This dissertation deals with language documentation from a theoretical and practical perspective. Alongside a theoretical discussion of language documentation, I present my own language documentation project carried out among the Kurripako of Venezuela. What is language documentation? How should it be carried out? And what does it include? These are some of the questions that I address in this dissertation through the analysis of my own project.I argue that language documentation is a multipurpose, data-driven gathering of records of a language that should take into account the needs of various interested parties such as researchers and community members. In this way, the needs and desires of many should be taken into account in the planning and execution of such a project.I provide examples of those different parts by looking at my project among the Kurripako. In this sense, each chapter addresses a different part of a project. Chapter 2 addresses theoretical and pragmatic issues. Chapter 3 addresses information about the language, its speakers, dialect variation and a grammatical sketch. Chapter 4 examines argument marking structure in depth and concludes that Kurripako is an active-stative language. Chapter 5 examines the active-stative split in texts and argues that text collection needs to be complemented by elicitation in order to be able to fully address grammatical aspects. Chapter 6 provides information on the context of the project and explains the interests that speakers had in collecting particular types of texts. I finalize by arguing that there needs to be much more work on the theory of language documentation as well as evaluations of these theoretical proposals in order to serve all needs that language documentation wishes to.I do not intend to present my language documentation project as an exemplar of how language documentation should be done, but merely use it to reflect upon issues that such a project raises. There more questions that came to light and that remain to be answered than issues that have been provided with an answer. It is my hope that this will also encourage others to tackle some of these questions.

archives@tulane.edu / The linguistic contexts of the African continent are undoubtedly complex and quite frequently polemic. In addition to a history of European colonialism, these are further complicated by matters of ethnic, political, and religious identity, oftentimes conflating some linguistic distinctions and establishing others without recourse to the data. A country like Zambia, on the other hand, which has nationally embraced and promoted ethnolinguistic differences—though arguably not at the expense of others—realizes extensive language intermixing that contributes to this complexity. This dissertation attempts to fulfill the request of the Aushi community of the Lwapula Province, Zambia, to provide an initial linguistic account of Ikyaushi. Following in the tradition of the “Boasian Trinity,” this is reached through three separate, though interrelated tasks, viz. the composition of a descriptive grammar, the development of a preliminary dictionary, and the compilation of culturally relevant texts. The descriptive grammar provides an introductory account of the phonetics and phonology, morphology, and syntax of Ikyaushi, and this account is based primarily on the analysis of fourteen collected narratives and secondarily upon naturally observed and elicited data. The narratives were recorded, transcribed, translated, and analyzed among male and female speakers from or in Matanda, Mansa, Kabunda, and Mabumba, and these narratives arrive in the form of fictional stories, trickster tales, a short history of the people, and descriptions of female initiation, brideprice, engagement procedures, and traditional education. As such, this dissertation contributes more broadly to the field of ethnographically- informed Bantu linguistics and provides more generally an avenue for speakers of Ikyaushi to be empowered, knowing that their linguistic differences are recorded and available for consideration, however small or large. Additionally, it is believed that these newly developed resources will contribute to future pedagogical materials to assist in literacy efforts among speakers and volunteers in the Lwapula Province. / 1 / Troy E. Spier

This thesis presents topics in the grammar of South Efate, an Oceanic language of Central Vanuatu as spoken in Erakor village on the outskirts of Port Vila. There has been no previous grammatical description of the language, which has been classified as the southernmost member of the North- Central Vanuatu subgroup of languages. In this description I show that South Efate shares features with southern Vanuatu languages, including a lack of serial verb constructions of the kind known for its northern neighbours and the use of an echo-subject marker. The phonology of South Efate reflects an ongoing change in progress, with productive medial vowel deletion and consequent complex heterorganic consonant clusters. (For complete abstract open document)

