This thesis is a French-Czech contrastive study which deals with schwa (mute E) in the two languages. It is conducted in a pedagogical perspective. After comparing the principal phonetic/phonological characteristics of the two languages, the topic of the theoretical part is schwa in spoken French. One of the chapters concerns the treatment of mute E in French songs and its treatment in poetry. The second part consists in an experimental study ; it contains the description and the results of three experiments concerning the two languages. The acoustic realization of schwa (or a "schwa-like" vocalic segment) is observed through production tasks and perception tests of syllabicity in French and Czech words. The acoustic analyses seem to support the well-known hypotheses that it is not just the effect of the phonological cribble, but also a mental representation of the graphical form that is operating in the process of the perception of a foreign language. Furthermore, the representation of the graphical form seems to play a role for French listeners when hearing their own language. The articulatory habits of native speakers can considerably influence not only the production of a foreign language being studied, but also its perception. In the case of our first two perception tests the French listeners were...

Die vorliegende Arbeit befasst sich mit der Akzentuierung im français du Midi, einer diatopischen Varietät des Französischen, die in ihrer Entstehung nicht unerheblich vom Okzitanischen beeinflusst wurde. Trotz der kaum zu überblickenden Literatur zur Prosodie des Französischen schien eine solche Untersuchung überfällig, zumal die Forschung fast ausschließlich die Standardsprache bzw. außerhalb Frankreichs gesprochene Varietäten fokussiert. Die kontrastiv angelegte Analyse zweier PFC-(Teil-)Korpora aus vorgelesenen und spontansprachlichen Äußerungen lässt auch Rückschlüsse auf die Besonderheiten des viel diskutierten Akzentsystems des Standardfranzösischen zu, das innerhalb der Romania eine außergewöhnliche Stellung einnimmt, da es als einzige die Entwicklung zum (finalen) Phrasenakzent durchgemacht hat. Die Prosodie des français du Midi zeichnet sich gegenüber dem Standard in erster Linie durch die große Zahl der Schwa-Silben aus. Die Unterschiede in der Silbenstruktur wirken sich auf Phrasenausdehnung und Akzentuierung aus: Da die finale Schwa-Silbe von der Akzentuierung ausgeschlossen ist und das français du Midi zahlreiche Schwa-Silben im Phrasenauslaut kennt, sind Paroxytona vermehrt anzutreffen. Auch lässt sich bezüglich der Sekundärakzente ein quantitativer Unterschied ausmachen: Die größere Ausdehnung der Phrase im Midi bedingt einen erhöhten Bedarf an rhythmusbildenden Nebenakzenten zur Aufrechterhaltung regelmäßiger Fußstrukturen. Letztere tragen ebenso zu einer Perzeptionsoptimierung bei wie Gegengewichte, die in der Regel von initialen Phrasenakzenten gebildet werden. Die OT-Modellierung der Untersuchungsergebnisse zeigt, dass der auf Reynolds & Nagy (1994, 1996) zurückgehende floating constraint-Ansatz geeignet ist, die in den Korpora zu beobachtende Variation adäquat darzustellen.

Although many studies have been carried out regarding the acquisition of Spanish as a second language, very few have focused on the acquisition of Spanish vowels. Studies that have compared the L2 production of Spanish vowels in learners who have spent an extensive time living abroad versus at home learners are scarce at best. The present study hopes to add to the literature by comparing the L2 pronunciation of the Spanish /a/ in these two groups using an acoustic analysis with the aid of speech-signal processing software and the inclusion of a native group for comparison. In addition, it hopes to provide insight into how these groups vary in their pronunciation of the Spanish /a/ in different tasks. Three tasks were administered—an oral interview, the reading of a short story, and the reading of a word list—whose range varied by less formal to more formal, respectively. The tokens were analyzed using Praat to find the F1 and F2 value at the midpoint of each. The results indicate that those who lived in a Spanish-speaking country for an extensive period of time (RM) demonstrated a significant difference (p<0.05) between their production of the Spanish stressed /á/ and the unstressed /a/ in the oral interview and short story tasks, but did not show a significant difference in the more formal word list task. The at-home (AH) group, who had spent no more than three weeks in a Spanish-speaking country, displayed a significant difference (p<0.05) between the two tokens in all three tasks. It was found that the RM group displayed a significant difference (p<0.05) in F2 values between it and the native speaker (NS) group in one of the tasks, indicating that language transfer was present in both its stressed and unstressed tokens of the Spanish /a/. Interestingly, the native Spanish-speaking group also displayed a significant difference (p<0.05) between its production of the stressed /á/ and the unstressed /a/ in the short story task.

