published_or_final_version / Language Studies / Master / Master of Arts

This thesis investigates the phonetic and phonological properties of laryngeal distinctions in the consonant inventory of Tahltan, a northern Athapaskan language. This thesis does not examine the phonetic properties of all Tahltan consonants. Instead, this thesis focuses on the phonetic acoustic properties of plain stop consonants, which have been described inconsistently in Tahltan, to determine their laryngeal specification. This thesis also examines the observed patterns of behavior governing syllable structure to help determine the laryngeal specification of consonants in Tahltan. In addition, several morphophonemic processes are examined to determine the phonological laryngeal specification of consonants in Tahltan. Based on the phonetic findings, and observed patterns of behavior governing syllable structure, I will argue that stop consonants in Tahltan exhibit four laryngeal articulations: voiced, voiceless unaspirated voiceless aspirated and glottalized. Based on the morphophonemic evidence, I will argue that fricative consonants exhibit two laryngeal articulations: voiced and voiceless. Furthermore, I will argue that glottal stop is specified for the laryngeal specification [constricted glottis] (henceforth [CG]) and that [h] is specified for the laryngeal specification [spread glottis] (henceforth [SG]).

In this thesis, we investigate several processes affecting coronals and velars in Tamil and Malayalam, two Dravidian languages spoken in southern India. We begin by discussing two assimilation processes which apply adjacent to front vowels, Palatalization, where anterior coronals become palatoalveolar, and Coronalization, where velars are fronted to palatoalveolar. We compare and contrast the feature geometries proposed by Sagey (1986) and Hume (1992) in their ability to adequately express these processes. In Sagey's model, front vowels are argued to be Dorsal. It is thus impossible to express either Palatalization or Coronalization as spreading. In Hume's model, where front vowels are Coronal, both processes involve spreading. However, the model does not formally distinguish between these two processes across languages; thus, it fails to capture the fact that Palatalization is widely attested but Coronalization seems to be restricted to diachronic alternations. In order to express this asymmetry, we adopt the model advanced by Goad & Narasimhan (1994), a revision of Goad (1993), where Palatalization involves spreading but Coronalization is a two-step process, spreading followed by reanalysis. In this model, a single feature (front), defined as "front of articulator", is doubly dependent on both Dorsal and Coronal nodes. Its interpretation is thus partly determined by the node to which it links; it marks apicality in coronals and front of tongue body in dorsals. In Chapter 3, we demonstrate how this model allows us to capture the fact that in Malayalam, only a subset of the anterior coronal consonants, the apicals, form a natural class with front vowels. In Chapter 4, we provide support for the model from languages other than Tamil and Malayalam, both Dravidian and non-Dravidian.

Thesis (Ph. D.)--State University of New York at Binghamton, Department of Psychology, 2009. / Includes bibliographical references.

The research is a study of male working class pronunciation in the Rhondda, part of the 'Valleys' area of South East Wales. It encompasses both segmental and suprasegmental (prosodic) phonology. The segmental analysis is primarily auditory although it has some supporting acoustic detail. It examines the consonant and vowel systems of Rhondda Valleys English (RVE), with phonetic realizations and lexical incidence. Comparisons with British R.P. are made and similarities with neighbouring varieties of English (e.g. the West Country) and the Welsh Language are observed. The suprasegmental (prosodic) analysis is of spontaneous conversational data, and is auditory and instrumental. The phonology of RVE intonation is described mainly via a system of intonation phrases (IPs), accents, and terminal tones. IP tunes (overall contours) are observed to contain accent profiles whose pitch obtrusions to the stressed syllable are, in the majority of cases, downwards and whose initial pitch movement from the stressed syllable is rising in over 80% of final accents and final accents. A large majority of IP terminal tones in the data are ultimately rising. Aspects of length and rhythm are examined. Evidence is found of rhythmic organization, e.g. of alternation between strong and weak beats. Strongly accented syllables can be accompanied either by lengthening of the vowel, or by shortening of the vowel with lengthening of the succeeding consonant. Which of these two strategies is adopted by the speaker depends partly on the vowel and partly on how the speaker syllabifies the word. The final 'weak' syllable of an IP may be phonetically stronger (with greater duration, envelope amplitude and pitch prominence) than the accented penult. Several of the prosodic features of RVE are found to bear strong influence from the Welsh Language.

