Browsing by Author "Flynn, Darin"
Now showing 1 - 20 of 21
Results Per Page
Sort Options
Item Open Access A Systematic Investigation of Mid-Vowel Sounds in Kinshasa Lingala: A Case of Extreme Contrast Reduction(2018-09-20) Kabasele, Philothé Mwamba; Flynn, Darin; O'Brien, Mary Grantham; Pounder, Amanda; Koffi, Ettien N'da; George, Angela MayThis dissertation examines the claims in the literature that Kinshasa Lingala (KL) speakers have merged /ɛ/ and /e/, and /ɔ/ and /o/, implying that the [±ATR] contrast between [-ATR] of mid-vowels no longer exists. This claim was tested with three production tests with native and loan data, and two perception tests on the contrast between /ɛ/ and /e/, and /ɔ/ and /o/. Picture Elicitation Task and Picture Audio Contrastive Task were used to elicit the data. Generalized Estimating Equation (GEE) was used to compare the measurements values of F1 and F2 of a vowel with the F1 and F2 of another vowel in a pair to determine their statistical differences. Lobanov (1971) z-score was used to normalize the data. The results show that KL speakers have merged /ɛ/ and /e/ in their sociolect, in both production and perception, while the contrast survives in loan data, which were used as a test of recoverability for the underlying contrast between the vowels. Therefore, the front phonetic space of KL displays absolute neutralization in KL, that is, the contrast between /ɛ/ and /e/ survives underlyingly, while it is neutralized in surface representation. Moreover, the results show that /ɔ/ and /o/ still contrast both underlyingly and on surface; but the said contrast is deactivated in perception. Evidence for the underlying contrast comes from vowel split in KL, suggesting a near-merger in the back phonetic space of KL. The results of two-way GEE, specifically Simple Effect Testing (SET) which aimed to determine the effect of syllable onset type and age group broken down by vowel respectively, show that the production of /ɛ, e/ is impacted by the onset of the syllable. /ɛ/ is phonetically implemented as [e̝] after an obstruent, while it is implemented as [e̞] after sonorants or nasal +voiced stops. Likewise, /ɔ/ is implemented as [o̞] after nasal + stops. This phonetic realization suggests an exemplar cloud in the phonetic space of KL in which the underlying forms /ɛ, e/ are realized as either [e], [e̞], or [e̝] in KL, while /ɔ, o/ are implemented as either [ɔ], [o], and [o̞].Item Open Access An exemplar-based model of intonation perception of statements and questions in English, Cantonese, and Mandarin(2017) Chow, Una Yu Po; Winters, Stephen; Winters, Stephen; Flynn, Darin; Pexman, PenelopeTo better understand how humans can perceive intonation from speech that includes natural variability, this study investigated whether exemplar theory could account for native listeners’ categorization of sentence intonation in English, Cantonese, and Mandarin. In each language, twenty native listeners classified gated utterances of statements and echo questions produced by native speakers. Then a computational model simulated the classification of these utterances, using an exemplar-based process of categorization that relied on F0 only. The computational model correctly classified these sentences above chance without normalizing F0 by speaker. Compared to the human listeners, the model was similarly sensitive to the cross-linguistic differences in the cues for questions, but performed worse when these cues were (partly) excluded from the utterances. These results suggest that human listeners store whole intonation patterns in memory and use additional acoustic information, along with F0, to categorize new statements and questions, in accordance with exemplar theory principles.Item Open Access The black vernacular versus a cracker's knack for verses(McFarland Books, 2014-10-24) Flynn, DarinThis chapter focuses initially on the fact that rap is deeply rooted in black English, relating this to Eminem in the context of much information on the language of (Detroit) blacks. This linguistic excursion helps to understand how Eminem and hip hop managed to adopt each other. The second part of the chapter focuses on Eminem's craft. Much like Eminem consciously primes our subconscious brains with his lyrics, this chapter aims to make explicit a selection of what his fans already know implicitly. Specifically, by documenting some of the sociolinguistics and psycholinguistics in his music, the chapter contributes to an appreciation of Eminem, "the first poet laureate of the white working class."Item Open Access Clipping as Morphology: Evidence from Japanese(2018-01-15) Daniel, Adam; Pounder, Amanda; Pounder, Amanda; Hayashi, Naotaka; Flynn, Darin; Winters, StephenThis thesis examines the role of morphology in clipping through an investigation of the Japanese language. The research objectives of this thesis are to outline the roles of morphology, phonology, and semantics in the clipping process, to distinguish patterns and types of clipping, and to explore whether the clipping process should be described in the grammar, as well as to determine whether clipped forms are organized lexically. The objectives of this thesis are met through