The delicate anatomical structures involved in infant cry production require intricate neurophysiological control especially in premature infants or those with a reduced respiratory or laryngeal function. Certain features like phonatory noise or subharmonics can be observed in infant cries using spectrograms. These features have a certain indicative valence for characterising the maturation stage of vocal control or its performance. One possible cause of deviation in neurophysiological coordination during voice production is disturbed CNS mechanisms, finally the consequences of orofacial clefts. Another is the influence of a familiar disposition for speech development disorders. The present paper studied the latter two relationships. For the evaluation and interpretation of a noise index (= average value of the noise portion within a cry) in infant’s pre-speech utterances, we analysed 1423 voice-signals emitted during the first 15 weeks of life by 10 orofacial cleft infants (5 females and 5 males), comparing these with a control group. The control group B of healthy infants was subdivided into B1 (FH- infants with a negative family history of speech developmental disorders) and B2 (FH+ infants with a positive family history of speech developmental disorders). Infants born with orofacial clefts are substantially exposed to severe difficulties for speech and language acquisition. Coupled with a premature muscle network, cleft infants are deprived in various ways (vocal nasality, limited consonant repertetoire, backward articulation etc) and their coordination of respiration, phonation and articulation is limited from a very early age. From birth until about 2 months of age, an infant's cry is characterised by a tuning phase between respiration and phonation. After training the production of more complex cry melodies with different rhythms, infants begin at 3 - 4 months of age (Wermke et al., 2005) to tune their phonation and articulation. Successfully absolving these stages of development is presumably a prerequisite for later acquisition of inconspicuous speech and language competence. The development of articulation is based on the tuning of melodies produced in the larynx and resonant frequencies from the vocal tract (Kempf, 2008). For an objective evaluation of pre-speech development in healthy and sick infants, this study produced comparable data on the appearance of selected parameters in age-appropriate control groups. In order to examine the connection between these selected cry properties and the physiological condition in infants, we made comparisons to 2623 voice-signals from 10 FH+ infants and 3002 voice-signals from 10 FH- infants (all without orofacial clefts and age-appropriates). For interpretations of future results, we also analysed 2684 voice-signals from 4 infants in the control group B1 (FH-) taken at closer time intervals until the 20th week of life. This study showed that the appearance of noise-like elements (NI) in the vocalizations of orofacial cleft infants and FH+ infants were identical during the first 15 weeks of life. Also, we could show that in both these groups (A and B2) there was a delayed development in the average signal length (phonation time). Although cleft infants and FH+ infants differ from each other physiologically, our results may propose a common neurophysiological retardation. Comparing prosodic elements in cries from FH+ and FH- infants showed differences (Blohm, 2007; Denner, 2007). Therefore, future research could apply this knowledge to a larger sample of infants in order to establish a better therapy concept, thus preventing late interventions. Infants from our control group B1 (FH-) met our expectations because when they got older, a development in their pre-speech capability was noticed. Our results support the hypothesis that in cry research, physiological differences (orofacial clefts or a family history for speech development disorders) in infants may encourage the appearance of noise-like elements in their vocalisations. However we believe that a period of training enables the infants to reduce their mean NI. The production of more complex melodies with age was better managed by the FH- infants and they also produced longer cries. To avoid a developmental retardation in speech and learning capabilities, it may be necessary in future to make more compact studies considering many other parameters and making comparisons with age-appropriates. Further studies also have to correlate these findings while investigating the consequences of these maturation processes on sound production. Despite physiological differences in the three groups of infants, the noise index (NI) as applied in this study can be used as an objective parameter for daily clinical diagnosis during the first four