Neste trabalho, investiga-se a distribuição dos sufixos -inho e -zinho nos dialetos de Porto Alegre e Curitiba. A partir de estudos que contemplaram tais afixos, olhamos para fatores de natureza lingüística, como classe gramatical (nomes, não-nomes), tonicidade (oxítona, paroxítona e proparoxítona), segmento final da forma primitiva (vogal baixa, vogal média-baixa frontal e posterior, -i, - u primitivos, -e, - o primitivos, -s / -z, -r / -m, -l), onset da sílaba final (nasal dorsal, labiais, coronais, dorsais, onset vazio), e para fatores de natureza extralingüística, como escolaridade (primário, ginásio, secundário), sexo (masculino e feminino), faixa etária (menos de 50 anos e mais de 50 anos), localidade (Porto Alegre e Curitiba). Esta pesquisa faz uso de dados de fala extraídos de 24 entrevistas do Projeto VARSUL e de um teste de produtividade, que se utiliza de pseudopalavras, aplicado a 20 informantes. Para a análise quantitativa dos resultados, fazemos um uso adaptado dos programas que compõem o pacote VARBRUL, lançando mão tão-somente de freqüências, já que não se trata de um fenômeno tipicamente variável. A análise dos resultados mostrou que -inho é o sufixo mais usado, ainda que -zinho seja o preferido no contexto de pseudopalavras. Além disso, observouse um padrão predominantemente alternante, conforme já previu a literatura, definido pela tonicidade, ainda que algum lugar para usos variáveis esteja reservado. / In this paper, the distribution of the suffixes -inho and -zinho is investigated in the dialects of Porto Alegre and Curitiba. Starting from studies that cover such affix, we looked at factors of linguistic nature, like grammatical class (names, nonames), tonicity (stressed on the last, second last, or third last syllable), final segment in the primitive form (low vowel, back and front low-mid vowels, -i, primitive -u, -e, primitive -o, -s / -z, -r / -m, -l), final syllable onset (dorsal nasal, labial, coronal, dorsal, empty onset), and extra linguistic nature factors, such as educational background (primary, elementary, high school), sex (male and female), age group (less than 50 and more than 50 years old), place (Porto Alegre and Curitiba). This research makes use of extracted speech data of 24 interviews from Projeto VARSUL and of a productivity test, which uses pseudo words, applied to 20 applicants. For the quantitative analysis of the results, we make an adapted use of the programs that compose the package VARBRUL, not taking into account only of frequencies numbers, since it is not a typically variable phenomenon. The analysis of the results showed that -inho is the most used suffix, although -zinho is the favorite in the context of pseudo words. Besides, a predominantly alternating pattern was observed, as already expected in literature, defined by the tonicity, although some place for variable uses may be reserved.