This study constitutes the first in-depth description and analysis of Texas Alsatian as spoken in Medina County, Texas, in the twenty-first century. The Alsatian dialect was transported to Texas in 1842, when the entrepreneur Henri Castro recruited colonists from the Alsace to fulfill the Texas Republic’s stipulations for populating his land grant located to the west of San Antonio. Texas Alsatian (TxAls)is a dialect distinct from other varieties of Texas German (Gilbert 1972: 1, Salmons 1983: 191) and is mainly spoken in Eastern Medina County in and around the city of Castroville. With a small and aging speaker population, it has not been transmitted to the next generation and will likely survive for only another two to three decades. Despite this endangered status, TxAls is a language undergoing death with minimal change. This study provides both a descriptive account of TxAls and discussions on extralinguistic factors linked to ethnic identity and language loyalty, which have enabled the maintenance of this distinctive Texas German dialect for 150 years. To investigate the extent of the maintenance of lexical, phonological, and morphological features, this study identifies the main donor dialect(s), Upper Rhine Alsatian, and compares its linguistic features to those presently maintained in the community, based on current data collected between 2007 and 2009 and Gilbert’s (1972) data collected in the 1960s. This discussion of TxAls is three-fold: (1) an analysis of social, historical, political, and economic factors affecting the maintenance and decline of TxAls, (2) a detailed structural analysis of the grammatical features of TxAls, supported by a description of its European donor dialect and substantiated by Gilbert’s (1972) data, and (3) a discussion of the participants’ attitudes toward their ancestral language, which have either contributed to the maintenance of TxAls, or are now accelerating its decline, based on responses to a survey developed for the TxAls community, the Alsatian Questionnaire. / text

This dissertation explores the nature, purpose, function and role of language documentation in order to further our understanding of mechanisms of language transmission and maintenance in the face of language endangerment and the repression of indigenous identity. Beyond its traditional use for generating linguistic data, I argue that the act and the process of language documentation can be understood as a comprehensive means to evaluate the interactions between speakers and researchers and as the stage where various beliefs and emotions are displayed. Extending the notion of “sites” developed by Silverstein (1998) and Kroskrity (2009), I argue that the act of language documentation can create “sites” of linguistic transaction, of self recognition, and of ideological and emotional stance shift. To attain this goal, this project linguistically and ethnographically documents and describes Belizean Mopan, an endangered Mayan language spoken in the southern Petén region of Guatemala and in the Maya Mountain region (Toledo District) of Southern Belize as a case study. Ethnographic and linguistic observation suggest that characteristics of Belizean Mopan do not simply stem from its linguistic features but rather are derived from ethnic complexity, language ideologies, identity politics, the history of Belize and speakers’ awareness of the self. Linguistic biographies, interviews, participant observation, and ethnographic accounts indicate that the individual’s emotional attachments to the language and the sense of belonging to one’s linguistic community are crucial keys for effective language documentation and revitalization. Discourse and grammatical analysis of sound symbolic words in narratives suggest that speakers’ linguistic affects can be evoked through sound itself. The devices used during language documentation, such as voice and video recorders can be understood as “signifying instruments” (J. D. Hill 2014), which amplify or evoke speakers’ and researchers’ linguistic ideologies and/or affects. Tzik ‘respect’ plays a pivotal role in distinguishing Mopans from other Maya groups and many stories and personal narratives either explicitly or subtly demonstrate the concept and importance of tzik for regulating and maintaining the traditional community and for having a successful life, which resembles the secretos ‘secrets’ described in Hofling’s (1996: 109) account of Itzaj Maya lives. Focusing on tzik gained through being a ch’ija’an kristiyanojo ‘the grown-up people’, I argue that storytelling is a primary device to transmit and circulate traditional knowledge, worldview, ideologies and memories of Maya people from the present, the immediate past, and the mythological past and that in a sense, the role and meaning of dream divination and my language consultant, Orlando Sho’s musical performances can be equated with the practice of storytelling. The act of language documentation is a portal to the site of linguistic and cultural transaction and of world learning, in which I see a key to successful language renewal and revitalization.