months of life. / Säuglinge mit einer orofazialen Spaltbildung sind bezüglich vieler Aspekte in ihrer sprachlichen Entwicklung benachteiligt (hypernasale Resonanz, eingeschränktes Konsonantenrepertoire, Rückverlagerung der Artikulation usw.) sowie bei der Koordination von Respiration, Phonation und Artikulation vor der chirurgische Behandlung. Ab Geburt bis zum Alter von 2 Monaten sind Säuglingslaute durch ausgeprägte Frequenzmodulationen charakterisiert, die auf einer Abstimmung von Respiration und Phonation beruhen. Nachdem Säuglinge in den ersten Wochen die Produktion komplexer Schrei-Melodien mit unterschiedlichen Rhythmen geübt haben, beginnen sie im Alter zwischen 3 - 4 Monaten die Phonation und die Artikulation fein abzustimmen (Wermke et al., 2005). Das erfolgreiche Absolvieren dieser Entwicklungsstadien ist möglicherweise eine Voraussetzung für einen späteren unauffälligen Sprech- und Spracherwerb. Die Artikulation beginnt sowohl mit einen intentionalen Tuning der Melodien, die laryngeal erzeugt werden, als auch mit der Bildung von Resonanzfrequenzen des Vokaltraktes (Kempf, 2008). Diese Entwicklungsprozesse sind bei Säuglingen mit orofazialen Spaltbildungen gestört. Die erhöhte nasale Impedanz führt durch einen rückgekoppelten Regelkreis zu einem Anstieg des subglottischen Druckes (Hauschildt, 2006). Die winzigen Stimmlippen der jungen Säuglinge sind diesem Druck nicht ausreichend gewachsen, so dass es regelmäßig zu phonatorischen Rauschphänomenen in deren erzeugten Lauten kommt. Dies verhindert das "Trainieren" melodisch-rhythmischer Elemente als Vorstufe für die spätere muttersprachliche Prosodie. Für eine objektive Auswertung solcher Phänomene in den vorsprachigen Entwicklungen wurden in dieser Studie das Auftreten und der Grad ihrer Ausprägung anhand ausgesuchter Parameter bei Säuglingen mit orofazialen Spalten (Gruppe A)) und bei solchen in einer altersähnlichen Kontrollgruppe (B) untersucht. Die Kontrollgruppe B bestand aus gesunden Säuglingen ohne (FH- oder B1) bzw. mit (FH+ oder B2) einem familiären Risiko für eine spezifische Spracherwerbsstörung. Letztere wurden einbezogen, da Säuglinge mit orofazialen Spalten (A) und Säuglinge mit einer familiären Disposition für Spracherwerbsstörungen (FH+) ähnlichen Abweichungen in der neurophysiologische Koordination der Lautproduktion aufweisen (Denner, 2007). In der Kontrollgruppe wurden 2623 einzelne Laute von 10 FH+ Kindern und 3002 Laute von 10 FH- Kindern miteinander verglichen. Um Entwicklungstrends besser beurteilen zu können, wurden in dichteren Intervallen (wöchentlich statt monatlich) 2686 Laute aus der Gruppe B1 (FH-) bis zur 20 Lebenswoche analysiert. Diese Studie konnte zeigen, dass während der ersten 15 Lebenswochen sehr identische, geräuschähnliche Phänomene (RI) in der Vokalisation von Spaltkindern und FH+ Kindern auftreten. Außerdem konnte gezeigt werden, dass in diesen beiden Gruppen eine verzögerte Entwicklung der durchschnittlichen Signallänge (Phonationszeit) einzelner Laute auftrat. Trotz allen physiologischen Unterschieden zwischen den Spaltkindern und den FH+ Kindern deuten unsere Ergebnisse auf eine gemeinsame neurophysiologische Entwicklungsverzögerung hin. Ein Vergleich prosodischer Elemente in der Vokalisationen von älteren FH+ und FH- Kindern ergab auch Unterschiede (Blohm, 2007; Denner, 2007). Die vorliegende Studie bestätigt dies und die Erkenntnisse könnten zukünftig verwendet werden um geeignete Therapieverfahren zu entwickeln. Säuglinge aus der Kontrollegruppe B1 (FH-) bestätigen die gestellten Hypothesen, da mit zunehmendem Alter eine signifikante Reduktion von Rauschphänomenen beobachtet wurde. Keine andere Studie hat bis heute eine vergleichbare objektive und detaillierte Analyse der Rauschelemente in Säuglingslauten durchgeführt. Die Ergebnisse bestätigen die Hypothese, dass neurophysiologische Besonderheiten, die auf eine orofaziale Spalte oder eine familiäre Disposition für eine Spracherwerbsstörung zurückzufuhren sind, das Auftreten von rauschähnlichen Elementen in der Vokalisation von Säuglingen beeinflussen könnten. Ganz offenbar gehört eine 'Trainingphase' dazu, um den mittlere RI zu reduzieren. Komplexere Melodien konnten mit zunehmendem Alter von den FH- Kindern besser beherrscht werden, da sie wohl auch längere Schreie koordinierter erzeugten. Die Arbeit bestätigt auch, dass der Rauschindex (RI), wie er in dieser Studie angewandt wurde, geeignet ist, um bei der klinischen Betreuung von orofazialen Spaltkindern, insbesondere während der ersten vier Lebensmonate, als Entwicklungsstandindikator dienen kann. Um einem Entwicklungsrückstand in der Spracherwerbsfähigkeit gefährdeter Säuglinge zu belegen, wird es in Zukunft nötig sein, umfassendere Studien durchzuführen, wobei weitere akustische Parameter berücksichtigt werden sollten.

