In modern urban Singapore, the variety of English spoken evolves through a continual negotiation of adhering to traditionally standard models and creating local norms in the environment of myriad social and substratum language influences. Singapore English (SgE) speakers constantly navigate a multilingual situation which requires them to simultaneously handle the language systems of the society’s main working language, English, and at least one other language while being immersed in a linguistic environment where interactions in countless other languages and varieties take place. Variation, thus, inevitably exists within SgE as depicted in models of variation developed throughout the years. While this variation manifests itself in many forms, this study focuses particularly on the sociophonetic variation of /r/ realisations, an area of SgE in which the little research done previously provides only impressionistic or preliminary descriptions. Here, /r/ variation is studied through an auditory and acoustic investigation of both read and conversational speech data collected from male and female SgE speakers of Singapore’s major ethnic groups (i.e. Chinese, Malay, Indian, and Eurasian). Through the use of an auditory perceptual strength index and a combination of fixed- and mixed- effects statistical modelling methods, this study reveals /r/ variation in SgE on several levels. Results show that SgE speakers produce a range of /r/ variants, both within and between speakers, and also highlight the intertwined roles of language-internal factors (e.g. phonological contexts, word class) and language-external factors (e.g. speech style, ethnicity, speaker sex) in determining variation in both the realisation and distribution of /r/ in SgE. Finer auditory and acoustic distinctions are found in approximant /r/, reflecting both the phonetic complexity of /r/ and the multifaceted nature of SgE. Additionally, supportive evidence for the presence of innovative trends in SgE /r/ realisation (i.e. labiodental /r/) and of fading ones (i.e. taps/trills) is also found. Taken together, these results provide the basis for discussions of a potential situation of natural /r/-weakening and the impacts of speech styles, cross-linguistic influences, and language dominance on /r/ variation. They also postulate trends of change in /r/ realisations in SgE affected by age, ethnicity and speaker sex. Besides contributing to the general on-going discussions of synchronic variation and diachronic change in the story of /r/, this study shares insights into the intricacies of studying linguistic patterns in multilingual urban communities and provides empirical evidence for the need of a multidimensional approach in researching multicultural varieties and/or ‘New Englishes’ like SgE.

This thesis investigates the construction of identity and authenticity through sociophonetic variation, focusing on British Hip Hop artist Amy Winehouse. Prior work on British vocal artists’ phonetic variation has relied upon regional categorical frameworks (Trudgill, 1983; Carlsson, 2001) and found variation to be evidence of production errors and speakers’ misidentification of targeted speech patterns, resulting in summative interpretations of conflict between speakers’ discreet identities and speech pattern categories. More recent work has attended to linguistic processes within cultural movements influenced but not strictly delimited by sociolinguistics’ canonical categories of region, class, race, etc. Within the context of the Hip Hop cultural movement, which demands members maintain authenticity via its mantra of keepin’ it real, scholars have described processes by which authenticity is redefined and re-localized (Pennycook, 2007), emphasized the performative process of the construction of identity rather than the categorical delineation of identity (Alim, 2009), explicated the construction of authenticity within Hip Hop as inextricable from Hip Hop’s roots in the Black American Speech Community (Alim, 2006), and shown how linguistic processes mediate the markedness of artists’ Whiteness as they construct authenticity within Hip (Cutler, 2007). This work applies sociophonetic analytic tools to sung and spoken speech informed by indexical theory. Through indexical theory, the construction of identity is examined via the employment of variants that do not convey fixed meanings but instead create complex fields of possible meaning (Eckert, 2008). The variables examined include postvocalic contexts of the liquids /l/ and /r/ and intervocalic instances of /t/. Findings indicate that Winehouse’s use of non-rhotic postvocalic /r/ in spoken language, rhotic postvocalic /r/ in singing language, glottal [ʔ] intervocalic /t/ in spoken language, intervocalic /t/ as [ɾ] in singing language, and categorical use of vocalized postvocalic /l/, demonstrates a negotiation between a Hip Hop identity and a White British non-posh identity. Her spoken and singing language represent a re-localizing of Hip Hop’s demand for authenticity within Winehouse’s British context. Findings indicate that phonetic features can index a redefinition of authenticity as forms of talk, such as Hip Hop, gain ownership in new contexts.