analyzing the relevant literature, compiling and analyzing a database of Japanese clipped forms, and conducting a survey study involving native Japanese speakers. A definition of clipping which is distinguished from other word-formation processes is proposed based on the analysis of the relevant literature. The analysis of the database reveals predictable patterns in clipped outputs which indicates the possibility of grammaticality in the clipping process. The results from the survey study provide further evidence of patternability in Japanese clipping. Based on the patterns of clipped forms observed in the database and survey study, clipping rules which account for these prototypical forms in Japanese clipping are proposed, as well as a conceptual model of the clipping process which pertains to components of the grammar. Finally, these clipping rules are adapted into a paradigm model which outlines the structures of clipping formation, lexical storage, and the processing involved in relating clipped forms with their sources and other inflected and/or derived forms.Item Open Access Enclisis in early old french(2011) Burnett, Ashley Layna; Flynn, DarinItem Open Access Explorations of Sound Symbolism and Iconicity(2019-09-17) Sidhu, David Michael; Pexman, Penny M.; Bodner, Glen E.; Curtin, Suzanne; Flynn, Darin; Reilly, Jamie PatrickSound symbolism refers to the finding that individuals have biases to associate certain language sounds (i.e., phonemes) with certain perceptual and/or semantic features (see Lockwood & Dingemanse, 2015; Sidhu & Pexman, 2018a). An example of this is the association between the phoneme /i/ (as in heed) and smallness. This is of special interest to language because it can enable iconic relationships between form and meaning: instances in which a word’s form maps onto its meaning via resemblance. For instance, the word teeny contains a vowel associated with smallness, and refers to something small. Iconicity can also exist through direct resemblance, in which a form imitates the meaning to which it refers (e.g., bang, woosh). In Chapter 2 I synthesize the existing sound symbolism literature to arrive at five potential mechanisms that could give rise to the associations between phonemes and features. I also discuss as yet unanswered questions for the field and propose ways in which future research might answer these questions. In Chapter 3 I demonstrate a novel form of sound symbolism, namely that between phonemes and personality factors. In a departure from much of the previous literature, I conduct this investigation using real first names, allowing exploration of sound symbolism in existing language. Further, by demonstrating an association between phonemes and an abstract dimension, I widen the scope of sound symbolism, and provide a novel test case for the potential mechanisms discussed in Chapter 2. In Chapter 4 I turn my attention to iconicity and its benefit to language processing. I demonstrate that iconic words are processed faster on a lexical decision task as well as a phonological lexical decision task, compared to arbitrary words. I consider how this finding might fit into an existing model of word processing. Finally, in Chapter 5, I explore the effect of iconicity on the structure of the lexicon. I demonstrate that iconic words tend to have more unique meanings, and to have a greater amount of associated sensory experience. I discuss how these findings could shed light on the emergence of iconicity in the lexicon over time. Across these diverse studies I explore non-arbitrariness in language both at the level of individual phonemes and entire words. A running theme throughout this work is a consideration of the mechanisms underlying these phenomena, as well as an exploration of their relevance to broad, existing language.Item Open Access From phonology to syntax — and back again: Hierarchical structure in Irish and Blackfoot(2017) Windsor, Joseph W.; Flynn, Darin; Ritter, Elizabeth; de Cuba, Carlos; O'Brien, Mary; McGillivray, Murray; Bennett, RyanThe interface between phonology and syntax is a tool that can be used to provide additional evidence for study in one grammatical component or the other. Through understanding how these components interact, one can use syntactic constituent structure to control for prosodic confounds in experimentation. Conversely, one can use phonological evidence to decide between competing syntactic analyses. In ideal cases, phonological and syntactic evidence can be used in tandem, allowing converging evidence to reinforce a hypothesis. In this dissertation, I undertake three case studies to highlight: i. how a knowledge of syntactic constituent structure can increase control over prosodic variables and enable more efficient phonological research; ii. how an understanding of prosodic constituent structure can be used to motivate an underlying syntactic structure at spell-out and enable analysis of morphosyntactic features and operations before spell-out; and, iii. how the use of phonological and syntactic study in tandem can help rule out competing analyses. The first case study utilizes an analysis of syntactic constituent structure to control for different levels of