Orofacial clefts (OFCs) are among the most common congenital birth defects and are characterized by incomplete development of the lip or the palate or both. The lip and palate develop separately at different times during the first trimester of pregnancy. The etiology of OFCs is multifactorial and includes a combination of genetic and environmental factors. This project aims to examine role of maternal diabetes mellitus in orofacial clefts through studies of medical histories, biomarkers, and genes. In a study of Utah birth certificates, mothers with pre-existing diabetes and gestational diabetes mellitus (GDM) had an increased risk of OFCs, and obese mothers also had an increased risk. Mothers of children with OFCs were more likely than mothers of unaffected children to develop obesity, metabolic syndrome and gestational diabetes mellitus later in life. These result were more strongly related to cleft palate than cleft lip. Many genes related to GDM were associated with OFCs through genetic effects alone and gene-environment interaction effects with periconceptional maternal multivitamin use, maternal smoking, and environmental tobacco smoke. These results support the hypothesis that GDM may be causally related to OFCs via multiple GDM susceptibility genes and interactions with environmental factors. Individuals with OFCs face both physical and mental health problem, which require multi-specialty team care. OFC prevention and prediction are important to public health. This dissertation reported that maternal diabetes mellitus, maternal pre-pregnancy weight and genes related to GDM had an association with the risk of OFCs. Mothers having an OFC child had an increased risk of developing metabolic abnormalities later in life. Potential risk factors were reported in this dissertation that may be useful for OFC prevention. This dissertation also reported potential biomarkers for predicting OFCs. Moreover, mothers having an OFC child may require regular monitoring of metabolic abnormalities later in life.

No description available.

As fissuras orais integram dois grupos segundo origens etiológicas distintas, quais sejam, fissuras labiais ou lábio-palatais e fissuras palatais. Realizou-se um estudo tipo caso-controle, com 450 casos e 450 controles, sendo que, entre os casos, 354 eram portadores de fissura labial ou lábio-palatina e 96 de fissura palatina. A fim de se estudar a associação entre fissuras orais e possíveis fatores de risco, foram objeto de análise as variáveis: local de moradia da mãe nos quatro primeiros meses de gestação (urbana/rural), poluição, aplicação de pesticida e herbicida na lavoura, doenças nos pais, doenças na mãe nos quatro primeiros meses de gestação, ingestão medicamentosa nesse período, hereditariedade, tabagismo, consumo de bebida alcoólica e exposição a raio-X durante a gestação ou um ano antes. Foram estimados os riscos relativos , segundo cada variável, por ponto e por intervalo de 95 por cento de confiança. Empregou-se análise multivariada, adotando-se o procedimento de máxima verossimilhança incondicional, para cada tipo de fissura. As variáveis hereditariedade (RR = 4,96), epilepsia na mãe (RR = 2,39) e ingestão de anti-inflamatório (RR = 2,59) são fatores de risco para fissuras labiais ou lábio-palatais. As variáveis hereditariedade (RR = 2,82) e poluição (RR = 2,58) são fatores de risco para fissuras palatinas. / There are two groups of clefts which differ in their etiology, namely the group with cleft lip with or without cleft palate and the group of cleft palate isolated. This study refers to a case-central analysis, with 450 cases and 450 controls. Among the cases, 354 have cleft lip with or without cleft palate and 96 have cleft palate. In order to study the association between oral clefts and possible risk factors, the analised variables were: place of mother\'s residence (urban/rural); pollution, parents\'s diseases, mother\'s diseases during the first four months of pregnancy, intake of drugs related to this period, heredity, smoking habits, alcohol consumption and X-ray examinations during pregnancy or X-ray examinations prior to pregnancy. It was estimated the relative risks CRR), for each variable by point and by 95 per cent of confidente interval. It was applied multivariate analysis, for unconditional maximum likelihood procedure, according to each type of cleft. Related to cleft lip with or without cleft palate, the risk factors are heredity (CRR = 4,96), epilepsy in the mother (RR = 2,39) and drug intake as anti-inflammatory in the first four months of pregnancy (RR = 2,59). Related to cleft palate, the risk factors are heredity (RR = 2,82) and pollution (RR = 2,58).

