This study is about voice system in Burushaski, focusing especially on the middle voice (MV) construction. It claims that the [dd-] verbal prefix is an overt morphological middle marker for MV constructions, while the [n-] verbal prefix is a morphological marker for passive voice. The data primarily come from the Hunza dialect of Burushaski, but analogous phenomena can be observed in other dialects. This research is based on a corpus of 120 dd-prefix verbs. This research has showed that position {-2} on the verb template is occupied by voice-marker in Burushaski. The author argues that the middle marker is a semantic category of its own and that it is clearly distinguished from the reflexive marker in this language. The analysis of the phenomenon in this study only comes from the dialect of Hunza Burushaski, so a lot of research remains to be done on the other three dialects of Burushaski: Yasin dialect, Nagar dialect and Srinagar dialect.

Based on 160 hours of recording collected in the villages of Săbăoani, and Pildeşti, Romania, the present research attempts to describe the middle voice system of Northern Moldavian Hungarian (NMH), an endangered language spoken by no more than 3000 speakers. Defining the middle voice category semantically rather than formally, it is argued that the various middle situation types in NMH can be placed relative to one another on a “semantic map” based on shared semantic properties such as 1) the confinement of the development of the action within the agent’s sphere to the extent that the action’s effect accrues back on the agent itself, 2) the degree of volitionality of the Initiator/Agent, and 3) the degree of affectedness of the Initiator/Agent. Polysemy structures are examined against the background of a common semantic map derived on the basis of cross linguistic investigation of a given grammatical domain. In working toward this end a detailed description of major patterns of meaning inherent in the NMH middle system, examining three types of morphological middles, syntactic middles, and lexical middles is presented. Cases in which the same verb can occur with or without a middle marker apparently having the same meaning are discussed. Moreover, seemingly minimal pairs in which two different morphological constructions occur with the same verb are analyzed. A detailed analysis of the differences in form and function of the two reciprocal syntactic middle constructions in NMH is provided. Regarding reflexive syntactic middles it will be shown that depending on the case marking taken by the reflexive anaphoric operator the function conveyed is different such as reflexives, intensifiers, causers, and experiencer. Finally, cases in which the same verb can convey a middle meaning by using a morphological middle marker or by using a syntactic middle construction are analyzed showing that there are main differences in the meaning those two strategies convey. Thus, the present paper identifies specific semantic properties relevant to the middle voice system in NMH, sets up some hypotheses regarding the relations among middle and related situation types and proposes some diachronic predictions regarding the middle voice system of NMH.

This thesis investigates "voice mismatch verbs" (deponents), verbs that take non-active morphology but are used in syntactically active environments. The focus is on the non-informant Indo-European languages Hittite, Vedic Sanskrit, Ancient Greek, and Latin, supplemented by data from Modern Greek. / Linguistics

O presente trabalho tem como objetivo investigar o uso do clítico se nas construções do português brasileiro, a partir de dados da modalidade oral. Tendo em vista que esse clítico está associado a uma diversidade de significados, buscamos identificar em quais tipos de construções ele é empregado e como ele contribui para a interpretação das sentenças em que aparece. Para isso, recorremos a dois corpora de língua oral: (i) C-ORAL-BRASIL e (ii) Projeto SP2010: Amostra da fala paulistana. O primeiro reflete a fala mineira, ao passo que o segundo retrata a produção de falantes paulistanos. Os dados foram tratados pelo viés da Gramática Cognitiva, uma vez que essa perspectiva teórica possibilitou-nos olhar para fatores relevantes para o entendimento da voz média. A proposta de Langacker (2008) acerca dos diferentes tipos de construções, assim como a de Maldonado (2006) no que concerne à elaboração da força indutora do evento, serviu de base para a análise das sentenças que se encontram em uma zona intermediária entre construções impessoais, de um lado, e construções absolutas, de outro. Dentro desse contínuo, a elaboração da força indutora se dá em graus variados. Nas sentenças impessoais, o clítico introduz no evento uma força