prosodic prominence. The analysis of prominence made possible by syntactic assumptions allows the establishment of a hypothesis into the origins of a stress-shift phenomenon in one dialect of Irish. The second case study correlates observable sound alternations to prosodic boundaries and morpho-syntactic categories in Blackfoot. The analysis of prosodic structure facilitates the formation of a hypothesis about suffixation that is suggested to be the result of syntactic agreement, rather than head-movement operations. The third case study uses the phonological and syntactic analyses from both of the preceding studies and applies those findings to analyze the prosodic and syntactic constituency of demonstratives in both Irish and Blackfoot. A hypothesis towards a common structure for nominal expressions in the two languages is suggested, despite obvious surface differences in realization. Finally, predictions based on that hypothesis are made with questions for future cross-linguistic research. Each of the case studies examined herein contribute to the over-arching goal of the dissertation: To understand how cross-component evidence can provide additional insight and research tools towards a specific problem in one grammatical component or the other.Item Open Access I Bag Your Pardon: The Albertan ae/ɛ Vowel Shift as a Window into Community Grammars(2015-12-24) Jones, Jacqueline; Winters, Stephen; Hayashi, Naotaka; Flynn, DarinThis thesis explores a vowel shift by speakers in Alberta in which [æ] is shifting before [g]. A production experiment was designed to examine the direction, extent, and sources of this change. I hypothesized that differing prompt modalities might elicit productions that could be used to support the existence of a triadic grammar, where productions are influenced by the community (auditory), the self (pictorial), and the standard (orthographic) grammars. I hypothesize a refinement of Ohala’s Active Listener hypothesis to include “super” and “inactive” listeners as other possible sources of sound change. The results show this is an in-progress merger by approximation and that changing the prompt modality altered listener productions. Auditory prompts had the greatest influence on production: Those most likely to merge the vowels [æ] and [ɛ] were more likely to mimic auditory cues. This indicates that super perceivers spread sound change to a greater degree.Item Open Access Idéologies linguistiques à l’Est de l’Ukraine à la suite du conflit russo-ukrainien de 2014: Représentations sociolinguistiques et choix de code intergroupes dans la ville de Kharkiv(2020-07-06) Pletnyova, Ganna; Amédégnato, Sénamin; Bratishenko, Elena; Zaidi, Rahat; Flynn, Darin; Le, Elisabeth; Amédégnato, SénaminThis study explores the effects of the Russian-Ukrainian conflict on the linguistic ideologies and practices of Russian-speaking Ukrainians from the city of Kharkiv, situated in the east of Ukraine, 30 kilometres from the border with Russia. Ukraine is a country with a single official language, Ukrainian, where as many as 40% of the eastern and southern population speak Russian as their first language. As Russia capitalized on the long-existing language issues at the start of the conflict in 2014, it is important to research how it influenced Russian-speaking Ukrainians. This conflict both polarized Ukrainian Russophones into competing ideological positions for or against Russia and gave fresh vigour to the linguistic debate in Ukraine. The Russian-speaking Kharkivites’ opposing political convictions and interpretations of the role of language in a nation-state affected their linguistic behaviour, motivating them to either attempt to switch to Ukrainian, to advocate bi- or multilingualism or to demonstratively use only Russian. I sought participants for my study based on these three language ideologies, determined by previous research and confirmed by my own observations. In my project, I used qualitative research methods, i.e. semi-structured interviews, which allowed me to obtain in-depth responses from participants. The field study was carried out in Kharkiv in the summer of 2018, where I interviewed 32 adult participants (an equal number of men and women) of different ages, levels of education and professions. The study results provide evidence of changes in the participants’ attitudes to Russian, Ukrainian, foreign languages and state language policy as well as shifts in their language practices and accommodation strategies of speakers of different language ideologies. As the study was carried out four years after the beginning of the conflict, its results also cast light on the evolution of language ideologies and behaviours over time.Item Open Access Japanese Prosody and Its Role in Comprehensibility: Exploration and Training(2024-04-25) Shimada, Masako; O'Brien, Mary; Bratishenko, Elena; Cai, Wei; George, Angela; Flynn, Darin; Lee, AndrewProsody has gained attention in L2 pronunciation research due to findings indicating that prosodic features, such as rhythm, stress, duration, and changes in pitch, have a great influence on listeners’ understanding. In Study 1, I investigated how Japanese prosody affects intelligibility and