Esta dissertação de mestrado tem como objetivo estudar a relação entre as propriedades semântico-pragmáticas das clivadas e as do advérbio somente. Mais especificamente, procura, por meio do estudo de ocorrências de clivadas com e sem somente, delimitar tais propriedades, verificando se os chamados “efeitos de exaustividade” (Kiss, 1998) são propriedades inerentes a esse tipo de construção. De acordo com Teixeira e Menuzzi (2015), a impossibilidade do uso de clivadas com somente em certos contextos indica que ela não é exaustiva – propriedades independentes de somente, contudo, também podem ser responsáveis por essa inadequação. Em nosso estudo, exploramos as razões que levam a essa inadequação, a partir de duas hipóteses: (i) da pressuposição de unicidade e (ii) das implicações de somente: a prejacente e a de exclusão. De acordo com Szabolcsi (1994); Wedgwood et. al (2006), a primeira hipótese afirma que clivadas pressupõem que apenas um único indivíduo satisfaz a predicação expressa em sua oração, o que nos leva a concluir que não pode ser adequada com somente, já que seria redundante assertar exclusão. A segunda hipótese, contudo (cf. Horn 1969, 1981), afirma que clivadas pressupõem apenas existência – não sendo incompatível com a noção de exclusão, permitiria que fossem usadas com somente. A partir de análises de clivadas com e sem somente em diversos contextos, nossos julgamentos de aceitabilidade indicam que, embora clivadas possam pressupor unicidade, esse não é, de fato, um requisito necessário para o seu uso. E, no caso de somente, também verificamos que não basta que clivadas pressuponham apenas existência, as implicações de somente também precisar ser satisfeitas contextualmente. Em resumo, acreditamos que a pressuposição da clivada não precisa ser de unicidade, não implica necessariamente exclusão e é, portanto, mais fraca do que a literatura em geral supõe. Com objetivo de verificar nossos julgamentos de aceitabilidade, aplicamos testes Likert em um grupo diverso de falantes de língua portuguesa, para que julgassem textos contendo clivadas com e sem somente. Casos em que a clivada com somente pressupõe apenas existência parecem se correlacionar com os nossos julgamentos, mas casos de unicidade e de satisfação dos requisitos de somente mostram-se problemáticos – em especial quando somente envolve alguma noção de “contra-expectativa”. Esses casos, estudados em pormenor, indicam a necessidade de controlarmos mais fortemente variáveis que envolvam a leitura dos textos aplicados e de investigar outros tipos de inferências possivelmente levantados por somente. / This thesis studies the relationship between the semantic-pragmatic properties of cleft sentences and the adverb only. More specifically, it aims at investigating whether the so-called “exhaustive effects” (Kiss, 1998) are inherent to clefts, through the study of clefts with and without only. According to Teixeira & Menuzzi (2015), the impossibility of clefts with only in certain contexts indicates that they are not exhaustive; independent properties of only may also be responsible for such inadequacy, though – requiring the application of other expressions to verify it. In our study, we analyzed the (in)adequacy of clefts with only based on two hypotheses: (i) the presupposition of uniqueness and (ii) the two implications of only, the prejacent implication and the exclusive implication. According to Szabolcsi (1994); Wedgwood et. all (2006), the first hypothesis assumes that clefts presuppose uniqueness, i. e., they presuppose that only one individual satisfies the predication expressed by the cleft clause. In such cases, only, it seems, is not appropriate in clefts, since it would be redundant to assert exclusion. Horn (1969, 1981), however, states that clefts presuppose existence, therefore allowing only to assert exclusion. Based on analysis of our own acceptability judgements of clefts with and without only, we find that, although clefts can presuppose uniqueness, this is not a necessary condition, and, when it comes to only, it is not enough that clefts simply presuppose existence – the implications of only need also be satisfied, as Teixeira & Menuzzi (2015) have pointed out. In summary, we believe that the presupposition of clefts need not be of uniqueness, and need not imply exclusion: they are weaker than the literature generally suggests. In order to confirm our own judgments, Likert tests were applied on a range of Portuguese speakers to judge texts containing clefts with and without only. The results show that cases in which the cleft with only presupposes only existence seem to confirm our hypotheses, but cases of uniqueness and requirements of the implications of only seem to be problematic – especially when only involves some notion of “counter-expectation”. These cases, studied in detail, indicate the need to better control variables such as reading proficiency and to investigate other possible inferences of only.

This research concerns the structural and discourse related properties of cleft constructions found in spoken New Zealand English. In particular, the main analysis focuses on one cleft type, namely on the demonstrative cleft; examples include (a) 'That's what I had in mind' and (b) 'That's what I thought'. The demonstrative cleft has received little attention in the literature, and this is reflected in its inconsistent classification (some believe it to be a reversed wh-cleft, others classify it together with it-clefts, and others still use the label ‘th-cleft’). The current work investigates the clefts exemplified in (a) and (b), in terms of 23 different structural and discourse related properties. These properties were identified by consulting existing literature on clefts, and data from the Wellington Corpus of Spoken New Zealand English (about 200,000 words of spontaneous conversation). Additionally, the same excerpts of conversation were also examined for it-clefts, wh-clefts and reversed wh-clefts, whose most significant properties in spoken language are also related here. The data were tagged manually for the various cleft constructions investigated, and difficult examples were cross-checked by and discussed with other linguists. The thesis consists of three introductory chapters (Chapters 1, 2 and 3), which introduce the data investigated and the constructions analysed. Chapter 4 presents the analysis of the demonstrative cleft. Following this, peripheral and problematic demonstrative clefts, that is, clefts which deviate from the prototypical demonstrative cleft model are discussed in Chapter 5. Finally, Chapter 6 deals with other cleft types in English, namely it-clefts, wh-clefts and reversed wh-clefts, and their most significant properties in spoken language. The thesis concludes with a summary chapter (Chapter 7). One innovative aspect of the research concerns the fact that in spoken language, clefts can be 'un-integrated' or loosely integrated inside the syntactic structure which they are part of, while still being tightly connected within the discourse portion in which they are found, e.g., 'That's what you have to do when moving into a new house is nest', and 'That is what the government wants you to do is to vote Labour' (termed here double cleft construction). Double clefts are discussed in Chapter 4. The corpus also contains un-integrated wh-clefts, such as 'What I want to do is I want to study clefts', treated in Chapter 5. Previous studies suggest that in spoken language, the distinction between the syntax of clauses and the overall organisation of discourse is not always clear; clauses which do not appear to be syntactically subordinate may nonetheless be subordinate in terms of the discourse role they play. This is problematic for existing syntactic theories which rely on tightly integrated structures. An adequate analysis of un-integrated constructions, in terms of their syntax and discourse function will be of interest both to theoretical syntax and to computational linguistics. The research contributes to existing knowledge of the grammatical constructions used by speakers of New Zealand English and English worldwide. Given the significant differences between the grammar of spoken language and that of written language, this work helps increase understanding of spoken language and of what it means to be a ‘speaker’.