xxxix, 729 p. : ill. (some col.) / Kurtop is a Tibeto-Burman language spoken by approximately 15,000 people in Northeastern Bhutan. This dissertation is the first descriptive grammar of the language, based on extensive fieldwork and community-driven language documentation in Bhutan. When possible, analyses are presented in typological and historical/comparative perspectives and illustrated with ample data, drawn mainly from texts but also elicitation as need be. Within Tibeto-Burman, Kurtop has been placed within the East Bodish sub-branch. Data presented in this study support this placement and confirm previous observations that the East Bodish languages are close relatives, but not direct descendants of Classical Tibetan. The link between the current East Bodish languages and Bhutanese prehistory remains unclear but the Kurtop grammar is a first step at understanding the historical relations. The most remarkable aspect of Kurtop phonology is the tonal system, which is contrastive following the sonorants, but incipient following the obstruents, except the palatal fricative, for which tone has completely replaced a previous contrast in voicing. Tone is present only on the first syllable of stems, where vowels are also slightly longer. Kurtop is agglutinating and polysynthetic. Words generally consist of two or three syllables, but may be as long as five or six, depending mainly on suffixing morphology. Like most languages of South Asia, Kurtop exhibits verb-final syntax and the typological correlations that follow, including postposition (or relator noun constructions), auxiliaries after the verb, and sentence-final particles. The case marking system is 'pragmatic' ergative, where an ergative marker is required in some transitive contexts, but not in others. In other contexts, including for some intransitive verbs, the ergative signals a variety of pragmatic or semantic factors. This ergative system, though typologically unusual, is characteristic of many Tibeto-Burman languages, including neighboring Dzongkha and Tshangla. Nominalization and clause-chaining are two essential components of Kurtop syntax, constituting a majority of clauses and a diachronic source for much of the main clause grammar. The evidential/mirative system in Kurtop is also of typological interest, encoding a wide range of values pertaining to speaker expectation as well as mirativity and source of knowledge. / Committee in charge: Scott DeLancey, Chairperson and Advisor; Spike Gildea, Member; Doris Payne, Member; Gyoung-Ah Lee, Member; William Ayres, Outside Member

This dissertation explores the phenomenon of ways of speaking in the Nanti speech community of Montetoni, in southeastern Peruvian Amazonia, between 1999 and 2009. In the context of this study, a 'way of speaking' is a socially meaningful, conventionalized sound pattern, manifest at the level of the utterance, that expresses the speaker's orientation toward some aspect of the interaction. This study closely examines both the sound patterns and patterns of use of three Nanti ways of speaking — matter-of-fact talk, scolding talk, and hunting talk — and describes each one in relation to a broader set of linguistic, social, and cultural practices characteristic of the speech community at the time. The data for this study is naturally-occurring discourse recorded during multi-party, face-to-face interactions in Montetoni. Bringing together methods developed by linguists, linguistic anthropologists, conversation analysts, and interactional sociologists, this study explores the communicative relations among participants, interactions, situations of interaction, and the utterances that link them all, attending to both the individual-level cognitive (subjective) facets of interpersonal communication and the necessarily intersubjective environment in which communication takes place. In order to disaggregate the multiple levels of signification evidenced in specific utterances, tokens are examined at four levels of organization: the sound form, the sentence, the turn, and the move. The data are presented via audio files; acoustic analyses; sequentially-organized and temporally-anchored interlinearized transcripts; and composite visual representations, all of which are framed by detailed ethnographic description. Nantis' ways of speaking are shown to consistently and systematically convey social aspects of 'meaning' that are crucial to utterance interpretation and, therefore, to successful interpersonal communication. Based on the robust correspondences between sound form and communicative function identified in the Nanti communicative system, this study proposes that ways of speaking are a cross-linguistically viable level of organization in language use that awaits discovery and description in other speech communities. The research project itself is framed in terms of the practical issues that emerged through the author's own experiences in learning to communicate appropriately in monolingual Nanti society, and the ethical issues that motivate community-oriented documentation of endangered language practices. / text

By the end of this century, half of the approximately 6000 extant languages will cease to be transmitted from one generation to the next. The field of language documentation seeks to make a record of endangered languages before they reach the point of extinction, while they are still in use. The work of documenting and describing a language is difficult and extremely time-consuming, and resources are extremely limited. Developing efficient methods for making lasting records of languages may increase the amount of documentation achieved within budget restrictions. This thesis approaches the problem from the perspective of computational linguistics, asking whether and how automated language processing can reduce human annotation effort when very little labeled data is available for model training. The task addressed is morpheme labeling for the Mayan language Uspanteko, and we test the effectiveness of two complementary types of machine support: (a) learner-guided selection of examples for annotation (active learning); and (b) annotator access to the predictions of the learned model (semi-automated annotation). Active learning (AL) has been shown to increase efficacy of annotation effort for many different tasks. Most of the reported results, however, are from studies which simulate annotation, often assuming a single, infallible oracle. In our studies, crucially, annotation is not simulated but rather performed by human annotators. We measure and record the time spent on each annotation, which in turn allows us to evaluate the effectiveness of machine support in terms of actual annotation effort. We report three main findings with respect to active learning. First, in order for efficiency gains reported from active learning to be meaningful for realistic annotation scenarios, the type of cost measurement used to gauge those gains must faithfully reflect the actual annotation cost. Second, the relative effectiveness of different selection strategies in AL seems to depend in part on the characteristics of the annotator, so it is important to model the individual oracle or annotator when choosing a selection strategy. And third, the cost of labeling a given instance from a sample is not a static value but rather depends on the context in which it is labeled. We report two main findings with respect to semi-automated annotation. First, machine label suggestions have the potential to increase annotator efficacy, but the degree of their impact varies by annotator, with annotator expertise a likely contributing factor. At the same time, we find that implementation and interface must be handled very carefully if we are to accurately measure gains from semi-automated annotation. Together these findings suggest that simulated annotation studies fail to model crucial human factors inherent to applying machine learning strategies in real annotation settings. / text

