Derived Environments and Consonant Clusters: The …cunyphonologyforum.ws.gc.cuny.edu/files/2016/01/Pillion-Handout... · 14/01/2016 · Derived Environments and Consonant Clusters:
Post on 04-Sep-2018
220 Views
Preview:
Transcript
Derived Environments and Consonant Clusters: TheCase of Lezgi
Betsy Pillion
University of Chicago
bpillion@uchicago.edu
14 January 2016
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 1 / 44
Introduction
Word inital clusters in Lezgi arise in two environmentsmonomorphemicmultimorphemic
Monomorphemic and multimorphemic words allow for similarsequences
Multimorphemic words have a higher incidence of stop-stop sequencesMonomorphemic words with these clusters are less common
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 2 / 44
Introduction
What could be responsible for this tendency?
Choapan Zapotec has been proposed to allow for more SSP-violatingclusters across morpheme boundaries (Donnelly, 2013)
Lezgi has been proposed to possess voiceless vowels (Chitoran andIskarous, 2008)
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 3 / 44
Introduction
Some multimorphemic words have been proposed to possess voicelessvowels
Do voiceless vowels exist in these environments?What kind of status do voiceless vowels have in sonority sequencing?
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 4 / 44
Introduction
Sound change where high vowels in pretonic syllables delete betweentwo voiceless obstruents (Haspelmath, 1993)
Historical process is similar to phenomenon seen in morphologicalalternation
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 5 / 44
Data
Derived Non-Derived
Singular Plural gloss Historical Current gloss
ki>tS khj>tS-er ‘dog’ kitab khtab ‘book’
q’yd qhwjt’-er ‘winter’>tSi
>tSeg
>tS
>tSeg ‘onion’
thyt thwj-ter ‘throat’>tSika
>tSka ‘place’
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 6 / 44
Overview of Phenomenon
Haspelmath (1993) describes it as a pattern of vowel syncope
Chitoran and Iskarous (2008) claim vocalic gesture is still present
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 7 / 44
Overview
Questions
Lezgi background
Proposals
Experiment
Derived vs. Non-Derived Patterns
Conclusions
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 8 / 44
Research Questions
What is the status of historically created vs. morphologicallyprompted clusters in Lezgi? (Haspelmath, 1993; Chitoran andBabaliyeva, 2007; Chitoran and Iskarous, 2008)
Do sequences in derived environments allow for less well-formedconsonant sequences than sequences in non-derived environments?(Donnelly, 2013)
What status do voiceless vowels have in the language if any? Is thisphenomenon gradient or categorical? Do they contribute to theallowance of these sequences?
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 9 / 44
Background
Lezgi is a Northeast Caucasian language spoken by about 800,000people (Ethnologue, 2015)
Spoken in Daghestan, Russia, and Azerbaijan
Speaker from northern Azerbaijan
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 10 / 44
Background
Consonant Inventory
Vowel Inventory: i, y, u, e, a
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 11 / 44
Derived Environments
singular �! plural creates environment for this phenomenon
Singular Plural gloss
thyt thwjter ‘throat’q’yd qhwjt’er ‘winter’
ki>tS khj>tSer ‘dog’
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 12 / 44
Acoustic Data - Derived Environments
[thyt] ‘throat’
[thwjter] ‘throats’
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 13 / 44
Non-Derived Environments
some historically disyllabic words now begin with complex clusters
Historical Modern gloss
kitab khtab ‘book’>tSi
>tSeg
>tS
>tSeg ‘onion’
>tSika
>tSka ‘place’
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 14 / 44
Proposals
Vocalic gesture bundle is deleted completely
Vocalic gesture bundle remains but voicing gesture is realized underparticular conditions
Vocalic gesture bundle remains but voicing gesture is never realized
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 15 / 44
Proposals
Vocalic gesture bundle is deleted completelyMeasure spectral energy of consonant preceding vowel
Vocalic gesture bundle remains but voicing gesture is realized underparticular conditions
Vocalic gesture bundle remains but voicing gesture is never realizedCan high vowels surface as voiced in derived environments at slowspeech rates?Can low vowels devoice or delete in derived environments at fast speechrates?
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 16 / 44
Previous Research
Previous looks at vowel devoicing due to gestural overlap havefocused on optional processes.
Tokyo Japanese (Beckman and Shoji 1984)Montreal French (Cedergren and Simoneau 1985)English schwa elision (Davidson 2006)
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 17 / 44
Previous Research
Chitoran and Iskarous (2008) look at spectral tilt in sibilant-initialstems in Lezgi.
Groupings of spectral energy in sibilant noise indicate that there aresignificant di↵erences between stems with “deleted” y, u, and i.
Chitoran and Babliyeva (2007) examine stress’s e↵ects on vowelduration in Lezgi
“For [pre-tonic] non-high vowels, their gestures are also overlapped, butbecause they are longer they are not completely hidden.”