This research concerns the structural and discourse related properties of cleft constructions found in spoken New Zealand English. In particular, the main analysis focuses on one cleft type, namely on the demonstrative cleft; examples include (a) 'That's what I had in mind' and (b) 'That's what I thought'. The demonstrative cleft has received little attention in the literature, and this is reflected in its inconsistent classification (some believe it to be a reversed wh-cleft, others classify it together with it-clefts, and others still use the label ‘th-cleft’). The current work investigates the clefts exemplified in (a) and (b), in terms of 23 different structural and discourse related properties. These properties were identified by consulting existing literature on clefts, and data from the Wellington Corpus of Spoken New Zealand English (about 200,000 words of spontaneous conversation). Additionally, the same excerpts of conversation were also examined for it-clefts, wh-clefts and reversed wh-clefts, whose most significant properties in spoken language are also related here. The data were tagged manually for the various cleft constructions investigated, and difficult examples were cross-checked by and discussed with other linguists. The thesis consists of three introductory chapters (Chapters 1, 2 and 3), which introduce the data investigated and the constructions analysed. Chapter 4 presents the analysis of the demonstrative cleft. Following this, peripheral and problematic demonstrative clefts, that is, clefts which deviate from the prototypical demonstrative cleft model are discussed in Chapter 5. Finally, Chapter 6 deals with other cleft types in English, namely it-clefts, wh-clefts and reversed wh-clefts, and their most significant properties in spoken language. The thesis concludes with a summary chapter (Chapter 7). One innovative aspect of the research concerns the fact that in spoken language, clefts can be 'un-integrated' or loosely integrated inside the syntactic structure which they are part of, while still being tightly connected within the discourse portion in which they are found, e.g., 'That's what you have to do when moving into a new house is nest', and 'That is what the government wants you to do is to vote Labour' (termed here double cleft construction). Double clefts are discussed in Chapter 4. The corpus also contains un-integrated wh-clefts, such as 'What I want to do is I want to study clefts', treated in Chapter 5. Previous studies suggest that in spoken language, the distinction between the syntax of clauses and the overall organisation of discourse is not always clear; clauses which do not appear to be syntactically subordinate may nonetheless be subordinate in terms of the discourse role they play. This is problematic for existing syntactic theories which rely on tightly integrated structures. An adequate analysis of un-integrated constructions, in terms of their syntax and discourse function will be of interest both to theoretical syntax and to computational linguistics. The research contributes to existing knowledge of the grammatical constructions used by speakers of New Zealand English and English worldwide. Given the significant differences between the grammar of spoken language and that of written language, this work helps increase understanding of spoken language and of what it means to be a ‘speaker’.

This research concerns the structural and discourse related properties of cleft constructions found in spoken New Zealand English. In particular, the main analysis focuses on one cleft type, namely on the demonstrative cleft; examples include (a) 'That's what I had in mind' and (b) 'That's what I thought'. The demonstrative cleft has received little attention in the literature, and this is reflected in its inconsistent classification (some believe it to be a reversed wh-cleft, others classify it together with it-clefts, and others still use the label ‘th-cleft’). The current work investigates the clefts exemplified in (a) and (b), in terms of 23 different structural and discourse related properties. These properties were identified by consulting existing literature on clefts, and data from the Wellington Corpus of Spoken New Zealand English (about 200,000 words of spontaneous conversation). Additionally, the same excerpts of conversation were also examined for it-clefts, wh-clefts and reversed wh-clefts, whose most significant properties in spoken language are also related here. The data were tagged manually for the various cleft constructions investigated, and difficult examples were cross-checked by and discussed with other linguists. The thesis consists of three introductory chapters (Chapters 1, 2 and 3), which introduce the data investigated and the constructions analysed. Chapter 4 presents the analysis of the demonstrative cleft. Following this, peripheral and problematic demonstrative clefts, that is, clefts which deviate from the prototypical demonstrative cleft model are discussed in Chapter 5. Finally, Chapter 6 deals with other cleft types in English, namely it-clefts, wh-clefts and reversed wh-clefts, and their most significant properties in spoken language. The thesis concludes with a summary chapter (Chapter 7). One innovative aspect of the research concerns the fact that in spoken language, clefts can be 'un-integrated' or loosely integrated inside the syntactic structure which they are part of, while still being tightly connected within the discourse portion in which they are found, e.g., 'That's what you have to do when moving into a new house is nest', and 'That is what the government wants you to do is to vote Labour' (termed here double cleft construction). Double clefts are discussed in Chapter 4. The corpus also contains un-integrated wh-clefts, such as 'What I want to do is I want to study clefts', treated in Chapter 5. Previous studies suggest that in spoken language, the distinction between the syntax of clauses and the overall organisation of discourse is not always clear; clauses which do not appear to be syntactically subordinate may nonetheless be subordinate in terms of the discourse role they play. This is problematic for existing syntactic theories which rely on tightly integrated structures. An adequate analysis of un-integrated constructions, in terms of their syntax and discourse function will be of interest both to theoretical syntax and to computational linguistics. The research contributes to existing knowledge of the grammatical constructions used by speakers of New Zealand English and English worldwide. Given the significant differences between the grammar of spoken language and that of written language, this work helps increase understanding of spoken language and of what it means to be a ‘speaker’.