This research work has several aims, demanded by a number of external and internal factors. First of all, it aims to identify Lala: whether it is simply a dialect of Tekela- Nguni, or perhaps the substratum Bantu language spoken in the eastern parts of Southern Africa since the fourth century of the first millennium A.D. on which every new migratory wave imposed its language (chapters one and two). The lexical-statistical method has been attempted (chapter three), but with mixed results. In fact, relexification is widely used in southern African societies, and this renders the identification of the core vocabulary a very difficult task. The result is that all that can be said is that Lala indeed shows the characteristics of a Tekela-Nguni language, but it is impossible to determine at this stage whether Lala is the mother or the daughter of Tekela. Another important aim, dictated by an earlier research by Wilkes (1981 ), was to compare the two main branches of Lala, called North and South Lala, to prove whether the differences between the two dialects are so wide as to justify a division into two separate entities or not. Here the verdict is emphatically negative. In fact the various stages of the research abundantly re-affirm the view that Lala is one language, with only marginal differences caused by the different linguistic environments: Zulu in the north and IsiZansi in the south. African societies in this part of the world have affirmed themselves in the realm of the written word only recently, i.e. during the last 150 years or so. The background culture of these societies is still vigorously oral, and this means that the spoken word is used for much more than simple communication of thought and feeling. Language is felt as the major binding element in a society that has seen constant political break-ups, upheavals, migrations, wars, attempted exterminations. This means that Lala, as a language or the dominant dialect of a group, is banded about as a cultural-historical flag around which people are proud to gather. This sense of unity is an emotion, a feeling, rather than a deep reality. The colours of the flag, or the distinguishing elements of the language, may be fading away, and only a core might be preserved. But this is quite enough to kindle emotions and to rally people. At least in areas where people are still proud of their cultural heritage. hnicity and language should not be the same, but most of the people interviewed (cf chapter two) felt very strongly that they were Lalas because they spoke Lala; and that they spoke Lala because they are Lalas. Their either glorious or sad histories are recorded in chapter two to demonstrate how Shaka's wars, and the Mfecane, forced them to leave their homeland and to migrate. The trauma of this latest migration is still felt so strongly that it is often superimposed and confused with the great migrations from Central and Western Africa that took place in mythological times, or very long ago. But through their histories, they keep alive their memories and the certainty of the unity of the Lala nation, even though it is now spread from Kranskop to Harding and IZingolweni. Chapter four examines the phonetics and phonology of the two Lala dialects in the context of Swati and Zulu. Lala is a Tekela dialect indeed, but with its own phonetic and phonological peculiarities. The same can be said with regard to the research exposed in chapter five, on the morphology of Lala compared to Swati and Zulu. The grammatical system among the three languages is extremely similar, and there is hardly any substantial difference between North and South Lala. This can also be said with regard to tonology (chapter six) The last chapter reflects on what has been achieved. The ancient Lala language, about which James Stuart stated in the 1920's that it was nearly extinct, is proving very strong and resilient. Possibly the very dynamism experienced in its ability to adapt to the colonial languages and the new material culture by assimilating many foreign lexical items, and that South Lala has adopted many words from IsiZansi, is proof that Lala has got a life of its own that cannot be taken for granted, nor extinguished. The more one is able to study the local languages, the more one dusts up some ancient treasure that needs to be admired and constantly re-valued. The functional word here is 'treasure': because all languages, as carriers of culture, are a treasure that together form the mosaic of our beautiful "Rainbow Nation". / African Languages / D.Litt et Phil.

Neste trabalho, investiga-se a distribuição dos sufixos -inho e -zinho nos dialetos de Porto Alegre e Curitiba. A partir de estudos que contemplaram tais afixos, olhamos para fatores de natureza lingüística, como classe gramatical (nomes, não-nomes), tonicidade (oxítona, paroxítona e proparoxítona), segmento final da forma primitiva (vogal baixa, vogal média-baixa frontal e posterior, -i, - u primitivos, -e, - o primitivos, -s / -z, -r / -m, -l), onset da sílaba final (nasal dorsal, labiais, coronais, dorsais, onset vazio), e para fatores de natureza extralingüística, como escolaridade (primário, ginásio, secundário), sexo (masculino e feminino), faixa etária (menos de 50 anos e mais de 50 anos), localidade (Porto Alegre e Curitiba). Esta pesquisa faz uso de dados de fala extraídos de 24 entrevistas do Projeto VARSUL e de um teste de produtividade, que se utiliza de pseudopalavras, aplicado a 20 informantes. Para a análise quantitativa dos resultados, fazemos um uso adaptado dos programas que compõem o pacote VARBRUL, lançando mão tão-somente de freqüências, já que não se trata de um fenômeno tipicamente variável. A análise dos resultados mostrou que -inho é o sufixo mais usado, ainda que -zinho seja o preferido no contexto de pseudopalavras. Além disso, observouse um padrão predominantemente alternante, conforme já previu a literatura, definido pela tonicidade, ainda que algum lugar para usos variáveis esteja reservado. / In this paper, the distribution of the suffixes -inho and -zinho is investigated in the dialects of Porto Alegre and Curitiba. Starting from studies that cover such affix, we looked at factors of linguistic nature, like grammatical class (names, nonames), tonicity (stressed on the last, second last, or third last syllable), final segment in the primitive form (low vowel, back and front low-mid vowels, -i, primitive -u, -e, primitive -o, -s / -z, -r / -m, -l), final syllable onset (dorsal nasal, labial, coronal, dorsal, empty onset), and extra linguistic nature factors, such as educational background (primary, elementary, high school), sex (male and female), age group (less than 50 and more than 50 years old), place (Porto Alegre and Curitiba). This research makes use of extracted speech data of 24 interviews from Projeto VARSUL and of a productivity test, which uses pseudo words, applied to 20 applicants. For the quantitative analysis of the results, we make an adapted use of the programs that compose the package VARBRUL, not taking into account only of frequencies numbers, since it is not a typically variable phenomenon. The analysis of the results showed that -inho is the most used suffix, although -zinho is the favorite in the context of pseudo words. Besides, a predominantly alternating pattern was observed, as already expected in literature, defined by the tonicity, although some place for variable uses may be reserved.

