This thesis studies the inter-group variation of the consonant endings among five principal subgroups of the Chaoshan dialect, a branch of the South Min dialect in Eastern Guangdong Province, from the perspective of language contact and horizontal transmission. I conduct a quantitative study to present the synchronic variance of the consonant endings among five Chaoshan subgroups and the diachronic variance from Middle Chinese to modern Chaoshan dialect on a numerical scale. The current literature tends to take the change of the consonant endings as a process of weakening governed by regular rules. My research findings challenge this conventional view. First, the change of the consonant endings from Middle Chinese to five subgroups of modern Chaoshan dialect is irregular, which is an exception to the linguistic laws proposed in the existing literature. Secondly, I find that some characters without consonant endings or with a weakened ending in Chaozhou in the 19th century reverse to have a consonant ending in modern Chaoshan dialect. This reversal contradicts to the weakening hypothesis that regards the change of the consonant endings as a process of simplifying. Finally, my quantitative research shows that Chaozhou dialect in the 19th century in much closer to modern Xiamen dialect than to five subgroups of modern Chaoshan dialect in terms of the relativeness in consonant endings, which is the opposite to the prediction that languages become more and more different and have no consequent contact with other daughter languages after separating from the proto-language. We propose that the actual situation of the consonant endings in different subgroups of the Chaoshan dialect can be better explained from the perspective of language contact and horizontal transmission. The interaction between Han Chinese and non-Han Chinese is the primary reason for the change of the consonant endings of the Chaoshan dialect. Also, the language contact between Chaoshan aborigines and migrants from the Fujian Province leads to the divergence of the consonant endings in different Chaoshan subgroups.Population structure and other social factors determine what phonetic features survive after several times of horizontal transmission.

The present study is an investigation of acoustic correlates corresponding to the category [voice] in two dialects of Venezuelan Spanish. The Andean mountain dialect Mérida (MER) and Caribbean coastal dialect Margarita (MAR) are thought to differ systematically in the phonetic implementation of the Spanish phonological stop series along the lines of lowland and highland divides commonly reported for Latin American Spanish. Specifically, MER has been characterized by a greater percentage of occlusive pronunciations, MAR by more fricative and/or approximant realizations of phonological stops. To test what repercussions these differences in consonant articulation have on the acoustic correlates that encode [voice], a production experiment was run. Informants were 25 adult monolingual speakers of Venezuelan Spanish from the areas of El Tirano (Margarita Island) and San Rafael de Mucuchíes (Mérida state). The materials were 44 CV syllable prompts. Target syllables were analyzed with respect to the following: consonant closure duration, VOT, %VF, RMS, preceding vowel duration, CV ratio, F1 onset frequency, F0 contour, and burst. Statistical analysis using a linear mixed model ANOVA tested for fixed effects of voicing category, dialect and condition (speeded/unspeeded) and interactions of voicing category * dialect and dialect * condition. Results showed that the dialects MER and MAR vary significantly in RMS. In addition, the following correlates were significant for the interaction of voicing category * dialect: consonant duration, VOT, %VF, RMS, CV ratio and burst. Generally, the nature of the differences indicates a greater separation between [± voice] values in MER than in MAR (notably divergent are VOT and RMS). These results imply that while the same acoustic correlates of [voice] are operative in both fortis and lenis dialects of Spanish, [± voice] categories relate differently. Furthermore, with regard to prosody and rate of speech, most significant differences in condition occurred in initial position while most significant differences in the interaction of voicing category * dialect were linked to medial position. The results of this study are relevant to current research on the specifics of dialectal variation in consonant systems. They also have wider implications for the general mapping of phonetics to phonology in speech. / text

Den underliggande idén för denna studie är frågan om en ljudrelaterad grammatik kan undervisas via fonologiska övningar och diktamen för att förbättra den mekaniska skrivförmågan och om språkundervisningsmetoder för årskurs 1–2 kan också utnyttjas för äldre elever. Huvudsyftet med den här studien är att undersöka hur sverigefinska elever behärskar ljud av standardfinska med fokus på långa och korta vokaler och konsonanter samt diftonger. Dessutom är målet att testa en diktamensmetod som bedömer utgångsnivån av elevernas färdighetsnivå för att skriva korta och långa vokaler och konsonanter samt diftonger. Forskningsfrågorna är följande: 1) Vilka slags avvikelser jämfört med standardfinska finns i sverigefinska elevers skrivning av korta och långa vokaler och konsonanter samt diftonger? 2) Hur mycket avvikelser jämfört med standard finska finns i sverigefinska elevers skrivning av korta och långa vokaler och konsonanter samt diftonger? 