comprehensibility, with the goal of advancing our understanding of L2 Japanese prosody, as it has seen limited progress thus far. In the study, deliberate speech errors involving vowel length, consonant length, and pitch accent were produced by Japanese and English learners of Japanese. Words with and without these prosody errors were then evaluated by Japanese listeners. Results show that intelligibility is significantly affected by a combination of segmental length and pitch accent errors, while comprehensibility is almost equally impacted by each prosodic feature, whether combined or independent. Furthermore, while there is no significant difference in intelligibility of speech produced by Japanese and English L1 speakers, there is a significant difference for comprehensibility, suggesting that processing L2 speech requires more effort. In Study 2, I designed prosody training for English learners of Japanese, with a focus on vowel length and pitch accent. The training incorporated two methods—embodied and computer-assisted techniques—which were compared to determine their effectiveness in training these features. The accuracy of L2 perception and production of the target features was examined at pretest, posttest, and delayed posttest, revealing a significant overall improvement regardless of the methods used. This highlights the positive effects of both methods, yielding results within a short timeframe. These findings shed light on the crucial role of prosody in Japanese in improving L2 intelligibility and comprehensibility, which can be achieved by focused instruction using embodied and computer-assisted techniques. These techniques can be readily used in language classrooms, empowering teachers to develop effective pronunciation instruction.Item Open Access Lexical Blending in Ukrainian: System or Sport?(2017) Winters, Svitlana; Pounder, Amanda; Flynn, Darin; Bratishenko, Elena; Amadegnato, Ozouf; Ronneberger-Sibold, ElkeThis dissertation examines the formal, semantic, and syntactic mechanisms involved in the formation and processing of blends in Ukrainian. The objectives pursued in this research are (i) to provide a comprehensive definition of blending, (ii) to determine whether there are any formal, semantic, and syntactic regularities in the formation of blends in Ukrainian, and (iii) to learn what factors facilitate recognition of source words, as well as interpretation of meanings of Ukrainian blends. The objectives are achieved by critically analyzing the relevant literature, conducting an analysis of a corpus of Ukrainian blends, and running three psycholinguistic experiments with native speakers of Ukrainian. Based on the analysis, a comprehensive definition of blending is proposed, which establishes the boundaries between blends and other complex words that resemble them (for example, clipped compounds). The analysis of the corpus of Ukrainian blends compiled in the course of this research reveals a number of regularities in blending. Based on these regularities, a system of formal, semantic and categorial rules which account for the formation of blends in Ukrainian are proposed. Finally, the results of the psycholinguistic experiments reveal that the following factors affect the likelihood of recognition of the source words and accurate understanding of the meaning of blends: the structure of the blend, the modality of presentation of the blend, and the amount of time allotted for its processing.Item Open Access Melodies of Hawaiʻi: The Relationship Between Hawaiʻi Creole English and ʻŌlelo Hawaiʻi Prosody(2013-10-02) Murphy, Kelly; Flynn, DarinThis dissertation offers a comparative study of speech melodies of Hawaiian and Hawaiʻi Creole English (HCE), two languages spoken in the Hawaiian Islands. HCE, which is an English lexified creole, developed from many languages such as Chinese, Portuguese, and Hawaiian. This research proposes that Hawaiian intonation patterns remained even after relexification from a Hawaiian based pidgin to an English lexified creole. This comparative study of speech melodies of Hawaiian and HCE provides documentation of falling yes/no question intonation found in both languages, as well as for other utterance types, i.e. statements, wh-questions, and listing/continuation. Patterns emerged from the major utterance types revealing insights into creole genesis from substrate influence, and language universals and variation. Of the utterance types examined, the most striking example is that of falling yes/no question intonation, present in both languages. This type of falling intonation is quite rare amongst the world’s languages, which strengthens the hypothesis I claim, that HCE received intonation patterns from Hawaiian. These findings also have implications for creole genesis as well as for language typology research for question intonation as well as universals and sociolinguistics. With regard to creole genesis, the evidence provided supporting Hawaiian as a contributing language strengthens substrate arguments and weakens a universalist view, specifically that of the Language Bioprogram Hypothesis (LBH). Also, the results show a typology of question intonation that makes use of a more latitudinal