indutora que, embora seja humana, é genérica ou arbitrária. Nas sentenças médias de proeminência terminal, o se representa uma força indutora externa ainda mais esquemática, o que faz com que o foco da sentença recaia sobre a porção final do evento. Além disso, há os casos em que a força indutora é ainda menos elaborada, remetendo a uma energia mínima, próxima das construções absolutas. Estas não envolvem a conceitualização de uma força indutora. Para entender o contínuo reflexivo-médio, partimos dos trabalhos de Kemmer (1993, 1994), que ressalta fatores como a distinguibilidade dos participantes e a prototipicidade da ação, e de Maldonado (2006), que destaca o grau de controle exercido pelo sujeito. Enquanto as construções reflexivas envolvem maior distinguibilidade na conceitualização das facetas do participante e maior controle por parte deste, as médias refletem uma conceitualização indistinguível do participante e, em diversos casos, menor grau de controle. Além disso, a proposta acerca do experienciador médio (Maldonado, 2006) mostra que o clítico tem relação com o nível de envolvimento do participante no evento. Com base nas análises desenvolvidas dentro desse quadro teórico, bem como nas evidências empíricas fornecidas pelos corpora, defendemos que o clítico está associado a um processo de energização do evento (Maldonado, 2006), seja conferindo a introdução de uma força indutora, seja indicando maior envolvimento do participante. Com este estudo, pretendemos contribuir para o entendimento do papel do clítico se no português brasileiro e sua relação com a conceitualização de eventos. / The present study aims at investigating the use of clitic se in Brazilian Portuguese constructions, by means of empirical analysis. Considering this clitic is associated with a variety of meanings, this research has tried to identify in what types of constructions it is used and how it contributes to the interpretation of the sentences. In order to accomplish this goal, two oral language corpora were examined: (i) C-ORAL-BRASIL and (ii) Projeto SP2010. The first one reflects the dialect spoken in Minas Gerais, and the second portrays the dialect spoken in the city of São Paulo. Data were treated according to Cognitive Grammars perspective, since this theoretical approach has pointed out some aspects that proved to be relevant for the understanding of the middle voice. Langackers (2008) proposal about the different types of constructions, as well as Maldonados (2006) thesis concerning the elaboration of the inductive force of the event, was taken as a starting point for the analysis of the sentences located in an intermediate zone between impersonal constructions, on one side, and absolute constructions, on the other side. Within this zone, the inductive force can be elaborated in different degrees. In impersonal constructions, clitic se introduces a human force that has a generic or arbitrary interpretation. In middle sentences that exhibit prominence in the final part of the event, the clitic represents an external and highly schematic inductive force it is precisely this schematization that causes the effect of final prominence. Furthermore, there are cases in which the inductive force has an even lower level of elaboration. In such cases, the clitic makes reference to a minimal level of energy, one that is closer to absolute constructions (in which no inductive force is conceptualized). In order to understand the reflexive-middle continuum, we departed from Kemmers study (1993, 1994) which emphasizes aspects such as the relative distinguishability of participants and the expectation that initiator and endpoint have the same identity and Maldonados work (2006), which highlights the degree to which the participant controls the event. While reflexive constructions involve greater distinguishability in the conceptualization of the participant and greater control exerted by him/her, middle constructions reflect that the participant is being conceived as a unitary entity and that, in many cases, this participant does not exert as much control over the event. Besides, Maldonados (2006) proposal regarding the middle experiencer shows that the clitic is related to the degree the participant is involved in the event. Based on the analyses developed within this theoretical frame and on empirical evidence, we defend that the clitic is associated with a process of energization of the event (Maldonado, 2006). This is accomplished either by introducing an inductive force or by indicating a greater involvement of the participant in the event. In this study, we intend to contribute to the understanding of the role played by clitic se in Brazilian Portuguese constructions and its relationship with event conceptualization.

