao Tokizaki
direct link: http://ling.auf.net/lingbuzz/003881
January 2018
In the principle and parameter theory of generative grammar in the 1980s, the variation of word order in languages was ascribed to the value of the head directionality parameter (head-initial/final). Following the minimalist program of linguistic theory, this paper argues that syntactic computation builds a structure without linear order, which is only linearized at the externalization to the sensory-motor system, according to the stress pattern of the language. We can derive word order variation in the world’s languages from their phonology, especially their stress systems. We can do away with the complement movement that derives head-final structure from the head-initial structure constructed by the Linear Correspondence Axiom (Kayne (1993)).

Grillo, Aguilar, Leah, Santi & Turco (2018) – Prosody of classic garden path sentences: The horse raced faster when embedded

Prosody of classic garden path sentences: The horse raced faster when embedded
Nino Grillo, Miriam Aguilar, Roberts Leah, Andrea Santi, Giuseppina Turco
direct link: http://ling.auf.net/lingbuzz/003868
January 2018
Prosody, it is assumed, does not always disambiguate syntax. We investigate one classic case at point from the psycholinguistics literature: garden path sentences involving the main-verb vs. reduced relative clause contrast (the horse raced past the barn (and) fell). Despite their centrality in shaping theories of sentence processing, no experimental work to date has investigated the prosody of these sentences. We show that, contrary to previous assumptions (Fodor 2002, Wagner and Watson 2010), this contrast is prosodically disambiguated, but that this disambiguation can only be observed when the relevant clauses are embedded within a matrix clause which provides a baseline pace. Prosodic disambiguation obtains through pace modulation, with faster pace associated with the embedded/reduced relative reading and regular pace (no change) with main verb analysis. The essential contribution of the matrix sentence is to provide a baseline pace without which it is impossible to establish whether a change took place. Importantly, duration is solely determined by prosody and independent from complexity: faster pace is associated with the more complex structure.

Format: [ pdf ]
Reference: lingbuzz/003881
(please use that when you cite this article)
Published in: Proceedings of Sophia University Linguistic Society No.32, 18-34.
keywords: interface, minimalist, linearization, morphology, syntax, phonology
Format: [ pdf ]
Reference: lingbuzz/003868
(please use that when you cite this article)
Published in: Speech Prosody 2018
keywords: syntax-prosody interface, complexity, garden path, pace, embedding vs. sisterhood, syntax, phonology

Shaw & Kawahara (2018) – Consequences of High Vowel Deletion for Syllabification in Japanese

Consequences of High Vowel Deletion for Syllabification in Japanese
Jason Shaw, Shigeto Kawahara
direct link: http://ling.auf.net/lingbuzz/003854
January 2018
The EMA study reported in Shaw and Kawahara (2018) showed that Japanese /u/ optionally deletes in devoicing environments, yielding consonant clusters. This paper presents both phonological and phonetic evidence which suggests that these consonant clusters are parsed heterosyllabically. The current results imply a rather surprising conclusion that Japanese allows consonantal syllables headed by a fricative or an affricate, a conclusion that is especially surprising in light of the view that considers Japanese a “strict CV-language”

Format: [ pdf ]
Reference: lingbuzz/003854
(please use that when you cite this article)
Published in: Proceedings of AMP 2017
keywords: high vowel deletion, ema, syllabification, gestural timing, consonantal syllable, phonetics, phonology

Hsu (2018) – Exceptional prosodification effects revisited in Gradient Harmonic Grammar

Exceptional prosodification effects revisited in Gradient Harmonic Grammar
Brian Hsu
direct link: http://ling.auf.net/lingbuzz/003846
January 2018
In exceptional prosodification effects, individual lexical items pattern phonologically as if they occur in a prosodic structure that is inconsistent with the regular syntax-prosody mapping. These patterns have been analyzed as cases of prosodic prespecification (Inkelas 1989, Zec 2005), where morphemes subcategorize for a (non-default) prosodic representation. This paper argues that prespecification approaches should be reconsidered, and shows that such patterns are predicted to arise without morpheme-specific prosody in a weighted constraint system with gradiently active symbols, Gradient Harmonic Grammar (Smolensky et al. 2014; Smolensky and Goldrick 2016). Exceptional prosodification effects result from the interaction of two constraint penalty manipulations with independent support: [1] scaling of constraint violations according to prosodic context (Hsu and Jesney 2016) and [2] contrastive levels of activity in underlying forms (Smolensky and Goldrick 2016). The interaction is illustrated in an analysis of the distribution of French nasal vowels and linking [n]. This approach reduces the amount of structure posited for URs, and provides new arguments for a more uniform syntax-prosody mapping.

Format: [ pdf ]
Reference: lingbuzz/003846
(please use that when you cite this article)
Published in: Submitted ms.
keywords: nasal vowels, liaison, exceptionality, syntax-prosody interface, gradient symbolic representations, harmonic grammar, phonology