Neste trabalho, investiga-se a distribuição dos sufixos -inho e -zinho nos dialetos de Porto Alegre e Curitiba. A partir de estudos que contemplaram tais afixos, olhamos para fatores de natureza lingüística, como classe gramatical (nomes, não-nomes), tonicidade (oxítona, paroxítona e proparoxítona), segmento final da forma primitiva (vogal baixa, vogal média-baixa frontal e posterior, -i, - u primitivos, -e, - o primitivos, -s / -z, -r / -m, -l), onset da sílaba final (nasal dorsal, labiais, coronais, dorsais, onset vazio), e para fatores de natureza extralingüística, como escolaridade (primário, ginásio, secundário), sexo (masculino e feminino), faixa etária (menos de 50 anos e mais de 50 anos), localidade (Porto Alegre e Curitiba). Esta pesquisa faz uso de dados de fala extraídos de 24 entrevistas do Projeto VARSUL e de um teste de produtividade, que se utiliza de pseudopalavras, aplicado a 20 informantes. Para a análise quantitativa dos resultados, fazemos um uso adaptado dos programas que compõem o pacote VARBRUL, lançando mão tão-somente de freqüências, já que não se trata de um fenômeno tipicamente variável. A análise dos resultados mostrou que -inho é o sufixo mais usado, ainda que -zinho seja o preferido no contexto de pseudopalavras. Além disso, observouse um padrão predominantemente alternante, conforme já previu a literatura, definido pela tonicidade, ainda que algum lugar para usos variáveis esteja reservado. / In this paper, the distribution of the suffixes -inho and -zinho is investigated in the dialects of Porto Alegre and Curitiba. Starting from studies that cover such affix, we looked at factors of linguistic nature, like grammatical class (names, nonames), tonicity (stressed on the last, second last, or third last syllable), final segment in the primitive form (low vowel, back and front low-mid vowels, -i, primitive -u, -e, primitive -o, -s / -z, -r / -m, -l), final syllable onset (dorsal nasal, labial, coronal, dorsal, empty onset), and extra linguistic nature factors, such as educational background (primary, elementary, high school), sex (male and female), age group (less than 50 and more than 50 years old), place (Porto Alegre and Curitiba). This research makes use of extracted speech data of 24 interviews from Projeto VARSUL and of a productivity test, which uses pseudo words, applied to 20 applicants. For the quantitative analysis of the results, we make an adapted use of the programs that compose the package VARBRUL, not taking into account only of frequencies numbers, since it is not a typically variable phenomenon. The analysis of the results showed that -inho is the most used suffix, although -zinho is the favorite in the context of pseudo words. Besides, a predominantly alternating pattern was observed, as already expected in literature, defined by the tonicity, although some place for variable uses may be reserved.