<p>Swedish learners of French often experience large difficulties in understanding spoken French. Words that the learners know very well when written or when pronounced separately are often hard to recognize in the speech flow. The aim of this study is to examine Swedish learners’ perception of French speech in order to identify the problems.</p><p>The thesis consists of two parts. The first part provides an introduction to the perception of a second language. It also describes the phonological structures of Swedish and French and gives an overview of studies of the perception of spoken French.</p><p>The second part of the thesis contains a presentation and an analysis of four perception experiments conducted with Swedish learners of French. The results show that the learners often confuse phonological contrasts that do not exist in Swedish. It is furthermore found that the phonological processes of <i>schwa deletion</i>, <i>liaison</i>, <i>enchaînement</i> and <i>voicing assimilation</i> contribute to the perception problems. However, although <i>liaison</i> may complicate word recognition the results indicate that the so-called <i>potential liaison</i> does so to an even greater extent. In a listening test using nonsense words, the learners seem actually to expect liaison when perceiving a word that can be linked to a following nonsense word. In fact, sequences like <i>un navas</i> and <i>un avas</i> are both perceived as <i>un avas</i>. Paradoxically, liaison thus seems to be most problematic when it does not occur.</p><p>As to schwa deletion, the results show that word recognition is delayed when the schwa in the first syllable is deleted, as in <i>la s’maine</i>. In addition, the learners make a large number of errors due to schwa deletion. This phonological process sometimes completely prevents word recognition, especially when combined with a voicing assimilation. Schwa deletion thus seems to strongly complicate Swedish learners’ word recognition in spoken French.</p>

Swedish learners of French often experience large difficulties in understanding spoken French. Words that the learners know very well when written or when pronounced separately are often hard to recognize in the speech flow. The aim of this study is to examine Swedish learners’ perception of French speech in order to identify the problems. The thesis consists of two parts. The first part provides an introduction to the perception of a second language. It also describes the phonological structures of Swedish and French and gives an overview of studies of the perception of spoken French. The second part of the thesis contains a presentation and an analysis of four perception experiments conducted with Swedish learners of French. The results show that the learners often confuse phonological contrasts that do not exist in Swedish. It is furthermore found that the phonological processes of schwa deletion, liaison, enchaînement and voicing assimilation contribute to the perception problems. However, although liaison may complicate word recognition the results indicate that the so-called potential liaison does so to an even greater extent. In a listening test using nonsense words, the learners seem actually to expect liaison when perceiving a word that can be linked to a following nonsense word. In fact, sequences like un navas and un avas are both perceived as un avas. Paradoxically, liaison thus seems to be most problematic when it does not occur. As to schwa deletion, the results show that word recognition is delayed when the schwa in the first syllable is deleted, as in la s’maine. In addition, the learners make a large number of errors due to schwa deletion. This phonological process sometimes completely prevents word recognition, especially when combined with a voicing assimilation. Schwa deletion thus seems to strongly complicate Swedish learners’ word recognition in spoken French.

The English diction texts by Madeleine Marshall and Kathryn LaBouff are two ofthe most acclaimed manuals on singing in this language. Differences in style between the two have separated proponents to be primarily devoted to one or the other. An in-depth study, comparing the precepts of both authors, and applying their principles, has resulted in an understanding of their common ground, as well as the need for the more comprehensive information, included by LaBouff, on singing in the dialect of American Standard, and changes in current Received Pronunciation, for British works, and Mid-Atlantic dialect, for English language works not specifically North American or British. Chapter 1 introduces Marshall and The Singer’s Manual of English Diction, and LaBouff and Singing and Communicating in English. An overview of selected works from Opera America’s resources exemplifies the need for three dialects in standardized English training. Chapter 2 reviews notational and diction resources, and use of the International Phonetic Association’s alphabet (IPA). Chapter 3 directly compares Marshall and LaBouff’s views of the importance of the unstressed syllable, often schwa [ә] or open I [ɪ], as vital to allowing the audience to understand the flow of the sung text, and contrasts their differences regarding < r >. Chapter 4 discusses observations in applying the tenets with singers, focusing on three arias coached for this dissertation. Chapter 5 states conclusions and opportunities for further research. Figures include materials from the Juilliard School Archives. Appendices include interviews.