3) Vilken typ av korrelation kan man hitta mellan avvikelser jämfört med standard finska och externa bakrundsvariabler. Denna studie av vilken teoretisk bakgrund läggs på fonologi, pedagogik och sociolingvistik har ett särskilt fokus på sverigefinsk kultur och flerspråkighet. Huvudmaterialet samlades in genom en diktamensövning med 22 skolelever i åldern 9–13 år och analyserades med hjälp av ordstrukturanalys. Utöver detta analyserades ett berättande textprov från tio elever. Bakgrundsinformation samlades in från föräldrar och jämfördes med resultaten av ordstrukturanalysen med hjälp av en implikationsskala. Studien visar att de viktigaste avvikelserna kring långa vokaler och konsonanter är de saknade bokstäverna, dvs enkelskrivning av ljuden. De mest utmanande långa vokalerna är oo, ee, yy och ää och de långa konsonanterna ss, rr och nn. Dessutom sker omkastning mellan bokstäver: för vokaler u > o, o > å och y > u och för konsonanter: m > n och kk > ck. De mest utmanande diftongerna är yi, ey, ie, uo, yö, öy, äy och iy och de lättaste att hantera är ai, oi, au, eu och öi. Dessutom blandas ibland o ihop u, u ihop y och ä ihop e. Det verkar som att bakgrundsvariablerna inte visar någon betydande korrelation med de språkliga färdigheterna vid hantering av vokaler, konsonanter och diftonger. Hur som helst säger en försiktig tolkning att det verkar ha en positiv effekt om man har lärt sig språket under de första tre åren och om man använder språket med släktingar. / The underlying idea of this study is a question if phoneme related grammar can be taught via phonological exercises and dictations in order to improve mechanical writing skills and if language teaching methods for grade 1–2 can also be utilized for older students. The main objective of this study is to study the command of standard Finnish phonemes among Sweden Finn pupils, focusing on long and short vowels and consonants, and diphthongs. In addition, the target is to test a dictation method which assesses the starting level of pupils’ competence in writing short and long vowels and consonants, and diphthongs. The research questions are the following: 1) What kind of deviations compared to standard Finnish can be found in Sweden Finn pupils’ writing of short and long vowels and consonants and diphthongs? 2) How much of the deviations compared to standard Finnish can be found in Sweden Finn pupils’ writing of short and long vowels and consonants and diphthongs? 3) What kind of correlation can be found between these deviations, compared to standard Finnish and the background variables? The theoretical background of this study is based on phonology, educational science and sociolinguistic, and it has its special focus on Swedish-Finnish culture and multilingualism. The main data was collected with a dictation from 22 pupils aged 9–13 years old and analyzed by using word structure analysis. In addition to this the narratives of text samples by ten pupils were analyzed. Background information was collected from parents and compared with the results of the word structure analysis, by using an implication scale. The study shows that the main deviations regarding long vowels and consonants are missing letters, that is, the use of single letters where two are expected.  The most challenging long vowels are oo, ee, yy and ää and the long consonants ss, rr and nn. In addition, a replacement between letters appears: with vowels u > o, o > å and y > u and with consonants: m > n and kk > ck.The most challenging diphthongs are yi, ey, ie, uo, yö, öy, äy and iy, and the easiest to manage are ai, oi, au, eu and öi. Besides this some disorders between o and u, u and y and ä and e appear. It seems that background variables do not show any significant dependencies on competences in managing vowels, consonants and diphthongs. However with a cautious interpretation, there seems to be a positive effect if the language have been learnt during the first three years and if the language is used with relatives.

Nesta Tese, investigamos as consoantes geminadas no italiano a partir da produção dessas em dados coletados de falantes italianos e de aprendizes brasileiros, alunos das disciplinas de graduação oferecidas pelo Setor de Italiano do Instituto de Letras da Universidade Federal do Rio Grande do Sul. A questão geral que norteia nossa análise acústica e fonológica é verificar se falantes brasileiros, aprendizes de italiano, adquirem essas consoantes. As geminadas são caracterizadas foneticamente como consoantes longas (Ridouane, 2010) e, fonologicamente, como portadoras de mora (McCarthy & Prince, 1986; Hayes, 1989; Morén, 1999). No italiano, esses segmentos têm função distintiva (Dardano & Trifone, 1999) quando se encontram no interior de palavras, mas não quando emergem na frase, entre palavras. Neste ambiente, a geminação consonântica é provocada pela gramática, em respeito a condições de boa formação prosódica. Com base nessas considerações, analisamos dados de realização de geminadas, especificamente coletados e analisados por meio do software PRAAT, versão 5.1.42 (Boersma & Weenink, 2010). A análise da duração das consoantes nos dados dos italianos e dos brasileiros nos possibilitou visualizar que: (1) o contexto prosódico interfere na duração dos segmentos; (2) embora o contexto prosódico interfira na duração, o contraste é preservado nos dados dos participantes italianos; (3) a princípio, os aprendizes relacionam a duração da consoante à distribuição do acento primário, mas, com o progresso no curso de italiano, os aprendizes passam a produzir, além das consoantes geminadas em ambiente tônico, geminadas em ambiente átono, embora não categoricamente. Construímos