This research concerns the structural and discourse related properties of cleft constructions found in spoken New Zealand English. In particular, the main analysis focuses on one cleft type, namely on the demonstrative cleft; examples include (a) 'That's what I had in mind' and (b) 'That's what I thought'. The demonstrative cleft has received little attention in the literature, and this is reflected in its inconsistent classification (some believe it to be a reversed wh-cleft, others classify it together with it-clefts, and others still use the label ‘th-cleft’). The current work investigates the clefts exemplified in (a) and (b), in terms of 23 different structural and discourse related properties. These properties were identified by consulting existing literature on clefts, and data from the Wellington Corpus of Spoken New Zealand English (about 200,000 words of spontaneous conversation). Additionally, the same excerpts of conversation were also examined for it-clefts, wh-clefts and reversed wh-clefts, whose most significant properties in spoken language are also related here. The data were tagged manually for the various cleft constructions investigated, and difficult examples were cross-checked by and discussed with other linguists. The thesis consists of three introductory chapters (Chapters 1, 2 and 3), which introduce the data investigated and the constructions analysed. Chapter 4 presents the analysis of the demonstrative cleft. Following this, peripheral and problematic demonstrative clefts, that is, clefts which deviate from the prototypical demonstrative cleft model are discussed in Chapter 5. Finally, Chapter 6 deals with other cleft types in English, namely it-clefts, wh-clefts and reversed wh-clefts, and their most significant properties in spoken language. The thesis concludes with a summary chapter (Chapter 7). One innovative aspect of the research concerns the fact that in spoken language, clefts can be 'un-integrated' or loosely integrated inside the syntactic structure which they are part of, while still being tightly connected within the discourse portion in which they are found, e.g., 'That's what you have to do when moving into a new house is nest', and 'That is what the government wants you to do is to vote Labour' (termed here double cleft construction). Double clefts are discussed in Chapter 4. The corpus also contains un-integrated wh-clefts, such as 'What I want to do is I want to study clefts', treated in Chapter 5. Previous studies suggest that in spoken language, the distinction between the syntax of clauses and the overall organisation of discourse is not always clear; clauses which do not appear to be syntactically subordinate may nonetheless be subordinate in terms of the discourse role they play. This is problematic for existing syntactic theories which rely on tightly integrated structures. An adequate analysis of un-integrated constructions, in terms of their syntax and discourse function will be of interest both to theoretical syntax and to computational linguistics. The research contributes to existing knowledge of the grammatical constructions used by speakers of New Zealand English and English worldwide. Given the significant differences between the grammar of spoken language and that of written language, this work helps increase understanding of spoken language and of what it means to be a ‘speaker’.