O presente trabalho tem como objetivo investigar o uso do clítico se nas construções do português brasileiro, a partir de dados da modalidade oral. Tendo em vista que esse clítico está associado a uma diversidade de significados, buscamos identificar em quais tipos de construções ele é empregado e como ele contribui para a interpretação das sentenças em que aparece. Para isso, recorremos a dois corpora de língua oral: (i) C-ORAL-BRASIL e (ii) Projeto SP2010: Amostra da fala paulistana. O primeiro reflete a fala mineira, ao passo que o segundo retrata a produção de falantes paulistanos. Os dados foram tratados pelo viés da Gramática Cognitiva, uma vez que essa perspectiva teórica possibilitou-nos olhar para fatores relevantes para o entendimento da voz média. A proposta de Langacker (2008) acerca dos diferentes tipos de construções, assim como a de Maldonado (2006) no que concerne à elaboração da força indutora do evento, serviu de base para a análise das sentenças que se encontram em uma zona intermediária entre construções impessoais, de um lado, e construções absolutas, de outro. Dentro desse contínuo, a elaboração da força indutora se dá em graus variados. Nas sentenças impessoais, o clítico introduz no evento uma força indutora que, embora seja humana, é genérica ou arbitrária. Nas sentenças médias de proeminência terminal, o se representa uma força indutora externa ainda mais esquemática, o que faz com que o foco da sentença recaia sobre a porção final do evento. Além disso, há os casos em que a força indutora é ainda menos elaborada, remetendo a uma energia mínima, próxima das construções absolutas. Estas não envolvem a conceitualização de uma força indutora. Para entender o contínuo reflexivo-médio, partimos dos trabalhos de Kemmer (1993, 1994), que ressalta fatores como a distinguibilidade dos participantes e a prototipicidade da ação, e de Maldonado (2006), que destaca o grau de controle exercido pelo sujeito. Enquanto as construções reflexivas envolvem maior distinguibilidade na conceitualização das facetas do participante e maior controle por parte deste, as médias refletem uma conceitualização indistinguível do participante e, em diversos casos, menor grau de controle. Além disso, a proposta acerca do experienciador médio (Maldonado, 2006) mostra que o clítico tem relação com o nível de envolvimento do participante no evento. Com base nas análises desenvolvidas dentro desse quadro teórico, bem como nas evidências empíricas fornecidas pelos corpora, defendemos que o clítico está associado a um processo de energização do evento (Maldonado, 2006), seja conferindo a introdução de uma força indutora, seja indicando maior envolvimento do participante. Com este estudo, pretendemos contribuir para o entendimento do papel do clítico se no português brasileiro e sua relação com a conceitualização de eventos. / The present study aims at investigating the use of clitic se in Brazilian Portuguese constructions, by means of empirical analysis. Considering this clitic is associated with a variety of meanings, this research has tried to identify in what types of constructions it is used and how it contributes to the interpretation of the sentences. In order to accomplish this goal, two oral language corpora were examined: (i) C-ORAL-BRASIL and (ii) Projeto SP2010. The first one reflects the dialect spoken in Minas Gerais, and the second portrays the dialect spoken in the city of São Paulo. Data were treated according to Cognitive Grammars perspective, since this theoretical approach has pointed out some aspects that proved to be relevant for the understanding of the middle voice. Langackers (2008) proposal about the different types of constructions, as well as Maldonados (2006) thesis concerning the elaboration of the inductive force of the event, was taken as a starting point for the analysis of the sentences located in an intermediate zone between impersonal constructions, on one side, and absolute constructions, on the other side. Within this zone, the inductive force can be elaborated in different degrees. In impersonal constructions, clitic se introduces a human force that has a generic or arbitrary interpretation. In middle sentences that exhibit prominence in the final part of the event, the clitic represents an external and highly schematic inductive force it is precisely this schematization that causes the effect of final prominence. Furthermore, there are cases in which the inductive force has an even lower level of elaboration. In such cases, the clitic makes reference to a minimal level of energy, one that is closer to absolute constructions (in which no inductive force is conceptualized). In order to understand the reflexive-middle continuum, we departed from Kemmers study (1993, 1994) which emphasizes aspects such as the relative distinguishability of participants and the expectation that initiator and endpoint have the same identity and Maldonados work (2006), which highlights the degree to which the participant controls the event. While reflexive constructions involve greater distinguishability in the conceptualization of the participant and greater control exerted by him/her, middle constructions reflect that the participant is being conceived as a unitary entity and that, in many cases, this participant does not exert as much control over the event. Besides, Maldonados (2006) proposal regarding the middle experiencer shows that the clitic is related to the degree the participant is involved in the event. Based on the analyses developed within this theoretical frame and on empirical evidence, we defend that the clitic is associated with a process of energization of the event (Maldonado, 2006). This is accomplished either by introducing an inductive force or by indicating a greater involvement of the participant in the event. In this study, we intend to contribute to the understanding of the role played by clitic se in Brazilian Portuguese constructions and its relationship with event conceptualization.