nossa análise com base nos pressupostos da Teoria da Otimidade (Prince & Smolensky, 1993/ 2004). Seguindo as predições da Teoria Mórica (Hayes, 1989) e a proposta de Morén (1999), mostramos, com base na literatura e em nossos resultados, que a aquisição de geminadas pelos aprendizes brasileiros depende do ranking alto da restrição de fidelidade MAXLINK-MORA[CON], ativa na gramática dos participantes italianos. As geminadas que ramificam sílabas tônicas são encontradas inclusive nos dados dos aprendizes brasileiros matriculados nos níveis iniciais do curso de italiano. Neste caso, o que garante sua produção é o ranking alto de WSP, que associa peso silábico a acento, em interação com as restrições de marcação FTBIN, que requer pés binários, *LONGV, que marca violação a candidatos com vogal longa, juntamente com a restrição EDGEMOST-R, que milita a favor do alinhamento do acento com a margem direita da palavra. A quantidade consonântica provocada pela gramática em favor da formação de pés harmônicos (como ocorre no caso da geminação na frase) corresponde à situação de mora por coerção (Morén, 1999). Nossa análise mostra que, nos dados dos aprendizes, a mora das consoantes geminadas que travam sílabas acentuadas é mantida pelo alto ranking das restrições de marcação que militam a favor de pés bem-formados e da limitação do acento primário à janela de três sílabas. Em relação à geminação na frase, apesar da relação entre acento e geminação, o fenômeno é opcional nos dados de um dos falantes italianos. Supomos que a posição de NOCODA/ (adaptada de Flack, 2009), sem ranking estrito em relação a WSP, justifique sua opcionalidade, apresentada também nos dados dos aprendizes brasileiros. Nossos resultados nos levam a crer que a opcionalidade de outputs com Raddoppiamento Sintattico (Nespor & Vogel, 1986) e a aquisição não categórica de geminadas lexicais indiquem que o algoritmo de aprendizagem seja gradual (Boersma, 1997; Boersma & Hayes, 2000). / In this dissertation, we investigate the geminate consonants of Italian using production data collected from Italian speakers and Brazilian learners who take undergraduate courses at the Italian Section of the Federal University of Rio Grande do Sul (UFRGS). The general question that guided our acoustic and phonological analysis is whether Brazilian speakers who are learners of Italian acquire such consonants. Geminates are characterized phonetically as long consonants (Ridouane, 2010) and, phonologically, they are mora-bearing (McCarthy & Prince, 1986; Hayes, 1989; Morén, 1999). In Italian, these segments are distinctive (Dardano & Trifone, 1999) in word-internal position, but not when they emerge at the phrase level, between words. In this environment, consonant germination is created by grammar, enforcing prosodic well-formedness conditions. Based on these facts, we have analyzed production data for geminates. Specifically, the data were collected and analyzed with PRAAT (version 5.1.42) software (Boersma & Weenink, 2010). The analysis of consonant length in the Italian and Brazilian data showed that: (1) prosodic context interferes with the length of segments; (2) even though prosodic context interferes with length, contrast is preserved in the data from Italian participants; (3) initially, learners tend to associate consonant length with the placement of primary stress; however, as they advance through the course, learners start to produce, in addition to geminate consonants in stressed environments, geminates in unstressed environments, though not categorically. We have developed our analysis based on the principles of Optimality Theory (Prince & Smolensky, 1993/2004). Following the predictions of Moraic Theory (Hayes, 1989) and the proposal of Morén (1999), we show, based on the relevant literature and our own results, that the acquisition of geminates by Brazilian learners depends on a high ranking of the MAXLINK-MORA[CON] faithfulness constraint, which is active in the grammar of Italian participants. Geminates that cause branching in stressed syllables can be found even in the data from Brazilian learners enrolled in beginner-level Italian courses. In this case, what ensures their production is the high ranking of WSP, which associates syllable weight with stress, and interacts with markedness constraints FTBIN, related to foot binarity, *LONGV, which assigns a violation to candidates with long vowels, as well as EDGEMOST-R, which militates in favor of alignment of the stress with the right margin of the word. The consonant quantity enforced by the grammar in favor of the creation of harmonic feet (as in the case of phrase gemination) corresponds to the scenario of coerced moraicity (Morén, 1999). Our analysis shows that, in the data of learners, the moras of geminate consonants at the end of stressed syllables are preserved due to the high ranking of markedness constraints that militate in favor of well-formed feet and to the limitation of primary stress to the three-syllable window. Concerning phrase gemination, in spite of the relationship between stress and gemination, such a phenomenon was optional in the data from one of the Italian speakers. We assume that the position of NOCODA/ (adapted from Flack, 2009), without a strict ranking with respect to WSP, justifies its optionality, which also appears in the data of Brazilian learners. Our results lead us to believe that the optionality of outputs with Raddoppiamento Sintattico (Nespor & Vogel, 1986) and the non-categorical acquisition of lexical geminates indicate that the learning algorithm is gradual (Boersma, 1997; Boersma & Hayes, 2001).