categorical distinction rather than the commonly used right-edge question intonation, which changes the perspective on language universals and variation, as well as biological codes. This dissertation also addresses language identity and how these two languages interact in the same linguistic landscape in Hawaiʻi . While providing a much-needed inventory of intonation for all utterance types in Hawaiian and HCE, this research project also addresses broader questions and linguistic topics such as creole genesis, language typology, universals, and sociolinguistics.Item Open Access Number in Saudi Arabic: Acquisition and Child-Directed Speech(2016) Alqahtani, Salma Musleh; Carroll, Susanne E.; Meisel, Jürgen; Flynn, Darin; Benmamoun, Abbas; Asatryan, MusheghThis thesis treats the acquisition of the Arabic number system, a complex system consisting of Singular, Dual, and several Plural forms. The research consists of two parts: (i) a study of Saudi Arabic parental speech directed to children aged 2 to 4 years old; (ii) a separate study of 2-to-6-year-old children’s comprehension and production of Arabic number-marking. The main findings of part (i) are that the parents use the Singular form more extensively than the other forms. They avoid using Dual and Masculine Sound Plural forms in speech to young children. Parents also produce ungrammatical speech. These results contradict claims that parents’ speech to children is always well-formed. The data for part (ii) was collected via a comprehension study (picture-pointing task) and a production study (picture-based elicitation task). These data were analysed quantitatively and qualitatively. The main results of the children’s data are: the Singular form is mastered before the Dual and Plural forms; among the non-singular forms, the Feminine Sound Plural is understood and produced before the other forms. The Dual form is the most difficult category for the children to acquire. Saudi children, like children learning other dialects of Arabic, master the Saudi Arabic number system late in comparison to the early emergence of number in other languages. In addition, the acquisition of the number system is slow and incremental. Analysis of parental speech suggests that input helps to explain these patterns. Importantly, the error analysis of production data shows that the children creatively used non-target expressions to express number meanings. Learning Arabic number involves pattern-detection and rule learning, not merely rote memorization of forms.Item Open Access Perception of Lexical Pitch Accents by Serbian and French Speakers(2023-04-27) Nikolic, Dusan; Winters, Stephen; O'Brien, Mary; Flynn, Darin; Athanasopoulou, Angeliki; Clopper, CynthiaThe study explored the perception of the Serbian lexical pitch accent by Serbian (native) and French (non-native) listeners. The motivation for the study came from the previous research on the processing of word-level prosodic categories in perception. With this study, I attempted to shed light on the ways the lexical pitch accent is perceptually processed by both native and non-native listeners in view of contributing to methodological and theoretical paradigms for studying the perception of word-level prosodic categories. The study included three experiments, all of which were carried out online. The first experiment included an AX discrimination task and an identification task in which only Serbian listeners participated. The results of the AX discrimination task revealed that the listeners were more sensitive to length than to pitch movement as the two main perceptual correlates of lexical pitch accent. The listeners were less accurate on the short-falling and short-rising contrasts, which suggested that the listeners were conflating the pitch movement on short lexical pitch accents. On the identification task, the lexical pitch accent types could not be reliably identified, which implied the lack of metalinguistic knowledge of lexical pitch accents by Serbian listeners. The second experiment was a Sequence Recall Task which investigated Serbian and French listeners' contrast and recall of lexical pitch accent types. The speakers of a language with no word-level prosodic category in their first language were expected to be perceptually insensitive to a non-native word-level prosodic category. The French and Serbian listeners were not significantly different from each other, which suggested that there was no perceptual insensitivity to lexical pitch accents by French (non-native) listeners. The third experiment was an online version of a free classification task, which was used to explore Serbian and French listeners' perception of Serbian lexical pitch accent types based on their similarity. The results were submitted to multi-dimensional scaling and clustering analyses. The analyses showed that the perceptual space between the short-falling and short-rising lexical pitch accents by the Serbian listeners was extremely reduced, while the French listeners displayed the confusion of the long-falling lexical pitch accents with the rest of lexical pitch accent types. Overall, the findings indicated that the listeners of a language with no word-level