xxii, 995 p. A print copy of this thesis (with two accompanying DVD videos) is available through the UO Libraries. Search the library catalog for the location and call number. / Since 2005, I have been working with members of the Kari'nja community of Konomerume, Suriname to document, describe, preserve, and revitalize their heritage language, the Aretyry dialect of Kari'nja (Cariban family). Simultaneously, I have worked to develop, pilot, and articulate a model of field research that depends on participation from speech community members. This dissertation combines exposition of this model of field research with presentation of a large body of the results from the application of that new model. Ethnically Kari'nja, Konomerume community members have witnessed a decline in language use in recent generations. Although I work primarily with members of the Konomerume community, a village on the banks of the Wajambo River in Suriname, in recent years, I have expanded my work to include two other communities in the region, Corneliskondre and Kalebas Kreek. My work with Kari'nja community members concerns four broad, interrelated areas of endangered languages research, each described in a section of the dissertation. Following Chapter 1, which provides an orientation to the dissertation as a whole, Chapter 2 reviews strengths and problems with prior models of fieldwork, then proposes a new model of fieldwork with members of Indigenous communities. Chapter 3 demonstrates some of the linguistic results of our work together in Konomerume, offering a more pedagogical overview of some aspects of Kari'nja grammar followed by more academic descriptions of nonverbal predication and an innovative main clause progressive construction. Chapter 4 addresses how documentation can be combined with applied linguistics to support revitalization through formal language teaching. Finally, Chapter 5 describes the documentary corpus that is found in the appendices, explaining procedural steps used in creating the corpus and outlining the actual documentary products that we have produced. The Appendices are the concrete representation of the body of collaborative work that the Kari'nja community and I have done together. They include DVD videos, a substantial collection of transcribed, translated, and grammatically annotated texts in multiple genres, a dictionary, a pedagogical grammar sketch, and a curriculum guide for formal teaching of introductory Kari'nja. / Committee in charge: Spike Gildea, Chairperson, Linguistics; Eric Pederson, Member, Linguistics; Susan Guion, Member, Linguistics; Janne Underriner, Member, Linguistics; Brian Klopotek, Outside Member, Anthropology

In the present study, I describe the linguistic expression of space in Caac, an Oceanic language spoken in New Caledonia, from both a descriptive and theoretical perspective. Caac is a minority language whose transmission process is not ensured anymore; it is also an under-documented language. Part I provides a concise description of Caac grammar, presenting thereby a first formal portrait of this language to the reader. Part II describes the formal and semantic features of the linguistic resources available in Caac to encode spatial relationships. Part III presents the theoretical framework based on and exploring further the vector analysis developed by Bohnemeyer (2012) and Bohnemeyer & O’Meara (2012). In particular, I propose an additional sub-category of vectors (Head-unspecified Vectors) which account for the uses of centrifugal forms in Caac. The resulting theoretical framework enables me to provide a systematic account of expressions of orientation as well as location and motion, and to combine the Frames of Reference typology (Pederson et al. 1998; Levinson, 1996, 2003; Bohnemeyer & Levinson, not dated) with an analysis of deictic expressions within a single framework. It also allows us to give a detailed analysis of the uses and combinations of Caac absolute and deictic directionals, which are spatial terms of primary importance for spatial reference in Caac. Special attention, moreover, is given to the use of directionals in spatial constructions involving Fictive Motion. The analysis of Caac data leads us to introduce an additional category of Fictive Motion beyond those previously recognised in the literature, labelled here ‘Anticipated Paths’. In the conclusion, I propose a functional and cultural-specific explanation for the emergence of this construction. Anticipated Path expressions in turn shed new light on the nature of vectors and the relationship between location, motion and orientation.