This presentation details a new approach to morpheme classification developed by the author and Janice L. Jake.  The model also is applicable to monolingual data, such as speech errors and data from Broca's aphasics.  The model emphasizes the notion that lemmas underlying different types of morphemes become salient at different levels of production.  The model offers indirect evidence for how language production works and how competence and performance are linked. Under this model the classification of morphemes is not basic; rather, the mechanisms for activating and combining morphemes are basic.  The interaction between conceptual information and complex grammatical structure in any entry in the mental lexicon and the rules of grammar in the formulator are what give any reality to the notion of morpheme.

(The model has been applied to various language contact phenomena by PhD students in the USC Linguistics Program, including Longxing Wei, Janet Fuller, Steven Gross, Elena Schmitt, and Sue Jenkins, among others.)

This paper seeks to shed new light on conditions governing extraction from NPs. It will show, contrary to general wisdom, that extraction out of NP objects is generally disallowed (even when definiteness effects are accounted for).  Rather, extraction from object NPs is found to be limited to NPs which project event (or argument) structures (in the sense of Grimshaw 1990 or Pustejovsky (1991).  The paper will show that: (i)  NPs lacking argument structure do not permit wh extraction at all, and (ii) NPs having agentive structure permit extraction even when they are definite.  The contrast in (1) illustrates the former claim, and the data in (2) support the latter.

 (1)  a.  What did Jake take a picture of?  (answer:  The Alamo)
        b.*What did Jake write a letter of?  (answer:  apology)
 (2) Which president did John Grisham write/finish/*read/*buy that book about?

Extraction in (1a) is licensed due to the activation of the telic structure (use and function) of metaphysical picture.  In (2), when a main verb such as write or finish activates the noun's agentive structure, wh extraction becomes possible even out of the definite NP that book about.  In both instances, extraction out of the NP is licensed by the activation of the "qualia structure" of the noun (Pustejovsky 1991 and 1995, Jackendoff 1997).   [This talk is based on collaborative work with William D. Davies, University of Iowa].

Field-initiated, community-based studies offer one of the greatest resources for addressing the most fundamental issues in language variation and change, yet their potential is surprisingly underutilized in sociolinguistics in general and dialectology in particular. What are the linguistic and sociolinguistic "issues" that make for a good community-based study? How do researchers practically set up and carry out such a study? How can a community itself be involved in the process? This workshop considers the theoretical, methodological, and practical issues involved in conducting field-initiated studies, based on extensive field experience in different sites in North Carolina in the last decade. A special emphasis includes the discussion and demonstration of applications of the linguistic gratuity principle in which researchers work with communities to return linguistic favors. A practical goal of the workshop is the encouragement of more extensive community-based sociolinguistic projects in South Carolina and the Southeast US.

A speaker's choice of referring expressions in discourse, including pronouns, names, lexical NPs, and other NP types, is a very complex phenomenon, especially given the fact that structural, semantic, pragmatic, and cognitive factors may affect both the use and felicitous interpretation of these expressions. This study focuses on the problematic issue of when and why a Spanish speaker uses a 'marked' (more effortful, complex, prolix) overt pronoun in place of an 'unmarked' (less effortful, simpler, more minimal) null pronoun. Specifically, I analyze Spanish narrative data, in order to identify the factors that determine native Spanish speakers' use of overt subject pronouns as opposed to null subjects in light of two current theoretical models of discourse anaphora. These models are known as the cognitive model (e.g. Gundel, Hedberg and Zacharski 1993; Freithem and Gundel 1996) and the neo-Gricean pragmatic model (e.g. Huang 1991, 1994, 2000; Levinson 1987a, b; 1991).

The cognitive model makes use of notions such as focus of attention, the activation of a referent in memory, and the cognitive status associated with the various referring expressions, in order to account for a speaker's use of referring expressions. On the other hand, the pragmatic model attempts to account for both choice and interpretation of NP types, including pronouns and zeros, via a set of neo-Gricean speaker and hearer principles, which interact to create a tension among three communication  strategies, informativeness, minimization, and the use of marked linguistic forms. In my analysis, I show that the cognitive model alone cannot explain instances of overt pronoun use in Spanish, since these instances appear to be motivated by pragmatic reasons, which are taken into account in the pragmatic model, and not solely by the cognitive statuses of the referents in question.

