We develop an extensible description logic for stating the content of optimalitytheoretic constraints in phonology, and specify a class of structures for interpreting it. The aim is a transparent formalisation of OT. We show how to state a wide range of constraints, including markedness, input–output faithfulness and base–reduplicant faithfulness. However, output–output correspondence and ‘intercandidate’ sympathy are revealed to be problematic: it is unclear that any reasonable class of structures can reconstruct their proponents’ intentions. But our contribution is positive. Proponents of both output–output correspondence and sympathy have offered alternatives that fit into the general OT picture. We show how to state these in a reasonable extension of our formalism. The problematic constraint types were developed to deal with opaque phenomena. We hope to shed new light on the debate about how to handle opacity, by subjecting some common responses to it within OT to critical investigation.

Abstract not found

This thesis is aimed at describing the phonology of Mayrinax, one of Formosan languages, and moves a step forward to observe and analyze the phonology as well as its interaction with morphology by way of Optimality Theory (OT), set forth by Prince and

This paper proposes an analysis of phonological vowel reduction according to which vowel contrasts are subject to neutralization in unstressed syllables because it is more difficult to keep vowels distinct where vowel duration is shorter. Where the durational difference between stressed and unstressed vowels is large enough, it may not be possible to realize the same number of vowel contrasts in stressed and unstressed syllables while keeping the unstressed vowels adequately distinct, in which case it is preferable to reduce the size of the unstressed vowel inventory. The analysis is formulated as a numerical model, extending Liljencrants and Lindblom’s (1972) model of vowel inventories by incorporating aspects of the prosodic and segmental contexts of vowels, and allowing for different vowel inventories in different contexts. The model contributes to our understanding of vowel reduction, providing an explanation for the observation that vowel reduction primarily neutralizes height contrasts, and demonstrates how Liljencrants and Lindblom’s modeling approach can be developed to analyze contextual restrictions on the distribution of speech sounds. 1.

The systematic positioning of phonology as an almost entirely separate discipline from phonetics by members of the Prague School in the 1930s (see e.g. Cercle Linguistique de Prague 1931) had several regrettable consequences. Considering phonological units to be psychological, not physical, attention to the detailed substance of speech seemed irrelevant; experimental phonetics was afforded a place, but that place was on the periphery, as an AUXILIARY discipline of linguistics. Largely divorced from the investigative methods of cognitive psychology, many of which arose later than structural linguistics anyway, phonologists have amassed a body of analyses and modes of enquiry rooted in traditional grammatical discourse. The ‘arbitrariness of the sign ’ may, if we are incautious, too easily lapse into a general arbitrariness of linguistic analysis. If we ask why high vowels are lowered to mid in stressed closed syllables in Chamorro (Chung 1983: 37) whereas low and mid vowels are raised when unstressed in Bulgarian (Trubetzkoy 1939 [1969]: 81; Klagstad 1958: 46; Jakobson 1962: 446), a phonological theorist may answer

• Mini-conference on Tuesday (10-1, but we should be done around 12:40) in conference room. Prepare handout for 15-minute presentation, 5 minutes of questions. • Papers due Friday (PDF by e-mail is fine). Overview: We’ll look at a bunch of phonological phenomena that show frequency effects—including a case from my own research—and consider where in our model of language those effects could reside.

• term paper: meet with me again by the end of next week (which is a Wednesday) • no problem set on this week’s material (so nothing due Fri., Nov. 23) Where we’ve been • Basic OT and basic SPE • Comparing their predictions for process application and interaction • Seeing some other rule theories (intrinsic or variable ordering, directional application...) • But two things held us back � Our theory of morphology-phonology interaction was primitive (“concatenate morphemes, then do all the phonology”) � We fixed this last week and will continue today � Certain cases of directionality, opacity, or rule-ordering paradoxes go away � Our theory of representations is still primitive (sequence of feature matrices) � We attack this next week, with better representations “below ” the segment (features) � We’ll continue in weeks 9 and 10 with better representations “above ” the segment (syllables and bigger constituents) Overview of today: A few last things about the Lexical Phonology model, then more examples 1. The complete model—can be implemented with rules or constraints