Ce travail se focalise sur le parler du tarifit de la province de Nador, afin d’étudier la gémination dans cette variante du berbère, parlée au Maroc. Il s’agit d’une investigation articulatoire cinéradiographique et acoustique. Un aperçu général du système phonologique du tarifit est proposé. Dans la revue critique de l’état de la question, nous avons tenté, d’abord, de montrer les différents arguments relatifs à la représentation des géminées en un seul segment ou en une suite de deux segments. Nous avons ensuite évoqué les débats contradictoires sur les notions de tension et de gémination. Notre recherche est conduite dans le cadre du paradigme de la perturbation et des réajustements, en variant la vitesse d’élocution. Les résultats principaux dans le domaine acoustique montrent, pour toutes les consonnes, simples et géminées, produites en vitesse d'élocution normale ou rapide, que la durée de la tenue consonantique est l’indice principal qui permet de les distinguer. Au niveau articulatoire, l’étude du timing des paramètres articulatoires indique, entre autres, que c’est le contact apical, vélaire et uvulaire, plus long pour la géminée, qui est le paramètre préférentiel de la distinction phonologique simple vs. géminée. L’analyse des vues de profil donne les résultats suivants en ce qui concerne l’étendue de contact (mm) : a) l’étendue de contact des occlusives est systématiquement plus importante pour les géminées que pour les simples ; b) l’étendue de contact augmente de la consonne apical, au vélaire (réalisée plutôt palatale), puis à l’uvulaire. Les résultats sont discutés en termes de relations articulatori - acoustiques. / This work focuses on tarifit of the provinces of Nador, in order to study gemination in this variant of Berber spoken in Morocco. This is an acoustic and articulatory X-ray investigation. A general overview of the tarifit phonological system is proposed. In a critical review of the literature, we tried, first, to show the different arguments concerning representation of geminates as one or as a sequence of two segments. We then discussed the contradictory debates on the concepts of tension and gemination. Our research is carried out within the perturbation and readjustments paradigm, by varying speech rates. Main results in the acoustic domain show for all consonants, singletons and geminates, produced in normal or fast speech, that consonantal closure is the main cue that allows distinguishing them. On the articulatory level, the study of the timing of articulatory parameters indicates, among other things, that it is the apical, velar and uvular contact, longer for geminates, which is the preferred parameter of the singleton vs. geminate phonological distinction. Analysis of profile views gives the following results regarding contact extent (mm) : a) contact extent for plosives are systematically larger for geminates than for singletons ; b) contact extent increases as one goes from the apical consonant to the velar (rather palatal) consonant, then to the uvular consonant. Results are discussed in terms of articulatory - acoustic relations.

Production errors made by second language (L2) learners of English have been attributed to markedness, L1 transfer or input frequency (cf. Major, 2001; Edwards & Zampini, 2008; Baptista, Rauber, & Watkins, 2009). This thesis examines the production of 17 English initial consonant clusters (e.g., /pr/ in “pray”) in a markedness relationship, whereby clusters with greater sonority distance between the first and second consonants are unmarked and clusters with smaller sonority distance between the first and second consonants are marked, by two groups of Saudi Arabian L2 English learners. It also explores the effect of input frequency and L1 transfer. Participants were asked to read 60 sentences and their reading was recorded for acoustic analysis. Analysis showed that “prothesis” was always used to simplify the clusters, and that, the duration of the prothetic vowel tended to get longer when clusters become more marked. Intermediate participants had greater degree of difficulty in producing the clusters and tended to insert a longer prothetic vowel in general. Markedness explained the performance on #sC clusters; however, performance on non #sC clusters was best explained by L1 transfer. Results further indicated that input frequency was irrelevant to this study.