For example, according to Gundel, Hedberg and Zacharski (1993), both null and overt yet unstressed subject pronouns in Spanish should fulfill the same condition of 'in focus', i.e., the referents of these referring expressions must be in short-term memory and also at the current center of attention. However, in the following example from my data, the cognitive status of 'in focus' does not provide an explanatory or descriptive account for why the speaker uses an overt subject pronoun after a null one to refer to the same referent, where a null pronoun would have been referentially successful (the narrator is referred to by
number (2 in this case); L refers to the 'listener'):

(1)  2: =Entonces se ha quedado extrañao de que venían comiendo una pera. No habían llegao a las peras pero ellos venían comiendo //una pera.
L:  Ya venían comiéndose una pera.
2:  Ellos venían comiéndose una pera en la mano. Y: y: y así lo he comprendido yo. No lo sé.

 2: '=Then he was surprised that they were coming along eating a pear.
They hadn't gotten to the pears but théy were coming along eating// a pear.
 L: 'They were already coming along eating a pear.
 2: 'Théy were coming along eating a pear in their hand. A:nd: a:nd that's what Í understood. I don't know.

On the other hand, the cognitive status of 'in focus' is not fulfilled in the following case, where the overt subject pronoun él ('he') is used at a point in the same speaker's narrative where the referent (a boy with a bicycle) is not the current center of attention:

(2) 2: *Pero que también los chicos, ha sido muy bonito para mí eran unos vagamundos (.) pero muy educaos. Ahora si él ha robao, un tío muy mal educao*

2: '*But also the boys, it was really beautiful for me they were a bunch of vagabonds (.) but very polite. Now if hé stole, a very impolite guy*'

In this paper I shall argue that my data lend support for the pragmatic model in that it provides a fuller explanation for the use of overt versus null pronouns in Spanish. The fifteen spoken Spanish narratives analyzed in this study were elicited under the same conditions after viewing "The Pear Film" (Chafe 1980). The objectives of my analysis are the following: (1) to identify the reasons for the use of null versus overt pronouns in the narratives; (2) to determine the extent to which the cognitive and the pragmatic theoretical frameworks are able to provide an explanatory account for the use of overt/null subjects in the Spanish narratives; and (3) to draw some analogies between the use of Spanish overt/null pronoun use and corresponding rules of English pronoun use. Finally, because the correct use of expressed subject pronouns versus null subjects is a problem for English speaking learners of Spanish, this study may also have important implications for a pedagogical grammar for teaching Spanish to native speakers of English.

The role of pedagogical grammar has always been a controversial issue in second language acquisition.  According to Terrell's Natural Approach (1977, 1982), grammatical explanations and exercises are to be written clearly and concisely. Likewise, in a reductive grammar approach, as I propose, focus is on a single distinguishing characteristic in usage of a particular grammar point.  The distinguishing characteristic is linked to a conceptualization, which aids in a clear presentation of the grammar point and facilitates the acquisition process.  This paper proposes that by using a reductive grammar approach the learners' oral and written performance is positively affected for some linguistic topics in Spanish.

A number of investigations into what nonlinguists believe about language structure and use have suggested that they base a great deal of what they know on the underlying concept of correctness. What is most interesting for linguists about this folk notion, however, appears to be its positioning outside the social facts of language. Apparently nonlinguists are unashamedly Platonist in their belief about a non-socially constructed but cognitively external existence for language, an "ideal form" which enfranchises all other language use but is, of course, more sinned against than utilized. The repercussions of this Platonist belief about language in language teaching and learning and policy and planning are also evident, although my major data will be taken from quantitative questionnaire and qualitative discoursal data collected from nonlinguist respondents over the last twenty years.
 
 

Dr. Jackson will be discussing the implications that recent SLI theories may have for African American English (AAE) speaking children.  Dr. Jackson will present relevant data on the tense and aspect system of AAE and how its manifestations might be misinterpreted as language disorder in light of recent research that identifies tense as a probable location for linguistic difficulty.  Dr. Jackson will also discuss the subtleties of the tense and aspect system of AAE that AAE speaking children must handle in order to operate the system adequately.  Finally, Dr. Jackson will discuss how recent research in AAE might shed light on issues of impairment for all language-impaired speakers of English.
 