The alternation between labial and velar fricatives (e.g. [x]uego fuego ‘fire’) and labial and velar approximants (e.g. a[ɣ]uelo abuelo ‘grandfather’) frequently co-occur in disparate Spanish dialects (Colombia, New Mexico, El Salvador, Ecuador, Chile, among others). I hypothesize that these alternations are triggered by the perceptual similarity between such variants in the context of [u] and [w]. I further hypothesize that the spread of these variables to the upper layers of society is prevented by formal education, since orthography can block sound change. Although the labio-velar alternations have been observed before, there are few experimental studies addressing their acoustic and perceptual motivations. Yet, the only way to understand the mechanisms of sound variation and change is to analyze the physical, acoustic and perceptual characteristics of the sounds involved. This dissertation uniquely combines three methodologies of data elicitation in order to achieve a better understanding of the alternations. Vernacular speech was collected through sociolinguistic interviews. Contextually controlled target words were elicited via a picture naming task. Finally, the hypothesis that the alternations were driven by the perceptual similarity between the sounds was tested via an AX discrimination test. The sociolinguistic data was correlated with the results from the perception experiment to determine whether more variation in speech correlates with higher rates of confusion in perception. The results reveal that Education and Following Context are two of the most powerful factor groups that influence the alternations. The alternation is almost exclusively found before the diphthongs [we, wi], and in stressed syllables. Knowing the orthography plays an important role in blocking the diffusion of this perceptually driven variation. The same factors affecting the variation in sociolinguistics interviews were found to be significant in increasing the confusion between [f] ~ [x] and [β] ~ [ɣ] in the perception experiment. The acoustic analysis (centre of gravity and F2 at vowel onset), however, did not support the hypothesis regarding the similarity of labial and velar fricatives and approximants.

The alternation between labial and velar fricatives (e.g. [x]uego fuego ‘fire’) and labial and velar approximants (e.g. a[ɣ]uelo abuelo ‘grandfather’) frequently co-occur in disparate Spanish dialects (Colombia, New Mexico, El Salvador, Ecuador, Chile, among others). I hypothesize that these alternations are triggered by the perceptual similarity between such variants in the context of [u] and [w]. I further hypothesize that the spread of these variables to the upper layers of society is prevented by formal education, since orthography can block sound change. Although the labio-velar alternations have been observed before, there are few experimental studies addressing their acoustic and perceptual motivations. Yet, the only way to understand the mechanisms of sound variation and change is to analyze the physical, acoustic and perceptual characteristics of the sounds involved. This dissertation uniquely combines three methodologies of data elicitation in order to achieve a better understanding of the alternations. Vernacular speech was collected through sociolinguistic interviews. Contextually controlled target words were elicited via a picture naming task. Finally, the hypothesis that the alternations were driven by the perceptual similarity between the sounds was tested via an AX discrimination test. The sociolinguistic data was correlated with the results from the perception experiment to determine whether more variation in speech correlates with higher rates of confusion in perception. The results reveal that Education and Following Context are two of the most powerful factor groups that influence the alternations. The alternation is almost exclusively found before the diphthongs [we, wi], and in stressed syllables. Knowing the orthography plays an important role in blocking the diffusion of this perceptually driven variation. The same factors affecting the variation in sociolinguistics interviews were found to be significant in increasing the confusion between [f] ~ [x] and [β] ~ [ɣ] in the perception experiment. The acoustic analysis (centre of gravity and F2 at vowel onset), however, did not support the hypothesis regarding the similarity of labial and velar fricatives and approximants.

Prior research has suggested that there may be differences in voice quality between black and white speakers of South African English who had attended well-resourced middle-class schools. The principal objective of the study is to address the question of whether there is any acoustic evidence of such differences. The study then proceeds to describe such acoustic evidence for differences in voice quality. The author interviewed 36 female South African English speakers (18 white and 18 black) between the ages of 18 and 22. The research subjects had all attended well-resourced middleclass schools. In order to control for the possibility of substrate influences on voice quality, all black participants were of an isiXhosa language background. High quality sound recordings were conducted, consisting of both a set of read sentences as well as semi-structured interviews, the latter of which formed the core dataset for the subsequent acoustic analysis. The acoustic data were analyzed using VoiceSauce, a program specifically designed for the acoustic analysis of voice quality. Measurements were based on automatically segmented speech samples using FAVE and PRAAT. The VoiceSauce measurement data were statistically analyzed by means of a linear mixed effects regression analysis and Wilcoxon rank sum tests using the statistical package R to evaluate the significance of ethnicity as a variable. The effect of ethnicity was found to be significant for several measures of spectral tilt (including for example, 2K*-5K, H4*-2K*, H1*-H2* and H1*-A1*) and cepstral peak prominence with a nearly significant effect for the subharmonics-to-harmonics ratio. Black speakers exhibited consistently higher values for most harmonic differential measures (for example, H1*-A1*) overall, while white speakers exhibited higher values for fundamental frequency, harmonics-tonoise ratio and cepstral peak prominence. The author concludes that the acoustic evidence is most consistent with the hypothesis that the white speakers overall typically use a voice quality iii characterized by greater vocal fold constriction, thickness and stiffness in comparison to the black speakers, hypothesized to use a voice quality characterized by more breathiness. By providing a description of voice quality variation, the research contributes towards a more complete account of sociolinguistic variation in South African English.