FundaÃÃo Cearense de Apoio ao Desenvolvimento Cientifico e TecnolÃgico / Este trabalho visa caracterizar formal, semÃntica e pragmaticamente as construÃÃes passivas, mÃdias e impessoais, com o fim de elucidar a natureza da relaÃÃo entre estas construÃÃes, no portuguÃs arcaico, relaÃÃo essa observada em trabalhos, como os de Camacho (2002, 2003, 2006), para o portuguÃs atual. Parte do princÃpio de que o traÃo comum a essas construÃÃes à a nÃo-atribuiÃÃo de causalidade, e de que estas construÃÃes refletem diferentes conceitualizaÃÃes de um evento. Com base nesses princÃpios, analisa, numa abordagem cognitivo-funcional (GIVÃN, 1993, 1995, 2002, 2005; KEMMER, 1993; CAMACHO, 2002, 2003; HOPPER & THOMPSON, 1980; POTTIER, 1992; CROFT, 1994, 1998; LANGACKER, 1987, 1991; DELANCEY, 1987; TALMY, 1988, 2000; LAKOFF, 1977), a codificaÃÃo da funÃÃo nÃo-atribuiÃÃo de causalidade na CrÃnica Geral da Espanha de 1344, prosa histÃrica representante do portuguÃs arcaico, editada por Cintra (1951). Nessa anÃlise, busca respostas para os seguintes problemas: (a) dos recursos verbais disponÃveis, no corpus analisado, para a expressÃo da nÃo-atribuiÃÃo de causalidade, qual a mais e a menos prototÃpica? (b) que traÃos semÃntico-pragmÃticos e formais caracterizam as diferentes estratÃgias? A hipÃtese central considera que as construÃÃes analisadas codificam a nÃo-atribuiÃÃo da causalidade em variados graus, conforme fatores ligados à noÃÃo de causalidade e à transitividade, como a perspectivaÃÃo e a saliÃncia. Foram coletadas 1061 ocorrÃncias e analisadas quanto a fatores pragmÃticos, como a topicalidade e o estatuto informacional; fatores semÃnticos, como saliÃncia cognitiva do Afetado e do Causativo, traÃo [animado], tipo de afetaÃÃo; e fatores formais, como a ordem e a expressÃo. AlÃm disso, todas foram graduadas quanto ao grau de transitividade e de relevo discursivo. Os resultados apontaram que a mÃdia à a mais prototÃpica das construÃÃes de nÃo-atribuiÃÃo de causalidade, por ser a que apresenta causativo menos saliente e menor grau de transitividade. A passiva, mais freqÃente no corpus e, muitas vezes, com Causativo omitido, caracterizou-se como menos prototÃpica por seu Causativo ser freqÃentemente evocado e apresentar maior grau de transitividade. Do mesmo modo, a impessoal, cujo causativo à freqÃentemente inferÃvel, manifestou-se como mais causativa e mais transitiva que a mÃdia, todavia menos que a passiva. A anÃlise demonstrou, enfim, que o fenÃmeno da nÃo-atribuiÃÃo de causalidade nÃo à um fenÃmeno isolado, limitado à manifestaÃÃo ou nÃo de um Agente. Submete-se a graus e acha-se ligado, especialmente, ao contorno tÃmporo-aspectual do evento. Devido à noÃÃo experiencial de causalidade, ligada à idÃia de movimento, eventos perfectivos sÃo mais causais que eventos imperfectivos. / This work aims to characterize formally, semantically and pragmatically the passive, middle and impersonal constructions, in order to elucidate the nature of relation between these constructions in a archaic Portuguese, a relation observed in researches, such as those of Camacho (2002, 2003, 2006), in modern Portuguese. The present research, thus, claims that the common feature of these constructions is the non-attribution of causality, and that they represent different event conceptualizations. Based on principles, founded on a cognitive-functional approach (GIVÃN, 1993, 1995, 2002, 2005; KEMMER, 1993; CAMACHO, 2002, 2003, 2006; HOPPER & THOMPSON, 1980; POTTIER, 1992; CROFT, 1994, 1998; LANGACKER, 1987, 1991; DELANCEY, 1987; TALMY, 1988, 2000; LAKOFF, 1977), the codification of the non-attribution of causality is analyzed in