grad program
undergrad program
courses
faculty
grad students
recent graduates
underling
student resources
spring colloquium
grad program
undergrad program
courses
faculty
grad students
recent graduates
underling
student resources
spring colloquium
Serial Harmonic Grammar
Prince and Smolensky (1993/2004: ch. 2) consider two versions of OT, one in which the entire representation is evaluated only once (parallel OT), and one in which the representation is changed and evaluated iteratively (serial OT). Though they (and most subsequent studies in OT), adopt the parallel version, McCarthy (2006 et seq.) has recently offered a series of arguments for serial OT. In this talk, I show that the choice of serialism also has consequences for a second basic architectural decision in Prince and Smolensky (1993/2004): the choice of constraint ranking over the constraint weighting of OT~Rs predecessor, Harmonic Grammar (HG; Legendre, Miyata and Smolensky 1990, Smolensky and Legendre 2006, Pater, Bhatt and Potts 2007). Prince and Smolensky (1993/2004: 232-233) suggest that a version of OT with weighted constraints would generate unattested languages that are not produced by ranking. I show that weighted constraints do indeed generate implausible patterns in parallel HG, but that these problems are resolved in serial HG. I further show that weighting leads to desirable simplifications in the formulation of constraints.
Traditionally, reference grammars represent the language as a
homogeneous entity, ignoring variation in favor of conciseness.
Sociolinguists focus on linguistic variation and its correlation to
culturally relevant distinctions (e.g., sex, age, attitude), and we
are interested in the issue of representing such variation in the
grammar. When a sociolinguist is confronted with the task of writing
a grammar, the traditional methods of homogeneous grammar preparation
clash with sociolinguistic goals. I discuss documentation of a
grammar of Faetar as a case in point.
Faetar is an endangered Francoproven?al dialect spoken by ~500
residents of a village in southern Italy. Speakers have been
predicting Faetar's demise for decades and lamenting its "imperfect
form" (due to contact with Italian). As part of their effort to fend
off the perceived death of Faetar, speakers encouraged me to write a
book describing their language. The linguistic challenge was to
efficiently and effectively codify this non-written language without
misrepresenting the facts of language change and variation. I will
describe my efforts to write a sociolinguistically accurate grammar
and problems encountered along the way.
Variations in Sonority Rankings: Gliding of Fricatives
Lisa Domby, University of North Carolina at Chapel Hill
Selected speech production data collected from female dizygotic (fraternal) twins at ages 3;3 and 3;5 will be analyzed using phonological constraint rankings proposed by optimality theory. Optimality theoretical analysis can explain child innovations such as the atypical error pattern of gliding of fricatives exhibited in the output of these twins. The sonority rankings expected in typically developing children and adults will be explained. Variations in rankings of constraints that govern sonority will explain why these twins glide the fricative /f/ to [w], and why their grammars select [w] as the output candidate that violates the least number of constraints. The findings of the optimality theoretical analysis will support the proposal that a constaint ranking system consistent with optimality theory offers an explanation for the twins' unusual output, gliding of fricatives.
Plural Markings in Classifier Languages
Jaeshil Kim, Liberty University (VA)
We investigate how the definiteness and specificity play roles in the parameterization of plural markings in classifier languages such as Chinese, Japanese, and Korean. Plural markers in classifier languages such as Chinese -men, Japanese -tachi, and Korean ~Vtul mark more than the plurality of referents of the nouns they attach to, unlike English plural marker ~V(e)s. It will be shown that they mark not only plurality but also definiteness/specificity of their base nouns. We also propose that there exists a demarcating feature which splits the requirement of obligatory plural marking in these languages and such semantic features are subject to parameterization: [?definite] in Chinese and Japanese and [?specific] in Korean. An interesting parallel emerges between these article-less languages and the article languages with respect to marking definiteness and specificity.