Nesta Tese, investigamos as consoantes geminadas no italiano a partir da produção dessas em dados coletados de falantes italianos e de aprendizes brasileiros, alunos das disciplinas de graduação oferecidas pelo Setor de Italiano do Instituto de Letras da Universidade Federal do Rio Grande do Sul. A questão geral que norteia nossa análise acústica e fonológica é verificar se falantes brasileiros, aprendizes de italiano, adquirem essas consoantes. As geminadas são caracterizadas foneticamente como consoantes longas (Ridouane, 2010) e, fonologicamente, como portadoras de mora (McCarthy & Prince, 1986; Hayes, 1989; Morén, 1999). No italiano, esses segmentos têm função distintiva (Dardano & Trifone, 1999) quando se encontram no interior de palavras, mas não quando emergem na frase, entre palavras. Neste ambiente, a geminação consonântica é provocada pela gramática, em respeito a condições de boa formação prosódica. Com base nessas considerações, analisamos dados de realização de geminadas, especificamente coletados e analisados por meio do software PRAAT, versão 5.1.42 (Boersma & Weenink, 2010). A análise da duração das consoantes nos dados dos italianos e dos brasileiros nos possibilitou visualizar que: (1) o contexto prosódico interfere na duração dos segmentos; (2) embora o contexto prosódico interfira na duração, o contraste é preservado nos dados dos participantes italianos; (3) a princípio, os aprendizes relacionam a duração da consoante à distribuição do acento primário, mas, com o progresso no curso de italiano, os aprendizes passam a produzir, além das consoantes geminadas em ambiente tônico, geminadas em ambiente átono, embora não categoricamente. Construímos nossa análise com base nos pressupostos da Teoria da Otimidade (Prince & Smolensky, 1993/ 2004). Seguindo as predições da Teoria Mórica (Hayes, 1989) e a proposta de Morén (1999), mostramos, com base na literatura e em nossos resultados, que a aquisição de geminadas pelos aprendizes brasileiros depende do ranking alto da restrição de fidelidade MAXLINK-MORA[CON], ativa na gramática dos participantes italianos. As geminadas que ramificam sílabas tônicas são encontradas inclusive nos dados dos aprendizes brasileiros matriculados nos níveis iniciais do curso de italiano. Neste caso, o que garante sua produção é o ranking alto de WSP, que associa peso silábico a acento, em interação com as restrições de marcação FTBIN, que requer pés binários, *LONGV, que marca violação a candidatos com vogal longa, juntamente com a restrição EDGEMOST-R, que milita a favor do alinhamento do acento com a margem direita da palavra. A quantidade consonântica provocada pela gramática em favor da formação de pés harmônicos (como ocorre no caso da geminação na frase) corresponde à situação de mora por coerção (Morén, 1999). Nossa análise mostra que, nos dados dos aprendizes, a mora das consoantes geminadas que travam sílabas acentuadas é mantida pelo alto ranking das restrições de marcação que militam a favor de pés bem-formados e da limitação do acento primário à janela de três sílabas. Em relação à geminação na frase, apesar da relação entre acento e geminação, o fenômeno é opcional nos dados de um dos falantes italianos. Supomos que a posição de NOCODA/ (adaptada de Flack, 2009), sem ranking estrito em relação a WSP, justifique sua opcionalidade, apresentada também nos dados dos aprendizes brasileiros. Nossos resultados nos levam a crer que a opcionalidade de outputs com Raddoppiamento Sintattico (Nespor & Vogel, 1986) e a aquisição não categórica de geminadas lexicais indiquem que o algoritmo de aprendizagem seja gradual (Boersma, 1997; Boersma & Hayes, 2000). / In this dissertation, we investigate the geminate consonants of Italian using production data collected from Italian speakers and Brazilian learners who take undergraduate courses at the Italian Section of the Federal University of Rio Grande do Sul (UFRGS). The general question that guided our acoustic and phonological analysis is whether Brazilian speakers who are learners of Italian acquire such consonants. Geminates are characterized phonetically as long consonants (Ridouane, 2010) and, phonologically, they are mora-bearing (McCarthy & Prince, 1986; Hayes, 1989; Morén, 1999). In Italian, these segments are distinctive (Dardano & Trifone, 1999) in word-internal position, but not when they emerge at the phrase level, between words. In this environment, consonant germination is created by grammar, enforcing prosodic well-formedness conditions. Based on these facts, we have analyzed production data for geminates. Specifically, the data were collected and analyzed with PRAAT (version 5.1.42) software (Boersma & Weenink, 2010). The analysis of consonant length in the Italian and Brazilian data showed that: (1) prosodic context interferes with the length of segments; (2) even though prosodic context interferes with length, contrast is preserved in the data from Italian participants; (3) initially, learners tend to associate consonant length with the placement of primary stress; however, as they advance through the course, learners start to produce, in addition to geminate consonants in stressed environments, geminates in unstressed environments, though not categorically. We have developed our analysis based on the principles of Optimality Theory (Prince & Smolensky, 1993/2004). Following the predictions of Moraic Theory (Hayes, 1989) and the proposal of Morén (1999), we show, based on the relevant literature and our own results, that the acquisition of geminates by Brazilian learners depends on a high ranking of the MAXLINK-MORA[CON] faithfulness constraint, which is active in the grammar of Italian participants. Geminates that cause branching in stressed syllables can be found even in the data from Brazilian learners enrolled in beginner-level Italian courses. In this case, what ensures their production is the high ranking of WSP, which associates syllable weight with stress, and interacts with markedness constraints FTBIN, related to foot binarity, *LONGV, which assigns a violation to candidates with long vowels, as well as EDGEMOST-R, which militates in favor of alignment of the stress with the right margin of the word. The consonant quantity enforced by the grammar in favor of the creation of harmonic feet (as in the case of phrase gemination) corresponds to the scenario of coerced moraicity (Morén, 1999). Our analysis shows that, in the data of learners, the moras of geminate consonants at the end of stressed syllables are preserved due to the high ranking of markedness constraints that militate in favor of well-formed feet and to the limitation of primary stress to the three-syllable window. Concerning phrase gemination, in spite of the relationship between stress and gemination, such a phenomenon was optional in the data from one of the Italian speakers. We assume that the position of NOCODA/ (adapted from Flack, 2009), without a strict ranking with respect to WSP, justifies its optionality, which also appears in the data of Brazilian learners. Our results lead us to believe that the optionality of outputs with Raddoppiamento Sintattico (Nespor & Vogel, 1986) and the non-categorical acquisition of lexical geminates indicate that the learning algorithm is gradual (Boersma, 1997; Boersma & Hayes, 2001).