Tableau d’honneur de la Faculté des études supérieures et postdoctorales, 2017-2018 / En français laurentien (FL), il existe un phénomène d’apparente interversion du schwa et de la consonne qui le précède. On le retrouve dans certains clitiques et en syllabe initiale de certains mots. À titre d’exemple, on peut entendre le prononcé [œl], je [œʒ] ou regarde [œʁgard]. L’absence de schwa à l’initiale absolue du mot est une généralisation classique de la distribution du schwa en français de référence (Dell 1973 ; Martinet 1960), mais celle-ci est indéniablement contredite en FL. Si le schwa initial laurentien a été remarqué par les linguistes au moins depuis La Follette (1960), il n’a encore jamais fait l’objet d’études spécifiques, notamment quantitatives. À première vue, le schwa à l’initiale absolue semble provenir d’une métathèse, qui consiste en une interversion de phonèmes, et certains linguistes l’ont déjà expliqué comme telle. Toutefois, d’autres linguistes analysent ce phénomène comme étant l’épenthèse d’un schwa prothétique. Par l’étude des productions de 48 locuteurs québécois, ce travail vise à établir la distribution du SI laurentien au regard de facteurs internes et externes pertinents ainsi qu’à en comprendre la nature phonologique. Nous verrons que le SI apparait principalement dans le pronom je, le pronom le et le préfixe re- et qu’il est sensible à la nature des contextes segmental, lexical et prosodique. Il s’avèrera aussi que le lieu d’origine des locuteurs a une influence considérable sur sa production. Pour ce qui est de la nature phonologique, il apparaitra que certains SI sont prothétiques, alors que d’autres sont lexicalisés.

Le double doublage francophone donne lieu à deux doublages différents pour un même film américain : un doublage français et un doublage québécois. Issue de la législation des deux aires géographiques, cette pratique donne l'occasion inégalable d'analyser les deux doublages produits d'un point de vue sociolinguistique. Au Québec, la langue du doublage est souvent nommée « français international », mais en vérité, il n'existe pas de source décrivant officiellement les normes prescriptives ou objectives de ce français, qui est davantage une idéologie (Sanders, 1996, p. 113) qui découle des représentations que les doubleurs ont de la langue française. Il a déjà été démontré que la langue du doublage québécois est particulière en raison du désir des doubleurs d'employer un français à la fois ancré au Québec et délocalisé, qui sont deux objectifs difficilement conciliables (Reinke et Ostiguy, 2012, Reinke et al.,2017 et 2019). En effet, il semble avoir dans le doublage québécois peu de traits de la variété québécoise, et ce, même si plusieurs de ces traits sont aujourd'hui acceptés en variété soutenue. Afin de connaître quel type de norme est suivi par les doubleurs, endogène ou exogène, l'étude du e variable est pertinente, principalement à cause de sa variation diatopique (Côté, 2012, p. 260) et diaphasique (Lucci, 1978, p. 46). Afin de nous éclairer sur les normes objectives suivies par les doubleurs, nous analyserons les usages des doubleurs quant à cette variable linguistique dans un corpus de transcriptions de 11 films et séries télévisées. La réalisation du e variable sera analysée selon la provenance géographique du doublage, la formalité de la situation de communication et le genre cinématographique. / The French language "double dubbing" phenomen on gives rise to two different dubbing tracks for the same American movie: a French dubbed version from France and one from Québec. As a result of legislation in the two geographical regions, this practice provides an unparalleled opportunity to study the two dubbings from a sociolinguistic point of view. In Québec, the language of dubbing is often called "international French", but in reality, there exists no source officially describing the prescriptive or objective norms of this particular French, which is more of an ideology (Sanders, 1996: 113) originating from the dubber's representations of the French language. It has already been shown that the language of the Quebec dubbing is peculiar, due to the dubbers' desire to use a language both rooted in Québec and delocalized, which are two objectives that are difficult to reconcile (Reinke et Ostiguy, 2012, Reinke et al., 2017 et 2019). Indeed, Québec's dubbing seems to offer few features typical of Québec's French, even if several of these traits are now accepted in formal speech. In order to elucidate what type of norm is followed by the dubbers, endogenous orexogenous, the study of variable e is relevant, mainly because of its diatopic (Côté, 2012,p.260) and diaphasic variation (Lucci, 1978, p. 46). In order to shed light on the objective norms followed by dubbers, we will study how they use the variable e in a corpus composed of the transcriptions of 11 films and television series. The usage of the variable e will be analyzed according to geographical origin of the dubbing, register and film genre.