West Coast Conference on Formal Linguistics (WCCFL) (38th : 2020)

“Epenthetic” vowels are not all equal : Gradient Representation in Yokuts roots and suffixes Guekguezian, Peter; Jesney, Karen 2020-03-07

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


73804-Guekguezian_P_et_al_Epenthetic_WCCFL38_2020_slides.pdf [ 2.34MB ]
JSON: 73804-1.0389854.json
JSON-LD: 73804-1.0389854-ld.json
RDF/XML (Pretty): 73804-1.0389854-rdf.xml
RDF/JSON: 73804-1.0389854-rdf.json
Turtle: 73804-1.0389854-turtle.txt
N-Triples: 73804-1.0389854-rdf-ntriples.txt
Original Record: 73804-1.0389854-source.json
Full Text

Full Text

“Epenthetic” vowels are not all equal: Gradient Representation in Yokuts roots and suffixesPeter Guekguezian – University of RochesterKaren Jesney – Carleton UniversityWCCFL 38 • March 7, 20201This talkFollowing Zimmermann (2019): The behaviour of Yokuts ghost segments is effectively captured through Gradient Symbolic Representation (Smolensky & Goldrick 2016, Rosen 2016)New claims:This approach has consequences for the analysis of segments that are not typically thought of as “exceptional”.The class of alternating vowels is not homogeneous.2Roadmap• Ghost vowels in Yokuts and Gradient Symbolic Representations• Alternating suffix and root vowels in Yokuts• Challenges for categorical analyses• Not all alternating vowels are equal in the UR• Interactions with vowel shortening• Conclusion3Ghost segmentsAlternating segments whose:• (non-)appearance is governed by phonological considerations• quality and/or position is lexically determined  (e.g., Hyman 1985, Archangeli 1988, 1993, Zoll 1993, Zimmermann 2019)English a vs. an• appears in contexts where hiatus would otherwise exist• selection of [n] is specific to the indefinite article4Ghost vowels in Chukchansi Yokuts (1)  Ghost vowel appearance vs. non-appearanceAppearing ghosts triggered by CVX maximum syllable shape. Data from Guekguezian (2011), Adisasmito-Smith (2016), Collord (1968)5Gradient Symbolic RepresentationsIn the UR, segments can be gradiently represented (Smolensky & Goldrick2016, Rosen 2016)• Segments differ in their degree of underlying activation• Appearing ghosts surface when markedness pressures are high enough; disappearing ghosts surface except when markedness pressures are too great  (Zimmermann 2019)Intuition:  The cost of deletion / insertion is reduced for segments that some underlying activation6(2) Non-gradient analysis(3) Additional possibilities with GSR7Yokuts ghost vowels in GSRHere:    Ghost vowels have 0.3 activation in the UR(4)  (Non-)appearance of ghost vowels8Other alternating suffix vowels(5) Other alternating suffix vowels9Root vowels(6)10Categorical analysesAlternating vowels in Yokuts are present in the UR and sometimes deleted• Suffixes with alternating non-high vowels: /-al/, /-xa/ (Kuroda 1967, Kenstowicz & Kisseberth 1977, Archangeli 1991)• (Some) suffixes with alternating high vowels: /-ith/ (Collord 1968, McGrew 2015)• Roots with alternating high vowels: /lihim/  (Collord 1968, McGrew 2015)11Alternating vowels in Yokuts are absent from the UR and sometimes epenthesized• Suffixes with alternating high vowels: /-ith/, /-wiʃ-/(Kuroda 1967, Kenstowicz & Kisseberth 1977, Archangeli 1991, Zoll 1993)• Root vowels: /lihim/ (Kuroda 1967, Kenstowicz & Kisseberth 1977, Archangeli 1983, 1991, Zoll1993, Guekguezian 2011)12Epenthetic vowel positionKenstowicz & Kisseberth (1977): ∅à V[+high] / C1 __C2C3(7)13Zoll (1993)ALIGN(Morpheme, R, σ, R):  The right edge of each morpheme aligns with the right edge of a syllable (8) Partially-correct predictions of ALIGN-M14(9) Simple constraint setMAX *HIATUS ALIGN-MDEP *COMPLEXNo “all epenthesis” or “all deletion” analysis – or “some epenthesis / some deletion” analysis – is possible with the constraint set in (9) –checked with OT-Help2 (Staubs et al. 2010).These problems are not solved by:• constraint weighting alone• adding further general constraints to the inventory15Claim:  A consistent analysis of the full data is available in GSR.   This requires:• At least three degrees of activation for alternating vowelsThere is no set of weights such that only 1.0 and 0.0 underlying activations work.