prosodic category were not insensitive to lexical pitch accents. In addition, native listeners processed lexical pitch accent through the filter of the prosodic structure of their first language, Non-native listeners were able to hear distinctions between lexical pitch accents which native listeners could not. Moreover, non-native listeners heard the contrasts that did not exist in their native language. Lastly, the results indicated both the phonetic and phonological encoding of lexical pitch accent in perception.Item Open Access Politiques éducatives en Tanzanie au prisme de l’approche communicative(2020-09-18) Mulinda, Alfred Fulgence; Amédégnato, Sénamin; Gbanou, Sélom Komlan; Flynn, Darin; Avolonto, Aimé; Vijayan, DevikaLa présente étude se propose pour objectif d’examiner l’efficacité des politiques éducatives en Tanzanie à travers l’application de l’approche communicative, conformément aux réformes curriculaires de 2005 dans l’enseignement secondaire de premier cycle. Tout particulièrement, l’étude a voulu déterminer si l’enseignement du français et de l’anglais dans le secondaire du premier cycle intégrait ou non l’approche communicative. Pour ce faire, nous nous sommes appuyé sur des approches de recherche qualitatives: analyse des documents pertinents à l’enseignement du français et de l’anglais du lendemain de 2005 ; observation des cours de français et d’anglais ainsi que des entretiens auprès des enseignants de français et d’anglais. L’intérêt était de découvrir en quoi les pratiques pédagogiques en classe de langue reflètent les principes de l’approche communicative. L’étude a révélé qu’il existe très peu de pratiques que l’on pourrait associer à l’approche communicative dans l’enseignement d’anglais et de français en Tanzanie. L’approche communicative se trouve assez intégrée dans les documents d’appui de l’enseignement mais les pratiques pédagogiques des enseignants ne reflètent que peu de pertinent pour la philosophie communicative de l’enseignement : approches d’enseignement, matériel pédagogique ainsi que leur savoir-faire théorique et politique demeurent non-communicatifs. De plus, les examens sommatifs ne s’alignent pas sur l’enseignement communicatif car la plupart des questions sont préparées de manière non-communicative. L’étude fait appel aux organes chargés de l’enseignement en question pour considérer l’implication des enseignants dans la planification de décisions pertinentes à l’enseignement, ce qui est utile pour une mise en œuvre efficace des réformes curriculaires. De plus, le gouvernement devrait s’assurer qu’il y a suffisamment de ressources d’enseignement avant de se lancer dans le processus de mise en œuvre des politiques éducatives. En outre, nous proposons que le gouvernement offre des formations continues fréquentes pour les enseignants en vue d’équiper ces derniers avec les outils théoriques comme pratiques d’enseignement. Avec la bonne formation, les enseignants seront en mesure de bénéficier d’une pléthore de ressources dans leur environnement pour proposer des cours communicatifs.Item Open Access The Nominal Inflectional Suffix -e(t) in Damascus Arabic: A Network Morphology Account(2017) Allihibi, Sameera Atiya; Flynn, Darin; Pounder, Amanda; Sedivy, JulieThis thesis is concerned with the nominal (noun and adjective) inflections in Damascus Arabic (DA). In particular, the focus is on the so-called feminine suffix -at ~ -et. Several phenomena associated with this suffix are examined: a) allomorphy, b) syncretism, and c) optional agreement. The thesis offers a default-based account within the framework of Network Morphology. It demonstrates that different notions of default (e.g. default inheritance, overriding default, global inheritance, and exceptional case) provide a unified and elegant account for the multiple phenomena under study. The vocalic alternation between -[a]t and -[e]t represents a case of the dissociation between morphology and phonology, where a single morpheme corresponds to two phonological realizations. I propose a declarative morphological account as opposed to the traditional procedural phonological one that relies on the notion of the underlying form. I argue that -[e]t is the default allomorph, contrary to the widely assumed belief that -[a]t is the underlying vowel which raises to -[e]t. The notion of “default inheritance” allows us to treat the shared inflection among nominals (noun and adjective) as a main generalization that is positioned at the top of the inheritance hierarchy. This generalization is inherited by all classes of nominal, thus, -[e]t is represented as the “elsewhere” allomorph. The limited distribution of -[a]t to nominals with stem-final pharyngeals is treated as an “overriding default” specific to this class of nominals that overrides the main default. The syncretism between plural and -et agreement with the plural of the human class manifests a dissociation between morphology and syntax, whereby morphology fails to make a distinction that is required by syntax. For this language-specific syncretism, rules of referral are necessary to predict that the plural and feminine singular share a cell in the adjectival inflectional paradigm. With the benefit