CrÃnica Geral de Espanha de 1344, a representative historical text in archaic Portuguese, edited by Cintra (1951). In this analysis, answers are sought for the following questions: (a) in the analyzed corpus, which is the most and the least prototypical verbal recourses available for the expression of the non-attribution of causality? (b) which semantic, pragmatical and formal features characterize the different strategies used for expressing the non-attribution of causality? The central hypothesis considers that analyzed constructions codify the non-attribution of causality at various degrees, according to aspects related to the causality notion, such as perspectivization and salience. A large number of occurrences (1061, in total) were collected and analyzed from the perspective of pragmatical aspects such as topicality and informational status; semantic aspects, such as cognitive salience of Affected and of Causative, animacy, affectation type; and formal aspects, such as order and expression. Moreover, all of such occurrences were graduated regarding the transitivity degree and the Grounding. The results indicate that middle construction is the most prototypical of the non-attribution of causality constructions, because it presents less salient causative and is posited in the least degree of transitivity. Passive, most frequent in the corpus and, often, with omitted causative, was characterized as the least prototypical construction because its causative is commonly evoked and it presents a higher degree of transitivity. Impersonal construction, whose causative is frequently inferable, was characterized as more causative and more transitive than the middle construction, however, less than passive. Therefore, the analyses demonstrated that non-attribution of causality is not an isolated phenomenon, limited to the manifestation of an Agent. It is subject to degrees of escalarization which is attached to temporal and aspectual event contours. Because of experiential notion of causality, linked to the idea movement, perfective events are more causal than imperfective ones.

Transitive reflexive Sätze des Deutschen lassen sich unter den Begriff des Mediums fassen. Genauso wie entsprechende Konstruktionen in anderen indoeuropäischen Sprachen sind sie auch im Deutschen mehrdeutig und erlauben eine reflexive, mediale, antikausative und inherent reflexive Interpretation. Nach einem Überblick über die für die folgende Diskussion wesentlichen syntaktischen und semantischen Eigenschaften transitiver reflexiver Sätze wird anhand der Medialkonstruktion gezeigt, daß weder lexikalische noch syntaktische Ansätze in der Lage sind, eine korrekte und einheitliche Analyse der Medialkonstruktion im speziellen und von transitiven reflexiven Sätzen im allgemeinen zu bieten. Deshalb wird für einen neuen, dritten Ansatz argumentiert: a) alle transitiven reflexiven Sätze sind syntaktisch einheitlich zu analysieren; b) vermeintliche syntaktische Unterschiede bzgl. Koordination, Fokus und Voranstellung des (Argument- und Nichtargument-) Reflexivums lassen sich semantisch ableiten; c) (schwache) Reflexivpronomen sind bzgl. ihrer morphosyntaktischen Merkmale maximal unterspezifiziert; d) im Deutschen muß zwischen strukturellen und obliquen Kasusformen unterschieden werden. (c) und (d) erlauben eine einheitliche Analyse der Mehrdeutigkeit des Reflexivpronomens in transitiven reflexiven Sätzen im Rahmen einer entsprechend modifizierten Bindungstheorie. Die in dieser Arbeit entwickelte Theorie wird abschließend auf weitere Eigenschaften der sog. Medialkonstruktion angewandt. / Crosslinguistically, transitive reflexive