Nesta Tese, investigamos as consoantes geminadas no italiano a partir da produção dessas em dados coletados de falantes italianos e de aprendizes brasileiros, alunos das disciplinas de graduação oferecidas pelo Setor de Italiano do Instituto de Letras da Universidade Federal do Rio Grande do Sul. A questão geral que norteia nossa análise acústica e fonológica é verificar se falantes brasileiros, aprendizes de italiano, adquirem essas consoantes. As geminadas são caracterizadas foneticamente como consoantes longas (Ridouane, 2010) e, fonologicamente, como portadoras de mora (McCarthy & Prince, 1986; Hayes, 1989; Morén, 1999). No italiano, esses segmentos têm função distintiva (Dardano & Trifone, 1999) quando se encontram no interior de palavras, mas não quando emergem na frase, entre palavras. Neste ambiente, a geminação consonântica é provocada pela gramática, em respeito a condições de boa formação prosódica. Com base nessas considerações, analisamos dados de realização de geminadas, especificamente coletados e analisados por meio do software PRAAT, versão 5.1.42 (Boersma & Weenink, 2010). A análise da duração das consoantes nos dados dos italianos e dos brasileiros nos possibilitou visualizar que: (1) o contexto prosódico interfere na duração dos segmentos; (2) embora o contexto prosódico interfira na duração, o contraste é preservado nos dados dos participantes italianos; (3) a princípio, os aprendizes relacionam a duração da consoante à distribuição do acento primário, mas, com o progresso no curso de italiano, os aprendizes passam a produzir, além das consoantes geminadas em ambiente tônico, geminadas em ambiente átono, embora não categoricamente. Construímos nossa análise com base nos pressupostos da Teoria da Otimidade (Prince & Smolensky, 1993/ 2004). Seguindo as predições da Teoria Mórica (Hayes, 1989) e a proposta de Morén (1999), mostramos, com base na literatura e em nossos resultados, que a aquisição de geminadas pelos aprendizes brasileiros depende do ranking alto da restrição de fidelidade MAXLINK-MORA[CON], ativa na gramática dos participantes italianos. As geminadas que ramificam sílabas tônicas são encontradas inclusive nos dados dos aprendizes brasileiros matriculados nos níveis iniciais do curso de italiano. Neste caso, o que garante sua produção é o ranking alto de WSP, que associa peso silábico a acento, em interação com as restrições de marcação FTBIN, que requer pés binários, *LONGV, que marca violação a candidatos com vogal longa, juntamente com a restrição EDGEMOST-R, que milita a favor do alinhamento do acento com a margem direita da palavra. A quantidade consonântica provocada pela gramática em favor da formação de pés harmônicos (como ocorre no caso da geminação na frase) corresponde à situação de mora por coerção (Morén, 1999). Nossa análise mostra que, nos dados dos aprendizes, a mora das consoantes geminadas que travam sílabas acentuadas é mantida pelo alto ranking das restrições de marcação que militam a favor de pés bem-formados e da limitação do acento primário à janela de três sílabas. Em relação à geminação na frase, apesar da relação entre acento e geminação, o fenômeno é opcional nos dados de um dos falantes italianos. Supomos que a posição de NOCODA/ (adaptada de Flack, 2009), sem ranking estrito em relação a WSP, justifique sua opcionalidade, apresentada também nos dados dos aprendizes brasileiros. Nossos resultados nos levam a crer que a opcionalidade de outputs com Raddoppiamento Sintattico (Nespor & Vogel, 1986) e a aquisição não categórica de geminadas lexicais indiquem que o algoritmo de aprendizagem seja gradual (Boersma, 1997; Boersma & Hayes, 2000). / In this dissertation, we investigate the geminate consonants of Italian using production data collected from Italian speakers and Brazilian learners who take undergraduate courses at the Italian Section of the Federal University of Rio Grande do Sul (UFRGS). The general question that guided our acoustic and phonological analysis is whether Brazilian speakers who are learners of Italian acquire such consonants. Geminates are characterized phonetically as long consonants (Ridouane, 2010) and, phonologically, they are mora-bearing (McCarthy & Prince, 1986; Hayes, 1989; Morén, 1999). In Italian, these segments are distinctive (Dardano & Trifone, 1999) in word-internal position, but not when they emerge at the phrase level, between words. In this environment, consonant germination is created by grammar, enforcing prosodic well-formedness conditions. Based on these facts, we have analyzed production data for geminates. Specifically, the data were collected and analyzed with PRAAT (version 5.1.42) software (Boersma & Weenink, 2010). The analysis of consonant length in the Italian and Brazilian data showed that: (1) prosodic context interferes with the length of segments; (2) even though prosodic context interferes with length, contrast is preserved in the data from Italian participants; (3) initially, learners tend to associate consonant length with the placement of primary stress; however, as