This study investigates sociophonetic stress variation in the Onondaga County area of Upstate New York. I argue that five variations of stress correlate to factors of age, education level, place of residence, frequency, and analogical change. Dinkin and Evanini (2010) have examined and discovered similar outcomes of stress variation in his work with dialectal features across the state of New York. Rather than analyze the state and its borders in their entirety, I focus on morpheme-specific analogical change of stress in specific social categories within the Syracuse, New York region. In terms of lexical items, I analyze stress placement within four-, five-, and six-syllable words containing the -mentary affix and explore how stress shifts in these words depending on those social and linguistic factors. Data were collected through formal and informal sociolinguistic interviews in which each instance of the target words were analyzed as belonging to one of five types of stress. Results indicate that Syracuse is one of the locations in the state that see all five stress patterns. To further investigate, I take the provided evidence of stress variation and filter for sociological relevance for factors of age, gender, and residence.

The change a speaker makes in response to background noise is known as the Lombard Effect (LE). This study investigated the acoustic changes that are undergone in the presence of broadband noise and two-talker babble. Of particular interest were vocal fundamental frequency (F0) and formant frequency vowel space measures across sex. Forty participants (20 male, 20 female) were recruited and asked to read phrases in quiet and in the presence of two-talker babble and broadband noise. These masker conditions were presented at 50 and 70 dB HL. The phrases were recorded and acoustically analysed. The results showed a significant sex difference for both F0 and vowel space. A masking condition effect was not displayed for either F0 or vowel space. A significant effect was however shown for F0 according to intensity level, suggesting a LE. While the sex difference in F0 values can be explained on the basis of differences in vocal anatomy, the sex difference in vowel space was indicative of a sociophonetic influence on speech production.

No description available.

A presente pesquisa de mestrado investiga como as características sociais dos ouvintes, bem como o significado social das variantes linguísticas, afetam o modo como são atribuídas diferenças acústicas a pares formados entre cinco variantes de (-r) em coda: (i) vibrante com três batidas, (ii) vibrante com duas batidas, (iii) tepe (vibrante simples), (iv) aproximante alveolar, (v) aproximante retroflexa. Em entrevistas sociolinguísticas (MENDES; OUSHIRO, 2012), informantes paulistanos comentam acerca de dois graus de retroflexão do (-r): um mais forte (que corresponderia ao retroflexo), típico de um falar caipira e um mais fraco (a aproximante alveolar) que seria falado na cidade de São Paulo. Essas duas variantes ainda se oporiam ao tepe, tipicamente paulistano. Tais discursos revelam que os falantes parecem perceber diferenças fonéticas bastante sutis e, ao mesmo tempo, associar significados sociais a elas. No intuito de avaliar quão salientes são essas sutilezas acústicas, desenvolveu-se um experimento de percepção sociofonética (CAMPBELL-KIBLER, 2006, 2007, 2009; HAY; DRAGER, 2007) no qual diferentes grupos de moradores da cidade de São Paulo deveriam atribuir diferenças a pares formados pelo mesmo item lexical mas com pronúncias de /-r/ diferentes: p.e, a palavra artéria primeiro pronunciada com o tepe e depois com a aproximante alveolar. Os resultados indicam que as respostas dos ouvintes variam de acordo com seu Sexo/Gênero, Região de Nascimento, Local de Residência na Cidade e Variante que Reconhece na Própria Fala. Isto confirma que fatores extralinguísticos de fato influenciam até mesmo a percepção fonética um fenômeno que, a priori, variaria de forma menos sistemática do que a produção linguística. / This study investigates how listeners social characteristics and the social meanings of linguistic variants impact how acoustic differences are assigned to pairs formed between five variants of post-vocalic (-r): (i) trill with three tongue hits, (ii) trill with two tongue hits, (iii) tap, (iv) alveolar approximant and (v) retroflex. In sociolinguistic interviews (MENDES; OUSHIRO, 2012), Paulistanos (people born and raised in the city of São Paulo) comment on two \"degrees of retroflexion\" of post-vocalic (-r) in Paulistano speech: one that is perceived as \"stronger\" (which corresponds to the retroflex), typical of the states countryside, and one that is perceived as \"weaker\" (the alveolar approximant). These two variants, on their turn, are opposed to the tap, the prototypical Paulistano variant. These evaluations reveal that speakers seem to perceive quite subtle phonetic differences and, at the same time, to associate social meanings to them. In order to measure how salient these acoustic subtleties are to different groups of speakers, a sociophonetic experiment was developed (CAMPBELL-KIBLER, 2006, 2007, 2009; HAY; DRAGER, 2007) in which residents of São Paulo were asked to assign differences to pairs formed by the same lexical item but with different pronunciations of /-r/. The results show that the listeners perceptions vary according to their gender, region of birth, region of residence in the city and the variants they recognize in their own speech. This reveals how extra-linguistic factors influence even the phonetic perception, a phenomenon that, a priori, would be more \"objective\" than linguistic production.