Esta dissertação de mestrado tem como objetivo estudar a relação entre as propriedades semântico-pragmáticas das clivadas e as do advérbio somente. Mais especificamente, procura, por meio do estudo de ocorrências de clivadas com e sem somente, delimitar tais propriedades, verificando se os chamados “efeitos de exaustividade” (Kiss, 1998) são propriedades inerentes a esse tipo de construção. De acordo com Teixeira e Menuzzi (2015), a impossibilidade do uso de clivadas com somente em certos contextos indica que ela não é exaustiva – propriedades independentes de somente, contudo, também podem ser responsáveis por essa inadequação. Em nosso estudo, exploramos as razões que levam a essa inadequação, a partir de duas hipóteses: (i) da pressuposição de unicidade e (ii) das implicações de somente: a prejacente e a de exclusão. De acordo com Szabolcsi (1994); Wedgwood et. al (2006), a primeira hipótese afirma que clivadas pressupõem que apenas um único indivíduo satisfaz a predicação expressa em sua oração, o que nos leva a concluir que não pode ser adequada com somente, já que seria redundante assertar exclusão. A segunda hipótese, contudo (cf. Horn 1969, 1981), afirma que clivadas pressupõem apenas existência – não sendo incompatível com a noção de exclusão, permitiria que fossem usadas com somente. A partir de análises de clivadas com e sem somente em diversos contextos, nossos julgamentos de aceitabilidade indicam que, embora clivadas possam pressupor unicidade, esse não é, de fato, um requisito necessário para o seu uso. E, no caso de somente, também verificamos que não basta que clivadas pressuponham apenas existência, as implicações de somente também precisar ser satisfeitas contextualmente. Em resumo, acreditamos que a pressuposição da clivada não precisa ser de unicidade, não implica necessariamente exclusão e é, portanto, mais fraca do que a literatura em geral supõe. Com objetivo de verificar nossos julgamentos de aceitabilidade, aplicamos testes Likert em um grupo diverso de falantes de língua portuguesa, para que julgassem textos contendo clivadas com e sem somente. Casos em que a clivada com somente pressupõe apenas existência parecem se correlacionar com os nossos julgamentos, mas casos de unicidade e de satisfação dos requisitos de somente mostram-se problemáticos – em especial quando somente envolve alguma noção de “contra-expectativa”. Esses casos, estudados em pormenor, indicam a necessidade de controlarmos mais fortemente variáveis que envolvam a leitura dos textos aplicados e de investigar outros tipos de inferências possivelmente levantados por somente. / This thesis studies the relationship between the semantic-pragmatic properties of cleft sentences and the adverb only. More specifically, it aims at investigating whether the so-called “exhaustive effects” (Kiss, 1998) are inherent to clefts, through the study of clefts with and without only. According to Teixeira & Menuzzi (2015), the impossibility of clefts with only in certain contexts indicates that they are not exhaustive; independent properties of only may also be responsible for such inadequacy, though – requiring the application of other expressions to verify it. In our study, we analyzed the (in)adequacy of clefts with only based on two hypotheses: (i) the presupposition of uniqueness and (ii) the two implications of only, the prejacent implication and the exclusive implication. According to Szabolcsi (1994); Wedgwood et. all (2006), the first hypothesis assumes that clefts presuppose uniqueness, i. e., they presuppose that only one individual satisfies the predication expressed by the cleft clause. In such cases, only, it seems, is not appropriate in clefts, since it would be redundant to assert exclusion. Horn (1969, 1981), however, states that clefts presuppose existence, therefore allowing only to assert exclusion. Based on analysis of our own acceptability judgements of clefts with and without only, we find that, although clefts can presuppose uniqueness, this is not a necessary condition, and, when it comes to only, it is not enough that clefts simply presuppose existence – the implications of only need also be satisfied, as Teixeira & Menuzzi (2015) have pointed out. In summary, we believe that the presupposition of clefts need not be of uniqueness, and need not imply exclusion: they are weaker than the literature generally suggests. In order to confirm our own judgments, Likert tests were applied on a range of Portuguese speakers to judge texts containing clefts with and without only. The results show that cases in which the cleft with only presupposes only existence seem to confirm our hypotheses, but cases of uniqueness and requirements of the implications of only seem to be problematic – especially when only involves some notion of “counter-expectation”. These cases, studied in detail, indicate the need to better control variables such as reading proficiency and to investigate other possible inferences of only.