of the notion of “global inheritance”, this sharing of the cell is generalizable to all agreeing categories (i.e. verbs, demonstratives, pronouns, and possessive markers) that have the same pattern of syncretism. The global inheritance principle is superior in its simplicity and predictability power, by which a simple rule of referral makes it possible for a single cell (i.e. plural) in one paradigm (i.e. adjectives) to predict other cells (i.e. plurals) in multiple paradigms (i.e. verbs, demonstratives, etc.). Finally, the alternation between the plural and -et agreement with human nouns is handled by the notion of the “exceptional case default”. The regular behaviour of plurals in human nouns is to have strict plural agreement. However, for non-concatenative plural nouns, alternations occur. The irregularity of the non-concatenative plural is still manageable and predictable if we assume that the -et agreement is an “exceptional” default that applies only when plural nouns have non-concatenative morphology. Then, the alternation can be explained based on whether speakers resort to the “normal” default (i.e. plural agreement) which is positioned at the top of the hierarchy or to the “exceptional” default (i.e. -et agreement) as an overriding default positioned lower at the level of the non-concatenative subclass of human nouns.Item Open Access Tumʔi: A Phonetic & Phonological Analysis of a Khoisan Variety(2020-01) Kilian, Kelly; Bennett, Wm G.; Flynn, Darin; Ritter, Elizabeth; Bennett, Wm G.; Winters, Stephen J.As part of a linguistic research team I recorded a Khoisan language currently spoken by a linguistic community of three in the northern cape of South Africa. As the location in which this language was discovered is situated geographically close to varieties of both the Khoekhoe and Tuu language families, the question of genetic affiliation and typological similarity within the Khoisan lineages becomes significant. This will be addressed through the analysis of phonetic, phonological and lexical similarities and oppositions between Tumʔi and representative Tuu and Khoekhoe languages (Beach 1938; Bleek 1930; Ladefoged & Traill 1994; Miller et al. 2007). Overall this project attempts to answer the question, how unique is this undocumented language Tumʔi in comparison to varieties of geographically neighboring Khoisan languages? The analysis is comprised of a detailed description of the vowel and consonant systems, as well as evidence of any contrastive phonetic and phonological features. The clear focus on the analysis of sound contrasts is a consequence of limited data due to speaker competence (Grinevald, 2007). As a result of incomplete acquisition and generational linguistic attrition, the recorded utterances constitute Khoisan content words produced within an Afrikaans framework (Killian 2009). Specific research questions include: • What is the sound inventory of this language? • Are there phonation or glottalization contrasts between vowels? • What click types and accompaniments make up the inventory? • Are there laryngeal contrasts between consonants? Results of the analyses indicate the following; Tumʔi shows traces of a phonation contrast, uvular click accompaniment, and evidence of laryngeally marked stops. The phonological typology shares more similarities with the Southern Khoisan varieties of the Tuu family than with varieties of the central Khoekhoe family. Direct implications for this project include contribution to the current areal typological isoglosses separating the varieties of Khoisan located in southern Africa (Güldemann 2006). The final contribution of this work is the documentation of a moribund Khoisan dialect which has undergone no prior linguistic or anthropological investigation.Item Open Access Vocalic dorsality in Articulator Theory(2012-01-01) Flynn, DarinOn the strength of Halle’s (1995) demonstration that assimilation processes spread only terminal elements, Halle et al. (2000) propose to revise the Articulator Theory of feature geometry (Sagey 1986 et seq.) by treating designated articulators not as nodes but as features (viz. [coronal], [labial], [dorsal]) since indeed they participate in assimilation. Halle (2005) goes on to conclude that all features are terminal (since all may spread) and that we should dispense with subsegmental nodes altogether. Aside from these revisions, Halle (like Padgett and Yip) retains the feature-geometric conception of (local) assimilation as spreading. Revised Articulator Theory also preserves a long tradition in taking the tongue body to be the designated articulator of vowels (Halle 2003; Levi 2008), a view that is anathematic to the now dominant school of feature geometry known as Vowel-Place Theory (Clements 1989 et seq.). Thus current Articulator Theory uniquely generates the strong hypothesis that vowels are specified for an articulator feature [dorsal] which can spread individually because it is terminal in the segment. On the further (also traditional) assumption that phonotactics intensify within syllable rhymes, a more precise prediction can be made: [dorsal] may spread from any vowel to any adjacent segment, and coda consonants are favored targets. This