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 18 / 44
Phase windows (Byrd 1996)Faster speech rates are expected to have higher rates of gesturaloverlap, and slower speech rates are expected have lower rates ofgestural overlap.
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 19 / 44
Experimental Design
Manipulation of speech rate
Increased and decreased rate of gestural overlap to allow for variationin voicing
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 20 / 44
Experiment
Data collection focused on di↵erences between singular and plural(bimorphemic)
Words collected over two sessions, two days, approximately a monthapart.
Carrier sentences‘You say X again.’ vna X sadra laXana
Three conditions: normal rate, slow rate, fast rate
Six total blocks, 3 blocks per day
Written prompting in Lezgi, with test words in Russian
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 21 / 44
Data Collection
Noun stems of form CVC(-Vr)
C1VC2 where...C1 is voiceless stop or voiceless fricativeC2 is not controlled forV is high, mid, or low
35 possible consonant-vowel combinations were searched, while 8other pairs lacked a singular or plural form
Total of 60 words, collected 4 times in each speech condition, 720total tokens
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 22 / 44
Data Collection
Stop series P, T, K, Q include plain, aspirated and ejective stops
Fricative series s, S and X
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 23 / 44
Acoustic Analysis
Sibilant’s center for gravity and peak frequency (Chitoran andIskarous 2008)
Script averages spectra across a 40ms period for 10 points throughoutthe period of frication, measures used taken from 80% through thefrication.
Voicing measured categorically and manually by the researcher
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 24 / 44
Results - Presence of Voiceless Vowels
A linear models constructed to establish e↵ects on center of gravity infricatives
If vowel identity in singular form of word, or C1*V interaction is apredictor of CoG, then this is indicative of voiceless vowels
If V*number interaction is a predictor of CoG, then this is notconsistent with voiceless vowels
Fricative model - lm(CoG8 - C1xVowel+Number+VowelxNumber)
Deviation coded
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 25 / 44
Results - Fricative Model
Fricative model - lm(CoG8 - C1xVowel+Number+VowelxNumber)
S, X and s are significant predictors
Only C:V pairs X:e, S:i are significant predictors
Importantly, no interactions between number and vowel reachsignificance
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 26 / 44
Results - [s]
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 27 / 44
Results - [S]
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 28 / 44
Results - [X]
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 29 / 44
Results - Fricatives
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 30 / 44
Results - Speech Rate
None of the lexical items varied in voicing
If devoicing were a gradient phenomenon, we would expect higherrates of devoicing at fast speeds, and lower rates of devoicing at slowspeeds
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 31 / 44
Results - Speech Rate
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 32 / 44
Discussion
Evidence is consistent with Chitoran and Iskarous (2008)’s conclusions
There is no variation with respect to voicing presence, indicating thatthis phenomenon is obligatory
Voiceless vowels may vary with deletion
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 33 / 44
Sequencing in Derived and Non-Derived environments
Data taken from Haspelmath (1993)
Both environments allow for similar sonority declinations (tSka,‘place’; Xtul, ‘grandchild’ vs. Xu
˚kwar, ‘stomachs’, thu
˚p’ar, ‘fingers’)
Stop-stop sequences are more common across derived environmentsThe presence of voiceless vowels accounts for this tendency, as theymay allow these sequences to maintain independent syllabic status
Non-derived environments are less likely to possess secondaryarticulation on initial consonants
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 34 / 44
Consonant clusters in Non-Derived Environments
Survey of Haspelmath (1993) dictionary results in 49 monomorphemicwords with CC initial clusters
Only 14/49 of clusters possess secondary articulation on C1
Only 12/49 of non-derived clusters are stop-stop (8 with SA),compared to 7/13 derived clusters (all with SA)51 words begin with Cw (secondary articulation)
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 35 / 44
Consonant clusters in Non-Derived Environments
Secondary Articulation in Non-Derived EnvironmentsNo SA SA
word gloss word gloss
khsun ‘fall asleep’ khjse ‘purse’khtab ‘book’ khwjtæhun ‘finish, use up’
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 36 / 44
Sequencing in Derived and Non-Derived Environments
Similar sequences appear across both
Sequence Non-Der gloss Der gloss
stop-stop thwp’al ‘ring’ thu˚p’-ar ‘fingers’
fricative-stop Xtul ‘grandchild’ Xu˚kw-ar ‘stomachs’
fricative-stop st’al ‘drop’ si˚k’-er ‘foxes’
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 37 / 44
Conclusions
Evidence is consistent with the presence of voiceless vowels in derivedenvironments in the language
The presence of voiceless vowels/vocalic gesture in derivedenvironments may contribute to the allowance of sequences thatseemingly decrease in sonority across both derived and non-derivedenvironments
Syllabic status of derived environment and non-derived environmentclusters purported to have voiceless vowels and/or secondaryarticulations on the initial consonant is still unknown
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 38 / 44
Future Directions
Collecting syllabicity judgments
Timing relations across monomorphemic and bimorphemic words
Larger lexicon sample to better reflect general trends
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 39 / 44
References
Beckman, Mary, and Atsuko Shoji. 1984. ”Spectral and perceptual evidence for CV coarticulation in devoiced/si/and/syu/in Japanese.” Phonetica 41.2:61-71.