sentences in German can be subsumed under the notion of middle voice. Like corresponding constructions in Indo-European languages they are also ambiguous in German and yield a reflexive, middle, anticausative, and inherent reflexive interpretation. First, we give a detailed survey of the syntactic and semantic properties of transitive reflexive sentences that will be relevant in the following discussion. Second, we illustrate by means of the middle construction that lexical and syntactic theories fail to formulate a correct and unified analysis of middle formation in particular and transitive reflexive sentences in general. Therefore, we develop a different and (so far new) approach: a) all kinds of transitive reflexive sentences do not differ in syntax; b) the differences concerning coordination, focus, and fronting of the argument and non-argument reflexive are derived in semantics; c) the morphosyntactic features of (weak) reflexive pronouns are maximally underspecified d) German distinguishes between between structural and oblique case forms. (c) and (d) permit a uniform derivation of all four interpretations of the reflexive pronoun in transitive reflexive sentences. This analysis is based on a corresponding modification of the binding theory. Finally, we apply the theory developed in this book to further propertiies of the so-called middle construction.

By investigating the discursive rules of hermeneutics and diagnosis, this study seeks to problematize particular presuppositions—most notably the presupposition of sense—of the modern disciplinary hermeneutic context. Following Barthes's consideration of the Greek modus of the middle voice as a useful notion in conceptualizing the modern scene of writing, the study advances itself toward conceptualizing a configuration of the modern reading scene in its middle-voiced permutation. In such a scene, the moment a reading attempts to read itself from without its parameters, it arrives at a spatial and temporal crisis (from the Greek krin-ein; to decide) between its action and the place (of not sense and not not sense) which exceeds the parameters delimiting the action of reading itself, but which nevertheless conditions its possibility. The grammar of this crisis is the middle voice; its condition, in the context of this study, is configured as madness. Madness is thus configured as a function of interrogation, reading and diagnosis. At the nucleus of the modem reading scene itself, this thesis opens with an introduction of the terms middle voice, crisis and madness, and then offers a consideration of three permutations of reading: Chapter Two, Chapter Three and the space between. Chapter Two considers a fictional representation of writing in the middle voice through a reading of Nabokov's Lolita, a text of fiction in the form of a "mad writer's" diary, whose historical reception has been marked by acts of appropriative censorship and clinical diagnosis. Chapter Three considers a permutation of the middlevoiced reading through a reading of Gertrude Stein's lectures on writing. This consideration is framed by fragments from the writing of Maurice Blanchot, connecting reading (as conceived by Stein) to madness, figuring the convergence of reading and madness in writing. The Interchapter, between chapters Two and Three, is an aporetic space entitled "Madness Itself." By allowing a brief and partial view of the modem clinical psychiatric setting, and by calling into question the parameters of the surrounding "chapters" themselves, this section seeks to perform, structurally and thematically, a moment of crisis recalling the middle voice.

