Fall 2009

"Sonority and the Acquisition of #sC Onsets in Spanish-English Bilingual Children with Phonological Disorders." Dr. Mehmet Yavas, Florida International University. November 18, 2009.

This presentation looks at the development of English #sC onsets in Spanish-English bilingual children with phonological disorders. The findings are discussed in light of the following questions:

Do the data reveal differential treatment of #sC onsets with repect to C2?

Is a binary split between SSP-following and SSP-violating onsets a reasonable explanation for the ease/difficulty of the target onsets? If not, do ‘s+nasal’ onsets behave similar to ‘s+stop’ onsets in the easy/difficult continuum?

How do the findings compare with the ones reported for monolingual English speaking children with phonological disorders and typically developing bilingual children?

"Verb-Particle Errors." Dr. Ellen Thompson, Florida International University. October 21, 2009.

Within the Generative framework of linguistics, "...evidence concerning production, recognition, recall, and language use in general can ... have bearing on the investigation of rules of grammar, on ... grammatical competence or knowledge of language" (Chomsky 1980: 200-201). Current approaches to speech error research argue that …”these errors are constrained to those which are statable in terms of the linguistic system" (Fromkin 1988). This talk will examine how speech error data argues in favor of a particular analysis of a construction historically associated with competing analyses. The pattern of observed errors in the production of the verb- particle construction supports an analysis according to which the particle moves, and argues against alternative analyses, according to which the construction is derived through verb and/or direct object movement.

"Pronominal Feature Specification and Diachronic Change: The Case of Dominican Spanish." Dr. Cristina Martínez-Sanz, Florida International University. September 23, 2009.

Dominican Spanish presents itself as a perfect ‘laboratory’ for the exploration of what the locus of cross-dialectal or historical variation may be located in a minimalist framework This variety seems to reflect typological distinctions with regards to null and overt subject distribution. This talk will explore an alternative account. Taking Camacho’s (2008) proposal as a point of departure, it will be proposed that the source of the Dominican variation in subject expression is to be found in the feature specification of pronominals. It will be suggested that this feature specification is responsible for the shift in referential meaning of pronominal forms, and generally for their ability to license EPP, the shifts in their capacity to bear reference, and their ability to stand as prosodical independent units. This talk will also review the syntactic phenomena that arise from this reformulation of the pronominal feature specification, such as the specific clefted focus constructions that have been typically described for Dominican Spanish.

Spring 2009

"Pronominal feature specification and diachronic change: The case of Dominican Spanish." Cristina Martinez-Sanz, Flodi September 23, 2009.

The aim of the talk is to show how several French -ment adverbs function both as intrapredicative and as extrapredicative adverbs. In this case, they are discourse markers with a contrastive meaning, close to that of the conjunction mais. To account for this fact, the talk will argue for the grammaticalization of these adverbs along a double cline, one which was originally proposed by Meillet (1916): lexicon > grammar, and a much more recent one, proposed by Traugott (1995): propositional > textual > expressive. Since grammaticalization is a diachronic process, it is by definition a gradual phenomenon. Therefore one can expect some of the adverbs to be more grammaticalized than others.

"Parsing English Phrasal Verbs." Peter A. Machonis, Florida International University. February 5, 2009.

In Natural Language Processing, there has been much recent attention paid to multiword expressions, and in particular phrasal verbs. This paper presents an attempt at parsing English phrasal verbs with and without insertion using manually constructed Lexicon-Grammar tables, along with NooJ, a powerful linguistic tool where formalized linguistics descriptions can be applied to large corpora in real time. For the moment, we have achieved an overall accuracy of 84%, which includes many discontinuous instances of phrasal verbs, such as show the gentleman up. We will also highlight some of the problems and ambiguities encountered with parsing phrasal verbs and examine ways to improve recall. This preliminary investigation shows that NooJ, along with exhaustive Lexicon-Grammar tables, can help solve a key problem in Natural Language Processing.