specific prediction of current Articulator Theory is borne out in a variety of velarization patterns across languages. Revising and extending an original proposal by Paradis and Prunet (1993), I show that vowels can spread [dorsal] to a following tautosyllabic consonant, be it nasal, obstruent, or liquid. I also argue that extant alternative analyses fail. My conclusion is that the frequency and range of velarization effects argue strongly in favor of the Articulator Theory view of vowels (Halle 2003; Levi 2008): they are [dorsal], and this is a terminal feature.Item Open Access Voice Onset Time in Plains Cree(2021-09-10) Hodgson, Merion Rose; Flynn, Darin; Athanasopoulou, Angeliki; Winters, Stephen; George, AngelaThis thesis represents the first in-depth acoustic study of voice onset time (VOT) in Plains Cree. The study is designed to gain a better understanding of where Plains Cree fits in the typology of single series stops, how these VOTs support empirical descriptions that Plains Cree stop consonants are voiceless and unaspirated (Wolfart, 1996), and examine the factors that influence Plains Cree VOT at each place of articulation. The data was collected from publicly available recordings, measuring VOT in milliseconds from the rising edge of the release burst until the onset of voicing in the waveform. Statistical analysis included ANOVA or Kruskal-Wallis tests followed by applicable post-hoc tests. The factors include word position, syllable stress, and vowel environments. VOT of Plains Cree stops behave as predicted for single series stops but with patterns to explore for the typology. Results indicate that in Plains Cree, /p/ (mean=11.6ms, sd=4.62ms) and /t/ (mean=15ms, sd=4.9ms), /k/ (mean=25.6ms, sd=9.59ms) are steadfast short-lag stops in (most) environments. VOT was found to be statistically significantly longer when in word-initial position than in other positions and in front of high vowels in comparison to low vowels. Establishing VOT in Plains Cree not only adds to the knowledge of Plains Cree acoustics, it enables future exploration of language attitudes and acquisition, and also sets up other acoustic studies of Wolfart’s descriptions. Contextualizing Plains Cree among other single series languages and VOT typology also brings up the question of what VOT is utilized for in single series languages.Item Open Access VOT and F0 in the production and perception of Swahili obstruents: From the island to the coast to the inland region(2021-06) Alsamaani, Mohammad; Flynn, Darin; Winters, Stephen; Pounder, Amanda; Athanasopoulou, AngelikiThe status of aspiration in Swahili has received conflicting historical and linguistic accounts. To date, it is not fully understood if this laryngeal setting in the language’s four voiceless obstruents (/p/, /t/, /k/, and /tʃ/) is phonemic or allophonic. This dissertation analyzes the phonetic laryngeal variations in four Swahili varieties, which are spoken in East Africa as a first language (Zanzibar in Tanzania and Mombasa in Kenya) or as a second language (Iringa in Tanzania and Nairobi in Kenya). Two experiments, one in production and one in perception, examined the acoustic cues of voice-onset time (VOT) and fundamental frequency (F0) to investigate how speakers employ these language-specific details. A total of 98 participants (male and female) took part in these experiments. This dissertation first explores statistically the productions of real words by all subjects. Linear mixed-effects models indicate that the phonetic cue that accounts for more variance in the data is VOT, with little significance found for F0. However, the VOT cue was not a significant dependent measure for three of the locations. Speakers from Zanzibar, both men and women, demonstrated that their dialect is most distinct from other dialects in that a) the language, as they speak it, has an aspiration contrast in minimal and non-minimal pairs; b) VOT measurements are different depending on word origin (long for English loanwords, intermediate for native Swahili words, and short for Arabic loanwords); and c) while females had significantly longer VOT durations in the other towns, the productions of both genders in Zanzibar were not statistically different. Next this dissertation analyzes statistically how the same subjects perceived and imitated modified VOTs and F0s (three levels each) of non-words. Imitation was found for VOT only among L1 speakers of the language, and a closer look revealed that the Zanzibar participants’ imitations were the strongest: that is, they mimicked all three levels with statistically significant accuracy. Mombasa participants, on the other hand, distinguished only between Level 1 and the other two. By contrast, L2 speakers of Swahili from Nairobi and Iringa showed no difference across the three levels in either VOT or F0. In short, only Zanzibar preserves produced VOT-based aspiration. The loss of aspiration along the coast and in the inland regions, which may be due to the linguistic influence of neighboring Bantu languages, is reflected by the abandonment of unstable orthographic systems that distinguished aspirated consonants in favour of the simplified orthography of later times.