 

It is clear that the Hittite Empire was multiethnic and multilingual.  However, the spheres of use of the two major languages, Hittite and Luvian, and their relationship to each other remain controversial.  I will present evidence that Hittite was a spoken language for at least some segment of the society at all times, contrary to some claims.  I will also reexamine the effects of Luvian on Hittite, both lexical and grammatical, with more careful attention to the relative chronology than found in previous scholarship.
 
 

This presentation reports on a study that investigated the effect of two conditions (attention to form and meaning vs. attention to meaning) on how learners process object pronouns in Spanish in extended discourse.  This was analyzed taking into consideration the distance between a pronoun and its antecedent.  In order to better understand how learners process input, think-aloud protocols produced by the participants as they performed the task were analyzed. The 155 participants performed a multiple-choice task consisting of a text that contained examples of direct object pronouns and 16 blanks where the antecedents of those pronouns had to be placed.  Results suggest that (a) a task designed to direct attention to form and meaning is more difficult than one that only requires attention to meaning, (b) this difference is greater in the cases of long distance reference, and (c) mention of the targeted structure in the think-aloud protocols may be directly related to better performance on the task.
 
 

Our keynote presentation explores the link between Africans in Sierra Leone with the Gullah people of coastal South Carolina and Georgia. Over the years Mr. Opala's scholarly research has ranged from an archaeological study of an 18th century British slave castle in Sierra Leone, to historical research on the slave trade in colonial Charleston and Savannah, to oral history research with the Gullahs in South Carolina/Georgia and the Black Seminoles in Oklahoma.  Featured on PBS in 1998, the companion documentary film “The Language You Cry In” based on the reunion of a Georgia family with their African relatives will also be presented.  Mr. Opala is a part-time instructor of History at James Madison University.  Sponsored by Organization of Africans at USC, Dept. of Anthropology, Linguistics Program, and International Programs for Students.
 
 

The fundamental division in the field of linguistics today is between 'formalists' and 'functionalists'. Formalists construct models of grammar in which the syntactic structures of a language are characterized independently of their meanings and uses. Functionalists reject autonomous syntax on the basis of the conviction that grammatical form has been too profoundly shaped by the (general) nature of cognition and communication to allow for the  compartmentalization of form, meaning, and use.

This paper will argue that autonomous syntax and functional explanation are fully compatible. First, it will present evidence for the autonomy of linguistic form. Second, it will show that aspects of formal grammatical systems are functionally motivated. Third, it will discuss mechanisms by which such systems might be shaped externally. The paper will conclude with some speculative remarks on why a formal autonomous system at the heart of language might have arisen in the course of human evolution.
 

Even though the Tomatis Method of sound stimulation training has been used in Europe for many years, until recently little research has been available to support the claim that sound stimulation training improves foreign language listening comprehension and pronunciation. This colloquium presentation will begin with a report on a recent large-scale European study of sound stimulation training that compared three treatment groups and found that the Tomatis group achieved the desired goals in listening comprehension and pronunciation 50% faster than the two comparison groups. This report will be followed by an update on smaller scale research conducted with students enrolled in beginning and intermediate French classes at this institution.

This study proposes and explores a model of social variables in second language (L2) acquisition.  In terms of theory, it weds three strains of sociolinguistic research: early social models of L2 acquisition (Gardner 1985, Giles and Byrne 1982), research on the ideology of language (Gal and Woolard 1995, Bakhtin 1981), and investigations of the role of language in ethno-national identity (Fishman 1999, Laitin 1998).  Although theories of L2 learning normally assume an indirect influence from learner-external factors such as group ideology and ethno-national identity (Ellis 1992:197), social models of L2 acquisition have been unable to capture the complexity of these influences and quantify their relative weight. In this study, I propose a model linked by two predictions: (1) that an individual's first language will constrain the set of accessible ethno-national identities and (2) that the chosen identity will assign values to any possible L2, which will, in turn, affect the individual's access, motivation, attitude, and proficiency with respect to that language. Together these predictions comprise a model of social transfer from the L2 learner's first language via ethno-national identity.  Data from a survey of 65 multilingual subjects in the emergent national context of Moldova demonstrate the mechanisms of the social transfer and identity model.  The data include indices of the subjects' L2 access, motivation, attitude, and proficiency.  Specifically, the contrasts in survey responses among those reporting Moldovan, Romanian, and Russian identities support the basic model and provide directions for further research to elaborate and test it.