they advance through the course, learners start to produce, in addition to geminate consonants in stressed environments, geminates in unstressed environments, though not categorically. We have developed our analysis based on the principles of Optimality Theory (Prince & Smolensky, 1993/2004). Following the predictions of Moraic Theory (Hayes, 1989) and the proposal of Morén (1999), we show, based on the relevant literature and our own results, that the acquisition of geminates by Brazilian learners depends on a high ranking of the MAXLINK-MORA[CON] faithfulness constraint, which is active in the grammar of Italian participants. Geminates that cause branching in stressed syllables can be found even in the data from Brazilian learners enrolled in beginner-level Italian courses. In this case, what ensures their production is the high ranking of WSP, which associates syllable weight with stress, and interacts with markedness constraints FTBIN, related to foot binarity, *LONGV, which assigns a violation to candidates with long vowels, as well as EDGEMOST-R, which militates in favor of alignment of the stress with the right margin of the word. The consonant quantity enforced by the grammar in favor of the creation of harmonic feet (as in the case of phrase gemination) corresponds to the scenario of coerced moraicity (Morén, 1999). Our analysis shows that, in the data of learners, the moras of geminate consonants at the end of stressed syllables are preserved due to the high ranking of markedness constraints that militate in favor of well-formed feet and to the limitation of primary stress to the three-syllable window. Concerning phrase gemination, in spite of the relationship between stress and gemination, such a phenomenon was optional in the data from one of the Italian speakers. We assume that the position of NOCODA/ (adapted from Flack, 2009), without a strict ranking with respect to WSP, justifies its optionality, which also appears in the data of Brazilian learners. Our results lead us to believe that the optionality of outputs with Raddoppiamento Sintattico (Nespor & Vogel, 1986) and the non-categorical acquisition of lexical geminates indicate that the learning algorithm is gradual (Boersma, 1997; Boersma & Hayes, 2001).

Submitted by Aline Batista (alinehb.ufpel@gmail.com) on 2016-06-29T19:34:38Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) A dinâmica dos gestos articulatórios da líquida lateral palatal.pdf: 6446781 bytes, checksum: 47d086436a73ebb11f6b2217fdbf858d (MD5) / Approved for entry into archive by Aline Batista (alinehb.ufpel@gmail.com) on 2016-06-30T20:24:48Z (GMT) No. of bitstreams: 2 A dinâmica dos gestos articulatórios da líquida lateral palatal.pdf: 6446781 bytes, checksum: 47d086436a73ebb11f6b2217fdbf858d (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2016-06-30T20:25:05Z (GMT). No. of bitstreams: 2 A dinâmica dos gestos articulatórios da líquida lateral palatal.pdf: 6446781 bytes, checksum: 47d086436a73ebb11f6b2217fdbf858d (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2016-06-06 / Sem bolsa / Devido as suas características linguísticas, articulatórias e acústicas, as laterais são concebidas como sons complexos. Promovem curiosidades, dúvidas e controvérsias (LADEFOGED, 1986; SILVA, 1996 e ALBANO, 2001) no meio acadêmico, mas são pouco investigadas. De fato, descrições da consoante líquida lateral palatal, por exemplo, ainda são escassas no Português Brasileiro (PB), sobretudo em populações atípicas. A presente pesquisa tem por objetivo a investigação da produção dessa consoante por usuários de implante coclear e por adultos falantes do português brasileiro como língua materna. As análises foram guiadas pela Fonologia Gestual, que concebe a fonologia das línguas como um sistema dinâmico. Foram analisadas as produções orais de 10 sujeitos adultos com idades entre 20 e 25 anos – 5 do gênero masculino e 5 do gênero feminino – e de uma usuária de implante coclear, de 13 anos. Os dados foram gravados por meio de um gravador digital, modelo Zoom H4N sincronizado a um ultrassom, modelo Mindray DP-6600. As coletas ocorreram em duas etapas. Na primeira, imagens foram apresentadas na tela de um computador, as quais deveriam ser nomeadas e inseridas na frase veículo Digo (palavra) bem bonito. Na segunda, o mesmo procedimento foi realizado com logatomas. A totalidade do corpus é de 3.420 itens. Os dados acústicos foram analisados por meio do software Praat, versão 5.3.77 (BOERSMA & WEENINK, 2007), e os dados articulatórios por meio do Software AAA (WRENCH, 2012). Os dados foram submetidos, ainda, à análise estatística através do software SPSS Statistics. Os resultados foram ao encontro dos obtidos por Silva (1996): Identificou-se as três fases acústico-articulatórias da consoante [?] quanto à trajetória dos formantes, bem como sua palatalização. Tais fatos acústicos demonstraram possuir correlatos com as imagens ultrassonográficas obtidas. A informante usuária de implante coclear realizou a consoante mais lentamente do que os informantes