LIMA, Maria Claudete. A não-atribuição de causalidade na Crônica geral da Espanha de 1344. 2009. 473 f. Tese (Doutorado em Linguística) – Universidade Federal do Ceará, Departamento de Letras Vernáculas, Programa de Pós-graduação em Linguística, Fortaleza-CE, 2009. / Submitted by Liliane oliveira (morena.liliane@hotmail.com) on 2012-09-25T14:29:45Z No. of bitstreams: 1 2009_TESE_MCLIMA.pdf: 5188441 bytes, checksum: e2cf67965c6e922472be995c5e7ca70e (MD5) / Approved for entry into archive by Maria Josineide Góis(josineide@ufc.br) on 2013-11-14T14:14:33Z (GMT) No. of bitstreams: 1 2009_TESE_MCLIMA.pdf: 5188441 bytes, checksum: e2cf67965c6e922472be995c5e7ca70e (MD5) / Made available in DSpace on 2013-11-14T14:14:33Z (GMT). No. of bitstreams: 1 2009_TESE_MCLIMA.pdf: 5188441 bytes, checksum: e2cf67965c6e922472be995c5e7ca70e (MD5) Previous issue date: 2009 / This work aims to characterize formally, semantically and pragmatically the passive, middle and impersonal constructions, in order to elucidate the nature of relation between these constructions in a archaic Portuguese, a relation observed in researches, such as those of Camacho (2002, 2003, 2006), in modern Portuguese. The present research, thus, claims that the common feature of these constructions is the non-attribution of causality, and that they represent different event conceptualizations. Based on principles, founded on a cognitive-functional approach (GIVÓN, 1993, 1995, 2002, 2005; KEMMER, 1993; CAMACHO, 2002, 2003, 2006; HOPPER & THOMPSON, 1980; POTTIER, 1992; CROFT, 1994, 1998; LANGACKER, 1987, 1991; DELANCEY, 1987; TALMY, 1988, 2000; LAKOFF, 1977), the codification of the non-attribution of causality is analyzed in Crônica Geral de Espanha de 1344, a representative historical text in archaic Portuguese, edited by Cintra (1951). In this analysis, answers are sought for the following questions: (a) in the analyzed corpus, which is the most and the least prototypical verbal recourses available for the expression of the non-attribution of causality? (b) which semantic, pragmatical and formal features characterize the different strategies used for expressing the non-attribution of causality? The central hypothesis considers that analyzed constructions codify the non-attribution of causality at various degrees, according to aspects related to the causality notion, such as perspectivization and salience. A large number of occurrences (1061, in total) were collected and analyzed from the perspective of pragmatical aspects such as topicality and informational status; semantic aspects, such as cognitive salience of Affected and of Causative, animacy, affectation type; and formal aspects, such as order and expression. Moreover, all of such occurrences were graduated regarding the transitivity degree and the Grounding. The results indicate that middle construction is the most prototypical of the non-attribution of causality constructions, because it presents less salient causative and is posited in the least degree of transitivity. Passive, most frequent in the corpus and, often, with omitted causative, was characterized as the least prototypical construction because its causative is commonly evoked and it presents a higher degree of transitivity. Impersonal construction, whose causative is frequently inferable, was characterized as more causative and more transitive than the middle construction, however, less than passive. Therefore, the analyses demonstrated that non-attribution of causality is not an isolated phenomenon, limited to the manifestation of an Agent. It is subject to degrees of escalarization which is attached to temporal and aspectual event contours. Because of experiential notion of causality, linked to the idea movement, perfective events are more causal than imperfective ones / Este trabalho visa caracterizar formal, semântica e pragmaticamente as construções passivas, médias e impessoais, com o fim de elucidar a natureza da relação entre estas construções, no português arcaico, relação essa observada em trabalhos, como os de Camacho (2002, 2003, 2006), para o português atual. Parte do princípio de que o traço comum a essas construções é a não-atribuição de causalidade, e de que estas construções refletem diferentes conceitualizações de um evento. Com base nesses princípios, analisa, numa abordagem cognitivo-funcional (GIVÓN, 1993, 1995, 2002, 2005; KEMMER, 1993; CAMACHO, 2002, 2003; HOPPER & THOMPSON, 1980; POTTIER, 1992; CROFT, 1994, 1998; LANGACKER, 1987, 1991; DELANCEY, 1987; TALMY, 1988, 2000; LAKOFF, 1977), a codificação da função não-atribuição de causalidade