Este trabalho trata da descrição e da análise do fraseamento entoacional do contorno de sentenças declarativas neutras do português falado na Guiné-Bissau (PGB), no que se refere, especificamente, à investigação da relação entre atribuição de eventos tonais ao contorno entoacional e formação de domínios prosódicos. Além disso, faz-se a comparação dos resultados obtidos para o fraseamento entoacional dos dados do PGB com os resultados obtidos para o fraseamento entoacional já descritos em trabalhos anteriores para o português brasileiro (PB) e para o português europeu (PE). Os corpora utilizados para o desenvolvimento deste estudo derivam de dois tipos de discurso (sentenças de leitura e de fala espontânea). A descrição e a análise do fraseamento entoacional das sentenças declarativas neutras contidas nesses corpora é desenvolvido à luz da abordagem Autossegmental e Métrica da fonologia entoacional (PIERREHUMBERT, 1980; BECKMAN; PIERREHUMBERT, 1986; LADD, 1996, 2008; entre outros) e da Fonologia Prosódica (SELKIRK, 1984, 1986, 2000; NESPOR; VOGEL, 1986, 2007; entre outros) e com base em estudos prévios desenvolvidos nesses mesmos quadros teóricos e aplicados ao PB (CUNHA 2000; FROTA; VIGÁRIO 2000; TENANI 2002; FERNANDES, 2007a, 2007b; SERRA 2009; entre outros) e ao PE (GRØNNUM; VIANA, 1999; FROTA, 2000, 2002a, 2002b, 2003; VIGÁRIO, 2003; VIGÁRIO; FROTA, 2003; CRUZ, 2013; entre outros). Os resultados alcançados nesta dissertação revelam características entoacionais do PGB que são similares às encontradas nas demais variedades de português: (i) a frequente atribuição de acentos tonais a palavras fonológicas do contorno entoacional (característica também encontrada no contorno entoacional do PB); (ii) a associação obrigatória de um acento tonal à palavra fonológica cabeça do último sintagma fonológico do sintagma entoacional, seguido por um tom de fronteira associado à fronteira direita desse sintagma (assim como no PB e no PE); (iii) a correspondência do padrão do contorno nuclear do sintagma entoacional das declarativas neutras do PGB com os padrões do contorno nuclear desse mesmo sintagma das sentenças declarativas neutras do PB e das variedades do PE; (iv) a possibilidade de associação de acentos frasais a fronteiras de sintagmas fonológicos (evento tonal que é encontrado associado a sintagmas fonológicos desse mesmo tipo de sentenças em uma das variedades centro-meridionais do PE); e (v) a ocorrência de eventos tonais adicionais H associados a sílabas pretônicas de palavras fonológicas longas do contorno entoacional (semelhante aos tons adicionais encontrados associados a palavras fonológicas também longas no PB). / In this study, we investigate the intonation of neutral declarative sentences of Portuguese of Guinea-Bissau (GBP), with regarding to the investigation of the relation between tonal events assignment and prosodic domains formation. In addition, we compare the results obtained for the analysis of GBP data with the intonational patterns of neutral sentences described on previous works for Brazilian Portuguese (BP) and European Portuguese (EP). For this research, a corpus of two different speech styles (read sentences and spontaneous speech sentences) was used. For the description and analysis of intonational of neutral declarative sentences in this corpus we followed the Autosegmental Metrical aproach within the intonational phonology framework (PIERREHUMBERT, 1980; BECKMAN; PIERREHUMBERT, 1986; LADD, 1996, 2008; among others), the Prosodic Phonology framework (SELKIRK, 1984, 1986, 2000; NESPOR; VOGEL 1986, 2007; among others) and previous studies conducted on those theoretical frameworks and applied to BP (CUNHA; 2000; FROTA; VIGÁRIO, 2000; TENANI 2002; FERNANDES, 2007a, 2007b; SERRA 2009; among others) and EP (GRØNNUM; VIANA, 1999; FROTA, 2000, 2002a, 2002b, 2003; VIGÁRIO, 2003; FROTA; VIGÁRIO, 2003; CRUZ, 2013; among others). The results achieved in this work show that the intonation properties of GBP are similar to those found in the other varieties of Portuguese already studied. These properties are as follow: (i) frequent association of pitch accents with Phonological Words of intonation contour (tonal characteristic also found in BP); (ii) a pitch accent is obligatorily associated with the Phonological Word head of the last Phonological Phrase of a Intonation Phrase, followed by a boundary tone associated with the right edge of that Intonation Phrase (as in BP and EP); (iii) there is a matching in the nuclear contour of GPB neutral declarative sentences with the nuclear contour of BP and varieties of EP neutral declarative sentences; (iv) the possibility of phrasal accents to be associated with phonological phrases boundaries (the same possibility is found for one of the center-southern varieties of EP); and (v) the possibility of H tones to be associated with pretonic syllables of long Prosodic Words (as in BP).