A presente pesquisa de mestrado investiga como as características sociais dos ouvintes, bem como o significado social das variantes linguísticas, afetam o modo como são atribuídas diferenças acústicas a pares formados entre cinco variantes de (-r) em coda: (i) vibrante com três batidas, (ii) vibrante com duas batidas, (iii) tepe (vibrante simples), (iv) aproximante alveolar, (v) aproximante retroflexa. Em entrevistas sociolinguísticas (MENDES; OUSHIRO, 2012), informantes paulistanos comentam acerca de dois graus de retroflexão do (-r): um mais forte (que corresponderia ao retroflexo), típico de um falar caipira e um mais fraco (a aproximante alveolar) que seria falado na cidade de São Paulo. Essas duas variantes ainda se oporiam ao tepe, tipicamente paulistano. Tais discursos revelam que os falantes parecem perceber diferenças fonéticas bastante sutis e, ao mesmo tempo, associar significados sociais a elas. No intuito de avaliar quão salientes são essas sutilezas acústicas, desenvolveu-se um experimento de percepção sociofonética (CAMPBELL-KIBLER, 2006, 2007, 2009; HAY; DRAGER, 2007) no qual diferentes grupos de moradores da cidade de São Paulo deveriam atribuir diferenças a pares formados pelo mesmo item lexical mas com pronúncias de /-r/ diferentes: p.e, a palavra artéria primeiro pronunciada com o tepe e depois com a aproximante alveolar. Os resultados indicam que as respostas dos ouvintes variam de acordo com seu Sexo/Gênero, Região de Nascimento, Local de Residência na Cidade e Variante que Reconhece na Própria Fala. Isto confirma que fatores extralinguísticos de fato influenciam até mesmo a percepção fonética um fenômeno que, a priori, variaria de forma menos sistemática do que a produção linguística. / This study investigates how listeners social characteristics and the social meanings of linguistic variants impact how acoustic differences are assigned to pairs formed between five variants of post-vocalic (-r): (i) trill with three tongue hits, (ii) trill with two tongue hits, (iii) tap, (iv) alveolar approximant and (v) retroflex. In sociolinguistic interviews (MENDES; OUSHIRO, 2012), Paulistanos (people born and raised in the city of São Paulo) comment on two \"degrees of retroflexion\" of post-vocalic (-r) in Paulistano speech: one that is perceived as \"stronger\" (which corresponds to the retroflex), typical of the states countryside, and one that is perceived as \"weaker\" (the alveolar approximant). These two variants, on their turn, are opposed to the tap, the prototypical Paulistano variant. These evaluations reveal that speakers seem to perceive quite subtle phonetic differences and, at the same time, to associate social meanings to them. In order to measure how salient these acoustic subtleties are to different groups of speakers, a sociophonetic experiment was developed (CAMPBELL-KIBLER, 2006, 2007, 2009; HAY; DRAGER, 2007) in which residents of São Paulo were asked to assign differences to pairs formed by the same lexical item but with different pronunciations of /-r/. The results show that the listeners perceptions vary according to their gender, region of birth, region of residence in the city and the variants they recognize in their own speech. This reveals how extra-linguistic factors influence even the phonetic perception, a phenomenon that, a priori, would be more \"objective\" than linguistic production.