Butskhrikidze, Marika. 2002. The consonant phonotactics of Georgian. Vol. 63. LOT.
Cedergren, H. , Simoneau, L. 1985. La chute des voyelles hautes en francais de Montreal, “As-tu entendu la belle syncope?” In Les tendances dynamiques
du francais parle a Montreal, Quebec, Gouvernement du Quebec, coll. Langues et societes,. Vol. 2: 25-56.
Chitoran, Ioana, and Babaliyeva, Ayten. 2007. An acoustic description of high vowel syncope in Lezgian. ICPhS 16: 2153-2156.
Chitoran, Ioana, and Goldstein, Louis, and Byrd, Dani. 2002. Gestural overlap and recoverability: Articulatory evidence from Georgian. LaboratoryPhonology 7. Berlin: Mouton de Gruyter. 419-448.
Chitoran, Ioana, and Khalil Iskarous. 2008. ”Acoustic Evidence for High Vowel Devoicing in Lezgi.”
Cho, Young-mee Yu. 1993. The Phonology and Phonetics of Voiceless Vowels. Proceedings of the Nineteenth Annual Meeting of the Berkeley LinguisticsSociety: General Session and Parasession on Semantic Typology and Semantic Universals. 64-75.
Clements, G. N. 1990. The role of the sonority cycle in core syllabification. In J. Kingston and M. E. Beckman (eds.) Papers in Laboratory Phonology I:Between the grammar and the physics of speech. Cambridge: Cambridge University Press. 283-333.
Davidson, Lisa. 2006. Schwa Elision in Fast Speech: Segmental Deletion or Gestural Overlap? Phonetica 63: 79-112.
Davidson, Lisa. 2003. ”The atoms of phonological representation: Gestures, coordination and perceptual features in consonant cluster phonotactics.”PhD diss., Johns Hopkins University.
Davidson, Lisa and Jason A. Shaw. 2012. Sources of illusion in consonant cluster perception. Journal of Phonetics 40: 234-248.
Donnelly, Erin. ”The sonority hierarchy in Choapan Zapotec.” Handout from the International Con- ference on Mesoamerican Linguistics (2013).
Goldstein, Louis, Ioana Chitoran, and Elisabeth Selkirk. 2007. ”Syllable structure as coupled oscillator modes: evidence from Georgian vs. TashlhiytBerber.” Proceedings of the XVIth International Congress of Phonetic Sciences.
Haspelmath, Martin. 1993. A grammar of Lezgian. Walter de Gruyter.
Kim, Kyumin and Alexei Kochetov. Phonology and phonetics of epenthetic vowels in loanwords: Experimental evidence from Korean. Lingua 121:511-532.
Lewis, M. Paul, Gary F. Simons, and Charles D. Fennig (eds.). 2015. Ethnologue: Languages of the World, Eighteenth edition. Dallas, Texas: SILInternational.
Ridouane, Rachid. 2008. ”Syllables without vowels: phonetic and phonological evidence from Tashlhiyt Berber.” Phonology 25.02: 321-359.
Selkirk, E.O. ”On the major class features and syllable theory.” In M. Arono↵, RT Oehrle (Eds.), Language sound structure, MIT Press, Cambridge, MA(1984).
Rose, Sharon. 1994. The Historical Development of Secondary Articulation in Gurage. Proceedings of the Twentieth Annual Meeting of the BerkeleyLinguistics Society: Special Session on Historical Issues in African Linguistics. 112-124.
Yu, Alan C. L. 2004. Explaining Final Obstruent Voicing in Lezgian: Phonetics and History. Language, Vol. 80 : 73-97.Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 40 / 44
Secondarily Articulated Obstruents
secondarily articulated onset consonants cause unique patterns insyncope contexts
Singular Plural gloss
k’wa>tS k’u
>tSar ‘leg’
>tswez >tsu>tswer ‘ant’khwak khukwar ‘worm’
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 41 / 44
Speech Rate - Linear Model
lm(Voicing - VowelHeight + SpeedCondition + C2Voicing +VowelHeight*SpeedCondition)
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 42 / 44
Fricative Model
Fricative model - lm(CoG8 - C1xVowel+Number+VowelxNumber)
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 43 / 44
Stop Model
Stop model - lm(CoG8 - C1xVowel+Number+VowelxNumber)
Betsy Pillion (UChicago) Lezgi Clusters 14 January 2016 44 / 44
top related