Generic Singular Definite Nominal Phrases in Spanish: A Notional Account
Carlos Molina-Vital, Duke University (NC)
In this presentation it will be argued that the notional definition of noun is relevant to the determination (grounding) process of the nominal phrase (nominals). In particular, it will be shown that the division of a class in instances (if made of countable exemplars) or parts (if a non-replicable mass), affects crucially the use of definite generic singular nominals. It has been argued in several places that the generic meaning of nominals such as El leon es un animal salvaje ('The lion is a wild beast') or El hombre no debe matar a otro hombre ('The man must not kill another man') is possible due to the existence of an inherently abstract type space or universe of types where the they are the instances referred and pointed by the definite article. Even though I consider this idea to be basically correct, it seems to propose the nature of the type space or the types themselves without motivating them in more basic conceptual constructions. In other words, the universe of types is taken for granted too easily. I will propose that Langacker's intuition about the similarity between the type space and quality space (the cognitive domain that equates all different manifestations of the same substance due to their qualitative identity) goes beyond the similarity: the type space is a quality space created to equate different instances that are easily conceived as identical to one another. This is demonstrated through the analysis of some definite singular nominals that should be interpreted generically if they referred to and abstract type, but fail to deliver that meaning: El mueble es mas grande que la joya. The conceptual structure of the nouns mueble ('furniture') and joya ('jewel') create a class where the similarity of members is hard to stablish, which produces the anomaly of the generic meaning for singular definite nominals.
Teaching Spanish Phonetics: A Pedagogical Approach Based on Markedness
Kara Moranski, Temple University (PA)
The objective of this study is to examine the effectiveness of a model for explicit phonetic instruction grounded in the notion of phonetic markedness. Over the course of six weeks, participants from two Spanish classes at a major American university received supplemental phonetic instruction in the pronunciation of several Spanish phones. The first experimental group received instruction focused on phones considered to be highly marked for native speakers of American English, while the second group~Rs instruction gave equal emphasis to phones considered to be both marked and unmarked for these speakers. Speech samples from the participants were digitally recorded both pre and post treatment. Tokens from these recordings were then acoustically analyzed in the PRAAT speech analysis program, measuring both voice onset time and degree of occlusion for the critical phones to determine whether explicit phonetic instruction yields articulatory accuracy for the L2 learner as a function of markedness.
Nascent Futures in Hungarian
Peter Sherwood, University of North Carolina at Chapel Hill
Recent grammars of Hungarian devote little attention to the category of "future". The Kiss-Kiefer-Sipter oj magyar nyelvtan (New Hungarian Grammar, 1998), for example, intended as an up-to-date summary primarily for schoolteachers (p. 13), makes only the negative statement that "The future in Hungarian is not a morphological category", pointing out that the verb FOG used to express it (in fact, this is used mainly for rather firm futures) is an auxiliary operating with an infinitive, and contrasts this with the past conditional (formally: past tense followed by invariable VOLNA) , which is asserted to be "a morphological word" (p. 210). This paper argues for two hitherto unrecognized ways of expressing "future~", by cliticization to the verb of the adverbial element MAJD: preverbally with the sense "subsequently (but not now)", and postverbally (thus formally paralleling the past conditional) with the sense "in due course".
Discussion on the Use of the Cooperative Principle and the Politeness Principle in Chinese
Chunmei Yu, University of Virginia
It is widely acknowledged that the Cooperative Principle (CP) (Grice 1975)
and the Politeness Principle (PP) (Brown & Levinson 1987, Leech 1983)
interact and are universal. However, the relative "weights" of each
principle vary among cultures. How do they do so and why? This paper offers
a case study of the interaction of Grice's CP and Brown and Levinson's PP in
Chinese. It also discusses how Brown and Levinson's Face Threatening Acts
(FTA) are related to the PP in Chinese cultural contexts by offering an
account of how and why politeness values and face semantics in Chinese
differ from those in Anglo-American culture. The paper reveals that Chinese
try to balance using CP and PP, but sometimes give more weight to certain
maxims in these principles than Westerners. This paper analyzes the reasons
for the phenomena and explains the differences between Western and Chinese
violation of these principles.
(NB: Subject to update - check back later!)
Programs from past years (pdf files)
2008
2007
2006
2005
2004
2003
2002
The Spring Colloquium is sponsored in part by the Graduate and Professional Student Federation of UNC.