• Some degree of activation for alternating root vowels, even though these are largely predictable.16Relative strength of alternating vowels(10)17Weak UR vowels are needed in roots(11) 0.3 activation in roots ensures vowels surface with ghost V suffixes(12) No activation of root vowel in yields incorrect result18Strong suffixes trigger root non-alignment(13) 1.0 activation of vowel in /-in-/ triggers correct root non-alignment(14) No activation of vowel in /-in-/ yields incorrect result19No UR vowels are needed in some suffixes(15) Some alternating vowels are “purely epenthetic”(16) Epenthesis is minimized 20(20)  Minimization of epenthesisN.B. Increasing activation on this suffix leads to selection of the other form.[li.him.wi.ʃith] – attested in current fieldwork[lih.miw.ʃith] – attested in Collord 1968 and current fieldwork21Interactions with vowel shorteningKenstowicz & Kisseberth note that epenthesis bleeds vowel shortening in Yokuts.(17) Remote past Recent pasthelp /ʔaːmi0.3l/ [ʔaː.mil.tʰaʔ] [ʔam.litʰ]sew /peːwi0.3n/ [peː.win.tʰaʔ] [pew.nitʰ]It is is “cheaper” to shorten a vowel than to realize a weakly activated alternating segment if this is sufficient to meet phonotactic requirements.22(18)  Better to shorten a vowel than to realize a ghost vowel.(19)  Better to shorten a vowel than to violate *HIATUS23DiscussionEven “predictable” vowels in Chukchansi Yokuts have some degree of underlying activation.• This approach allows a consistent analysis with simple constraints“We must give up the assumption that two or more conflicting analyses cannot be simultaneously correct for a given phenomenon” (Smolensky & Goldrick 2016, citing Hankamer 1977)24Further directionsWhy are most alternating vowels [+high] in Yokuts?• Possibility:  The underlying vowel in these cases is a bare root nodeAre there additional degrees of underlying activation?• What does this mean for the set of contrasts found in the language?How do these alternating vowels relate to those in templatic forms?• Templatic forms have alternating vowels that are morphologically triggered, but in the same places as those seen here.25THANK YOU!26References• Adisasmito-Smith, Niken. 2016. Chukchansi-English Dictionary. Ms., California State University, Fresno.• Archangeli, Diana. 1983. The root CV-template as a property of the affix: evidence from Yawelmani. Natural Language & Linguistic Theory 1, 347-384.• Archangeli, Diana. 1988. Underspecification in Yawelmani Phonology and Morphology. New York & London: Garland Publishing Inc.• Archangeli, Diana. 1991. Syllabification and prosodic templates in Yawelmani. Natural Language & Linguistic Theory 9, 231-283.• Collord, Thomas. 1968. Yokuts Grammar: Chukchansi. PhD Dissertation, University of California, Berkeley.• Guekguezian, Peter. 2011. Topics in Chukchansi Yokuts Phonology and Morphology. MA Thesis, California State University, Fresno.27References• Hankamer, Jorge. 1977. Multiple analyses. in C. Li (ed.), Mechanisms of Syntactic Change, 583-607. University of Texas Press.• Hyman, Larry. 1985.  A Theory of Phonological Weight.  Dordrecht:  Foris.• Kenstowicz, Michael & Charles Kisseberth. 1977. Topics in Phonological Theory. New York, NY: Academic Press.• Kuroda, S.Y. 1967. Yawelmani Phonology. PhD Dissertation. MIT• McGrew, Heather. 2015. High Vowel Deletion in Wikchamni Geminates. MA Thesis, California State University, Fresno.• Newman, Stanley. 1944. Yokuts Language of California. New York, NY: Viking.• Noske, Rolan. 1985. Syllabification and syllable changing processes in Yawelmani. In H. van der Hulst & Smith (eds.), Advances in Nonlinear Phonology. Dordrecht: Foris.28References• Rosen, Eric. 2016. Predicting the unpredictable: Capturing the apparent semi-regularity of rendaku voicing in Japanese through Harmonic Grammar. In E. Clem, et al. (eds.), Proceedings of the Berkeley Linguistics Society 42, 235-249. • Smolensky, Paul & Matthew Goldrick. 2016. Gradient symbolic representations in grammar: The case of French liaison. Ms., Johns Hopkins University and Northwestern University. [ROA-1286].• Staubs, Robert, Michael Becker, Christopher Potts, Patrick Pratt, John J. McCarthy and Joe Pater. 2010. OT-Help 2.0. Software package. Amherst, MA: University of Massachusetts, Amherst.• Zimmermann, Eva. 2019. Gradient Symbolic Representations and the typology of ghost segments. In K. Hout et al. (eds.), Proceedings of the 2018 Annual Meeting on Phonology • Zoll, Cheryl. 1993. Directionless syllabification and ghosts in Yawelmani. Ms., University of California, Berkeley. [ROA-28].29


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items