Handout #7Background to Myers (1998):Underspecification in phonetics andphonologyEquipollent vs. Privative• Trubetskoy (1939) classified some phonologicalcontrasts as equipollent, meaning the contrastingcategories have opposite properties.• But other contrasts he classified as privative: acontrast between presence vs. absence of aproperty.• He also argued that when a contrast is neutralizedin some context, the result is an archiphoneme,belonging to neither of the contrasting classes.Underspecification in GenerativePhonology• An underspecified representation is one in whichsome elements are not specified for all of thefeatures.• Halle (1959) used such representations in hisinfluential phonology of Russian.• But Stanley (1967) argued that such a moveallowed for an unattested three-way contrastbetween +F, -F, and øF for any feature F.• Chomsky and Halle (1968) thus assumed that allphonological representations are fully specifiedfor all features.Underspecification in GenerativePhonology• Phonological arguments for underspecification inunderlying representations were given by Hooper(1975), and Pulleyblank (1986, 1988).• Some phonologists argued that all redundant(predictible) feature specifications should beeliminated from underlying representations, andinserted by rule.• Kiparsky (1981) argued that phonologicalcontrasts were all privative, in all levels ofrepresentation.Underspecification in OptimalityTheory• Underlying representation plays little role inOptimality Theory, so the question of whethersuch representations are underspecified as fadedaway (but see Inkelas 1994).• Redundancy of feature specifications can beexpressed through markedness constraints(Smolensky 1993).• Some work in OT therefore assumes that allcandidates and all outputs are fully specified (e.g.Smolensky 1993, Beckman 1997).Underspecification in phonetics• In contrast, phoneticians have always assumedthat representations were underspecified.• Not every segment has a target for everyarticulator.• During a bilabial stop, for example, there is nospecific goal for the position of the tongue.• The position of an articulator during unspecifiedstretches relapses toward a default position,subject to transitions out of and into neighboringtargets (Öhman 1966, Browman and Goldstein1990).Underspecification in phonetics• This long-held assumption has in more recentwork been supported by experimental work.• Pierrehumbert (1980) and Pierrehumbert andBeckman (1998) argued that the representation ofintonation is sparse, with long stretches ofunspecified syllables.• Keating (1988) argued that /h/ can have nospecification for place of articulation.• The scope of underspecification in surfacerepresentation is currently not fully understood.References• Beckman, J. (1997). Positional faithfulness, positional neutralizationand Shona vowel harmony. Phonology 14, 1-47.• Browman, C. and L. Goldstein (1990). Tiers in ArticulatoryPhonology, with some implications for casual speech. In J. Kingstonand M. Beckman (eds.) Papers in Laboratory Phonology I. CambridgeUniversity Press, Cambridge, 341-376.• Chomsky, N. and Halle, M. (1968). The Sound Pattern of English.Harper and Row, New York.• Halle, M. (1959). The Sound Pattern of Russian. Mouton, 'SGravenhage.• Hooper, J. (1975). The archisegment in Natural Generative Phonology.Language 51. 536-560.• Inkelas, S. (1994). The consequences of optimization forunderspecification. Rutgers Optimality Archive.• Keating, P. (1988). Underspecification in phonetics. Phonology 5, 275-292.References• Kiparsky, P. (1981). Vowel harmony. Unpublished ms., MIT,Cambridge.• Öhman, S. (1966). Coarticulation in VCV utterances: Spectrographicmeasurements. JASA 39, 151-168.• Pierrehumbert, J. (1980). The Phonetics and Phonology of EnglishIntonation. Distributed by IULC, Bloomington.• Pierrehumbert, J. and M. Beckman (1988). Japanese Tone Structure.MIT Press, Cambridge.• Pulleyblank, D. (1986). Tone in Lexical Phonology. Reidel, Dordrecht.• Pulleyblank, D. (1988). Vocalic Underspecification in Yoruba.Linguistic Inquiry 19. 233-270.• Stanley, R. (1967). Redundancy rules in phonology. Language 43.393-436.• Trubetskoy, N. S. (1939). Grundzüge der Phonologie, Travaux deCercle Linguistique de Prague 7,