ouvintes, mas os aspectos acústicos de suas produções são semelhantes aos desses informantes. Entretanto, ao analisar os dados articulatórios, detectou-se uma variabilidade na palatalização da consoante, fato não detectado para os ouvintes. / Due to their linguistic, articulatory and acoustic characteristics, the lateral consonants are conceived as complex sounds. They promote curiosities, doubts and controversies (Ladefoged, 1986; SILVA, 1996 and ALBANO, 2001) in the academia, but it is scarcely investigated. In fact, descriptions of the palatal lateral approximant consonant, for example, are still rare in Brazilian Portuguese (BP), especially in atypical populations. This research aims to investigate the production of this consonant by people who received cochlear implant and adult speakers of Brazilian Portuguese as their mother tongue. The analysis was guided by the Gestual Phonology that conceives the phonology of the languages as a dynamic system. Ten adult subjects with ages between 20 and 25 years - 5 males and 5 females - and one female user of cochlear implant with the age 13 years had their oral production analyzed. The data were recorded by a digital recorder Zoom H4N model synchronized to an ultrasound, Mindray DP-6600 model. The samples were collected in two stages. First, images were displayed on a computer screen, whose names should be inserted in the gap of the vehicle phrase Digo (word) bem bonito – I say (…) pretty well. In the second, the same procedure was performed with logatomas. The totality of the corpus is of 3,420 items. The acoustic data were analyzed by Praat software version 3.5.77 (BOERSMA & Weenink, 2007), and for the articulatory data was used AAA Software (WRENCH, 2012). The data were also submitted to statistical analysis using SPSS software. The results agree with those obtained by Silva (1996), it means: the three acoustic-articulatory phases of the consonant [x] were identified regarding the trajectory of the formants, as well as palatalization. Such acoustic facts demonstrate correlations with the obtained ultrasound images. The informant with cochlear implant articulated the consonant more slowly than listeners, but the acoustic aspects of her productions are similar to those informants. However, when we analyze the articulatory data it was detected a variability in palatalization of the consonant, a fact that was not detected for the listeners.

Les mots étant rarement produits hors contexte phrastique, les enfants vont devoir les extraire du flux de parole pour les apprendre. Les probabilités de transition (Saffran et al., 1996) et les unités rythmiques (Nazzi et al., 2006) seraient cruciales à l'émergence de la segmentation. Les formes sonores segmentées seront par la suite mémorisées en représentations phonétiquement détaillées. Cependant, selon Nespor et al. (2003) les consonnes seraient plus impliquées au niveau lexical que les voyelles, proposant un biais consonantique dans la reconnaissance des formes sonores segmentées. Le premier axe de ma thèse s'intéresse aux capacités précoces de segmentation chez les enfants francophones nés à terme et prématurés afin de déterminer quand elles émergent et dans quelle mesure les unités rythmiques sont impliquées. Le second s'intéresse à l'émergence et à l'origine du biais consonantique. Les résultats montrent que (1) les enfants nés à terme et prématurés sont capables de segmenter la parole en utilisant l'unité syllabique dès 6 mois ; (2) d'un biais vocalique à 6 mois, les enfants francophones, acquièrent un biais consonantique à 8 mois dans la reconnaissance de formes sonores segmentées ; (3) le biais consonantique proviendrait donc de l'acquisition des propriétés acoustiques/phonétiques de la langue maternelle. / Since words are rarely produced in isolation, one of the first steps in acquiring new words is to segment them from continuous speech. Transitional probabilities (Saffran et al., 1996) and rhythmic units (Nazzi et al., 2006) have been proposed to be crucial at segmentation onset. Segmented word forms will then have to be stored as phonetically-specified representations for future recognition. However, Nespor et al. (2003) hypothesized that consonants, more than vowels, are involved at the lexical level, proposing a consonant bias in early word processing.The first part of my dissertation investigates preterm and full-term infants' segmentation abilities to determine when they emerge and to what extent rhythmic units are involved. The second part investigates the emergence and origin of the consonant bias in recognizing segmented word forms with full-term infants. Results show that (1) both preterm and full-term 6-month-olds are able to segment speech by using syllabic units; (2) French-learning infants switch from a vowel bias at 6 months to an adult-like consonant bias in recognizing segmented word forms; (3) the consonant bias emerging between 6 and 8 months of age, it would result from the processing and learning of the acoustic/phonetic properties on the language being acquired.