na Crônica Geral da Espanha de 1344, prosa histórica representante do português arcaico, editada por Cintra (1951). Nessa análise, busca respostas para os seguintes problemas: (a) dos recursos verbais disponíveis, no corpus analisado, para a expressão da não-atribuição de causalidade, qual a mais e a menos prototípica? (b) que traços semântico-pragmáticos e formais caracterizam as diferentes estratégias? A hipótese central considera que as construções analisadas codificam a não-atribuição da causalidade em variados graus, conforme fatores ligados à noção de causalidade e à transitividade, como a perspectivação e a saliência. Foram coletadas 1061 ocorrências e analisadas quanto a fatores pragmáticos, como a topicalidade e o estatuto informacional; fatores semânticos, como saliência cognitiva do Afetado e do Causativo, traço [animado], tipo de afetação; e fatores formais, como a ordem e a expressão. Além disso, todas foram graduadas quanto ao grau de transitividade e de relevo discursivo. Os resultados apontaram que a média é a mais prototípica das construções de não-atribuição de causalidade, por ser a que apresenta causativo menos saliente e menor grau de transitividade. A passiva, mais freqüente no corpus e, muitas vezes, com Causativo omitido, caracterizou-se como menos prototípica por seu Causativo ser freqüentemente evocado e apresentar maior grau de transitividade. Do mesmo modo, a impessoal, cujo causativo é freqüentemente inferível, manifestou-se como mais causativa e mais transitiva que a média, todavia menos que a passiva. A análise demonstrou, enfim, que o fenômeno da não-atribuição de causalidade não é um fenômeno isolado, limitado à manifestação ou não de um Agente. Submete-se a graus e acha-se ligado, especialmente, ao contorno têmporo-aspectual do evento. Devido à noção experiencial de causalidade, ligada à idéia de movimento, eventos perfectivos são mais causais que eventos imperfectivos

Orientador: Suzi Frankl Sperber / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Estudos da Linguagem / Made available in DSpace on 2018-08-21T02:25:16Z (GMT). No. of bitstreams: 1 Santi_Julio_M.pdf: 2956974 bytes, checksum: 33010fe8b1c0be1edcf0254f1cabbdd2 (MD5) Previous issue date: 2012 / Resumo: Análise de seis contos do livro "Primeiras Estórias", de Guimarães Rosa. O ponto de partida da análise foi uma estrutura linguística chamada voz média, uma espécie de síntese entre a voz ativa e voz passiva. Com ponto de partida linguístico, o trabalho se inspira na obra Was Geht Uns Noah An, de Wolfgang Von Schöfer, e tenta resgatar significados de palavras que se perderam ao longo do tempo. Assim, encontra-se um ponto de diálogo com a obra de Suzi Frankl Sperber, que defende que Guimarães Rosa utiliza um recurso que a autora chamou de abertura do sintagma e também o de fundir espaço e personagem na narrativa com o objetivo de potencializar, ampliar os sentidos. A presente dissertação explora as dimensões simbólicas da obra de Rosa, mostrando de que maneira o autor se abria e investigava os mistérios da vida. Foram levantados alguns temas que se repetem em todos os contos, e a partir de uma análise minuciosa sobre os contos, encontra-se um tema principal que é desenvolvido: a importância de viver é compreendida através do aprimoramento da alma - e isso se dá através do conhecimento não de ordem racional, mas prática / Abstract: Analysis of six shorts stories from the book "Primeiras Estórias", written by João Guimarães Rosa. The analysis' starting point was a linguistic structure called "Middle Voice", some kind of synthesis between active and passive voice. This work is inspired in the book from Wolfgang Von Schöfer, Was Geht Uns Noah An, and tries to retrieve some word's meanings that have got lost through time. Therefore, this work was able to dialogue with the work from Suzi Frankl Sperber, who defends that Rosa uses a technique that Suzi has called "syntagm 's opening", and also an important structural characteristic: Rosa merges the space and the character in the narrative, in order to amplify the possible meanings and interpretations. This current dissertation explores the symbolic dimension of six shorts stories from Primeiras Estórias and tries to analyze how the author explored the meaning and mysteries of life. Some subjects were noticed, and after a detailed analysis of each short story, a main subject is found: the importance of living is understood as the development of the soul - and this happens only through the acquisition of practical knowledge / Mestrado / Literatura Brasileira / Mestre em Teoria e Critica Literaria