Monthly Archives: June 2017

Vago 2017: *VV in Hungarian

Direct link: http://roa.rutgers.edu/content/article/files/1662_robert_vago_1.pdf

ROA: 1311
Title: *VV in Hungarian
Authors: Robert Vago
Comment: To appear in: Approaches to Hungarian. Vol. 15: Papers from the 2015 Leiden Conference, Harry van der Hulst & Aniko Liptak (eds), 239-251. Amsterdam: John Benjamins, 2017. This work is under copyright; the publisher should be contacted for permission to
Length: 17 pp
Abstract: In Hungarian, suffixation derives robust instances of heteromorphemic vowel sequences (V1+V2). This work reports on an investigation of the facts (as culled, inter alia, from traditional grammars; Kenesei et al. 1998; Siptar & Torkenczy 2000; Siptar 2008) and provides analyses for the data within the framework of Optimality Theory. It brings together a conspiratorial web of mechanisms to respect a *VV constraint: V1 deletion; V2 deletion; suffix allomorphy. Exceptional cases are treated in terms of constraint reranking (Gouskova 2013). The main finding is that the story of vowel sequences across suffixes, not well studied in the phonological literature, as told from the perspective of Hungarian, provides strong support for Casali?s (1997, 1998, 2011) model of hiatus resolution.
Type: Paper/tech report
Area: Area: Phonology. Keywords: Allomorphy, constraint reranking, exceptionality, hiatus, Hungarian, optimality theory, positional faithfulness, vowel deletion

 

Round 2017: Phonological exceptionality is localized to phonological elements: the argument from learnability and Yidiny word-final deletion

Direct link: http://roa.rutgers.edu/content/article/files/1668_round_1.pdf

ROA: 1313
Title: Phonological exceptionality is localized to phonological elements: the argument from learnability and Yidiny word-final deletion
Authors: Erich Round
Comment: In Bowern, Horn & Zanuttini (eds) On looking into words (and beyond): Structures, relations, analyses. pp. 59-97. Berlin: Language Science Press. http://langsci-press.org/catalog/book/151
Length: 40pp
Abstract: Phonological exceptionality is a subject of intense and productive research in phonological theory. Here it is argued that the most coherent theory of exceptionality is diacritic (e.g. Pater 2000), but diacritic not at the level of morphs but of phonological elements. Relative to indices on phonological elements, indices on morphs add computational cost for no benefit during constraint evaluation and learning; and a theory without indices on phonological elements is empirically insufficient. On the other hand, approaches which represent exceptionality by purely phonological means (e.g. Zoll 1996) are ill-suited to efficient learning. Concerns that a phonologically-indexed analysis would overgenerate (Gouskova 2012) are misplaced under realistic assumptions about the learner. Second, it is argued that successful phonological learning via Biased Constraint Demotion (Prince & Tesar 2004) will require a Morphological Coherence Principle (MCP) which operationalizes morphological analytic bias (Moreton 2008) that is needed ensure correct constraint cloning (Pater 2009). Focusing on word final deletion in Yidiny (Dixon 1977a), I show that the learning of exceptional phonological patterns is improved if we assume the existence of an MCP. Furthermore, the MCP allows the initial state of Con to be simplified, and thus shifts explanatory burden away from the representation of the grammar per se and towards the learning device, consistent Anderson’s (2008) argument, that the space of possible human grammars must be constrained not only by limits on what is cognitively representable, but by what is learnable.
Type: Paper/tech report
Area: exceptionality, grammar learning, indexed constraints, yidiny, CON, Morphological Coherence Principle

 

Potsdam post-docs

From Diamandis Gafos
A number of postdoc positions will become available at Universität Potsdam as part of a just funded new institutional grant (whose concept is here: http://www.uni-potsdam.de/sfb1287/index.html).
Interested applicants are welcome to write to the PIs (for speech, Audrey Bürki, Barbara Höhle, Adamantios Gafos).

Clickable IPA Illustrations map

From Valerie Freeman

Here’s a great new resource I’d like to pass on: a map with the locations of the languages in the Illustrations of the IPA, with links to the articles in the Journal of the IPA and links to download zips of the sound files directly from the map. https://richardbeare.github.io/marijatabain/ipa_illustrations_all.html

As own of its creators, Marija Tabain says:

We have produced a clickable map showing the Illustrations of the IPA. The map is being updated with each new issue of the Journal of the International Phonetic Association.

 

Society for Computation in Linguistics Call for Papers

The Society for Computation in Linguistics  (SCiL) invites submissions to its inaugural meeting, SCiL 2018, which will be co-located with LSA 2018 as a sister society in Salt Lake City, Utah, January 4-7, 2018. SCiL 2018 will be held jointly with a one-time workshop on “Perceptrons and Syntactic Structures at Sixty” (PSS@60) and the 2018 meeting of Cognitive Modeling in Computational Linguistics (CMCL).

We seek high-quality research on computational and mathematical approaches in any area of linguistics. There will be two submission tracks: papers and abstracts. Accepted papers and abstracts will be presented (either orally or as posters) at the SCiL 2018 meeting. Papers will be published prior to the conference in the Association for Computational Linguistics (ACL) Anthology. Authors of accepted abstracts will have the option to submit extended abstracts prior to the conference. Both papers and extended abstracts will be published online in the forthcoming open-access SCiL proceedings.

Invited Speakers

  • Jacob Andreas
  • Emily M. Bender
  • Sam Bowman
  • Chris Dyer
  • Jason Eisner
  • Robert Frank
  • Matt Goldrick
  • Sharon Goldwater
  • Paul Smolensky

Important Dates

Submission Deadline (papers and abstracts): August 1, 2017
Notification of Acceptance: mid-September, 2017
Camera Ready Papers and Abstracts Due: November 1, 2017
Conference: January 4-7, 2018

Submissions

Papers (8pp) and abstracts (2pp) are due August 1, 2017. Links to the submission site will be posted by July 1, 2017. Papers and abstracts must be anonymous and prepared in PDF format according to the following guidelines.

Papers

Paper submissions must describe original, completed, and unpublished work. They are limited to 8 content pages (plus unlimited pages for references) and should follow the two-column ACL format. Style templates are available from the NAACL website:

Accepted papers will be published in the ACL Anthology prior to the conference, and will be presented either as oral or poster presentations at SCiL 2018. Papers will also be published online in the forthcoming open-access SCiL proceedings.

Abstracts

Abstract submissions must describe original and completed work. To facilitate exchange of research ideas across disciplines, this track will consider work that has been previously presented (and potentially published) at venues with distinct scope and target audiences from SCiL. Submissions describing previously presented/published work must indicate so at submission time.

Abstracts length is limited to a maximum of two single-spaced pages (US Letter), figures and references included. Font should be 12-point Times or Times New Roman throughout, and the document should be single-spaced, left justified, with margins of exactly one inch on all sides. Title and section headings (if used) should bold.

Accepted abstracts will be presented at SCiL 2018 as posters or oral presentations. Authors of accepted abstracts will have the option to extend abstracts up to 4 pages (extended abstracts). Abstracts and extended abstracts will be published before the conference in the forthcoming open-access SCiL Proceedings.

Organization

Co-Chairs

Advisory Committee

  • Emily Bender, University of Washington
  • Leon Bergen, UCSD
  • Jennifer Culbertson, University of Edinburgh
  • Naomi Feldman, University of Maryland
  • Tim Hunter, UCLA
  • Roger Levy, MIT
  • Giorgio Magri, CNRS and University of Paris 8
  • Brendan O’Connor, UMass Amherst
  • Christopher Potts, Stanford University