Acquisition of Rule-Governed vs. Lexical Stress (NIH Grant R03 DC05132)
Susan Guion (PI)
Tetsuo Harada, East Asian Languages and Literatures, University of Oregon (Co-PI)
Ratree Wayland, Program in Linguistics, University of Florida (Co-PI)
J.J. Clark, Department of Linguistics, University of Oregon (Graduate Assistant)

The goal of the funded research is to gain further insight into processes involved in second language acquisition. The first aim is to investigate the effects of type of learning (i.e., that which has traditionally been called rule-based vs. that which has traditionally been called irregular or associative), age of learning, and their interaction. Our second aim is to investigate the role of first language transfer in the formation of rules in a second language. Our third aim is to investigate the ability to acquire production proficiency of stress accent by learners whose native language exhibits non-stress accent or tone. The results and interpretation of the behavioral studies will provide a theoretical and empirical background for further investigation with neuro-physiological measures. Insight into the mechanisms of language acquisition and its interaction with age of exposure can contribute to advancements in linguistic rehabilitation after strokes and development of second language pedagogy.

The acquisition of English stress, both regular and irregular, is being investigated. Speakers of a variety of first languages are included in the research. Native speakers of Japanese (a language that specifies accent lexically), French (a language that has regular phrase-level accent), Spanish (a language that has both regular and irregular accent), and Thai (a language that has lexically specified tone) will participate. Native English controls are also tested. Both early and late bilinguals are included in order to investigate the proposal that, in the phonological domain, some types of learning (abstraction or generalization) are subject to a critical period whereas other types of learning (associative or memorized) are possible throughout the life span. In the experiments currently being conducted, the effect of age of learning on acquisition of regular and irregular English stress placement is investigated. In production and perception experiments, early and late bilingual groups are asked to syllabify nonsense words in noun, verb sentence frames. Nearest lexical neighbors for the nonsense words are being collected as well. The acquisition of stress accent production characteristics are also being investigated. Measures of duration, fundamental frequency, amplitude and airflow are compared in stressed vs. non-stress syllables in minimal pairs such as pÈrmit/permÌt. Effects of native language phonology are investigated in cross-linguistic comparisons.

Nilotic Languages Network

Nilotic languages are spoken in Tanzania, Kenya, Uganda, Sudan, Ethiopia and Congo. The Nilotic family is said to be one of the two most-complex sub-families within the larger Nilo-Saharan phylum. It contains some 29-to-53 languages, depending on complex issues concerned with degrees of mutual intelligibility and ethnic self-identity. Numbers of speakers per language range from about 3.5 million (Dholuo) to 50 or fewer (e.g., Omotic). A number of Nilotic languages and cultures have, for decades, experienced significant disruption due to war, famine, and movements to refugee camps in search of food, medical attention, and safety. Over the past century other groups have experienced significant and sustained contact with Western influences via colonialism, mission activities, western-oriented educational systems, establishment of international cities in their tradiational areas, tourists, government land policies, etc. Other small groups have been, and are being, absorbed into larger language groups. While some languages of the family have been reasonably well documented, others have received essentially no documentation and attempts to further understand the family are hampered by lack of adequate data.

The Nilotic Languages Network supplement grant is supporting three to four workshops involving currently-active Nilotic linguists. The first workshop was held in Kisumu, Kenya (March 2002), and the second in Leipzig, Germany (July 2002). These workshops are enabling Nilotic scholars to learn of each other's work and in so doing aim to increase the quality and quantity of research results beyond what individuals working in isolation have achieved. Together, the workshops are addressing standardization issues in the creation and further development of lexicographic and text data bases for Nilotic languages; exploring possible coordination between data bases; and encouraging cross-family typological study of linguistic phenomena salient in the family (e.g., lexical semantic features, phonological features, morphological features).

A Nilotic web site has been established at: http://darkwing.uoregon.edu/~dlpayne/Nilotic/NiloticNetwork.htm

Maa (Maasai) Lexicography and Text Databases

The Maa (Maasai) language is currently spoken by some 800,000 Maasai, Samburu, Camus and Okiek peoples in Kenya and Tanzania. In all cultures there is both basic and specialized vocabulary which describes activities, traditions, cosmology, religion, and the myriad ways of life specific to that culture. These features cannot be thoroughly understood without an understanding of the vocabulary which expresses the concepts comprising the traditions of the culture. The extensive lexicographic and text database will be valuable documents for preserving and transmitting the cultural knowledge of the Maa people, who are being impacted by rapid cultural change at the turn of the 21st century. Previously, however, no linguistically accurate set of texts or dictionary of the language has existed.

The current project is producing computerized text and lexicography databases, with a comprehensive set of fields, such that a variety of dictionaries and text materials can eventually be published depending on the needs of various audiences, including linguists, anthropologists, historians, bilingual school teachers, non-governmental organizations, and Maa speakers themselves. The project is also studying Maa tone, vowel harmony, semantic and morphosyntactic properties of verbs, and syntactic constructions.

The first stage of the project focused on the variety of Maa spoken in central to southern Kenya by Il-Keekonyokie and Il-Purko sections in the Narok and Kajiado regions of Kenya; the second stage extended to north Maa as spoken by the Samburu and to southern Maa as spoken by the Kisongo of Tanzania. Current work is extending the research to western Maa as spoken by the IlWuasinkishu section.

URL to NSF award and original abstract: https://www.fastlane.nsf.gov/servlet/showaward?award=9809387

URL to my Maasai pages: http://darkwing.uoregon.edu/~dlpayne/maasai/madict.htm

Akawaio Grammar (NSF Grant No. BCS-0117619)
Spike Gildea, University of Oregon and Rice University, and Desrey Fox, Rice University and University of Oregon

Project Summary

Lowland South America is one of the least studied linguistic areas of the world, and in South America, the indigenous languages of Guyana are among the least known. We propose to create the first comprehensive reference grammar of an indigenous language of Guyana, Akawaio (Cariban). The PI has conducted field work on 11 Cariban languages in Venezuela and Brazil, has written a book on comparative Cariban grammar, and has directed/is directing students in writing descriptive grammars of South American languages. The co-PI, Desrey Fox, is a native speaker of Akawaio who is in the early stages of writing her Ph.D. dissertation in linguistics/anthropology at Rice University. Fox's dissertation will include a large component of transcribed and glossed texts of various genres. This three-year project will make it possible to: (1) create - and publish on the web - an annotated, computerized database of Fox's transcribed and glossed texts; (2) begin co-writing a grammar of Akawaio (partly on the basis of this database) before Fox leaves the U.S. to return to her academic position in Guyana; (3) conduct joint field research in Guyana, checking Fox's linguistic intuitions against those of a range of Akawaio speakers (especially as regards sociolinguistic variation); and (4) work jointly in the U.S. during later stages of the writing, combining the benefits of an academic setting (library, computer support) and native speaker intuition at crucial stages of the work. The grammar of Akawaio offers typologists and theoreticians a previously unattested type of split ergativity, a case of reflexive morphology evolving into a middle voice and then apparently lexicalizing into the majority of intransitive verbs, and interesting morphophonological phenomena at the boundaries between verbs and person-marking morphology. Also, due to the participation of a native speaker, this project will be able to go farther than most in documenting the many varieties of Akawaio speech, from the tremendous sociolinguistic variation associated with the many communities of Akawaio speakers to the more archaic variations seen in ritual speech.