Segmentações não convencionais no EF II: há regularidades linguísticas em dados frequentes?
Carregando...
Data
2023-03-02
Autores
Borsato, Ana Carolina Teodoro
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Nesta dissertação, analisamos as chamadas segmentações não convencionais de palavra, que podem ser de três tipos: as hipossegmentações, que ocorrem quando estão ausentes as fronteiras gráficas entre palavras, como em “porisso” e “oque”; as hipersegmentações, quando estão presentes fronteiras gráficas dentro de uma palavra ortográfica, como em “que ria” e “morava-mos”; e os híbridos, que se caracterizam pela ocorrência simultânea de dois tipos de segmentação não convencional de palavras (hipossegmentação e hipersegmentação) em uma sequência de duas palavras, como em “po risso” e “dere pente”. As grafias aqui analisadas foram identificadas em um conjunto de textos escritos por alunos do Ensino Fundamental II (EF II), categorizados, de uma perspectiva estatística, como outliers, segundo Fiel (2018), por produzirem uma alta quantidade de hipossegmentações em relação aos demais sujeitos do Banco de Dados de Escrita do EF II. Nesta dissertação, selecionamos para análise as segmentações não convencionais desses alunos, as quais não foram analisadas por Fiel (2018), com o objetivo central de caracterizar essas grafias quanto aos aspectos prosódicos dos enunciados falados. Para alcançar esse objetivo, nos baseamos na Fonologia Prosódica, de Nespor e Vogel (1996), que concebe a existência de sete constituintes prosódicos que estruturam os enunciados das línguas do mundo, e na abordagem da escrita como constituída de modo heterogêneo, proposta por Corrêa (2004). Dos resultados obtidos na análise, constatamos que informações linguísticas de natureza prosódica e da própria convenção ortográfica estão envolvidas nas ocorrência de um limite não convencional, o que corrobora estudos anteriores sobre o mesmo tema. Avançamos ao mostrar que as características das grafias dos sujeitos denominados outliers vão ao encontro do que é descrito para os demais sujeitos do Banco de Dados de Escrita do EF II. Identificamos um sujeito que, apesar de se diferenciar das tendências gerais encontradas, possui características dos primeiros anos de escolarização ao apresentar, em algumas circunstâncias, recursos gráficos que demarcam palavras como formas de destacar sentidos privilegiados do texto e de informações linguísticas que permeiam as regularidades observadas. Em suma, os dados de segmentação não convencional de palavra fornecem evidências gerais e particulares a respeito da noção de palavra escrita mobilizada pelos escreventes, evidenciando as marcas de trânsito do escrevente por práticas de uso da linguagem, oral/falada e letrada/escrita, em que se inserem (CORRÊA, 2004).
In this dissertation, we analyze unconventional word segmentations, which can be of three types: hyposegmentations, defined by the lack of boundaries between words, such as in “porisso” and “oque”; hypersegmentations, which are characterized by the presence of boundaries within a word, as in “que ria” and “morava-mos”; and hybrids, which are characterized by the simultaneous occurrence of two types of unconventional word segmentation in a sequence of two words (hypossegmentation and hypersegmentation), as in “por risso” and “dere pente”. The unconventional word segmentations analyzed were identified in texts written by students from Junior High School (EF II), in Brazil, that were categorized, from a statistical perspective, as outliers, according to Fiel (2018), because they produce a high amount of hyposegmentations in comparison to the other subjects of the EF II Writing Database. In this work, we selected for analysis the unconventional segmentations of these students, which were not analyzed by Fiel (2018), aiming to characterize these datas in terms of the prosodic aspects of the utterances. To achieve this goal, we based ourselves on Prosodic Phonology, by Nespor and Vogel (1996), which conceives the existence of seven prosodic constituents that structure the utterances of the languages of the world, and on the approach of writing as heterogeneity, proposed by Corrêa (2004). The main results of this analysis we found are that linguistic information of a prosodic nature and the orthographic convention were important for the occurrence of spaces between syllables in a certain position of the word, which corroborates previous studies on the same subject. We advanced by showing that the characteristics of the datas of the subjects called outliers are in line with what is described for the other subjects in the EF II corpus. We identified a subject who, despite differing from the general trends found, continues to mobilize hypotheses characteristic of the first years of schooling by making use, in some circumstances, of graphic resources that demarcate words as ways of highlighting privileged meanings of the text and linguistic information that permeate the observed regularities. In short, the unconventional word segmentation data provide general and particular clues regarding the notion of the written word mobilized by the writers, evidencing the marks of the transit of the writer through practices of language use, oral/spoken and literacy/written, in that are inserted (CORRÊA, 2004).
In this dissertation, we analyze unconventional word segmentations, which can be of three types: hyposegmentations, defined by the lack of boundaries between words, such as in “porisso” and “oque”; hypersegmentations, which are characterized by the presence of boundaries within a word, as in “que ria” and “morava-mos”; and hybrids, which are characterized by the simultaneous occurrence of two types of unconventional word segmentation in a sequence of two words (hypossegmentation and hypersegmentation), as in “por risso” and “dere pente”. The unconventional word segmentations analyzed were identified in texts written by students from Junior High School (EF II), in Brazil, that were categorized, from a statistical perspective, as outliers, according to Fiel (2018), because they produce a high amount of hyposegmentations in comparison to the other subjects of the EF II Writing Database. In this work, we selected for analysis the unconventional segmentations of these students, which were not analyzed by Fiel (2018), aiming to characterize these datas in terms of the prosodic aspects of the utterances. To achieve this goal, we based ourselves on Prosodic Phonology, by Nespor and Vogel (1996), which conceives the existence of seven prosodic constituents that structure the utterances of the languages of the world, and on the approach of writing as heterogeneity, proposed by Corrêa (2004). The main results of this analysis we found are that linguistic information of a prosodic nature and the orthographic convention were important for the occurrence of spaces between syllables in a certain position of the word, which corroborates previous studies on the same subject. We advanced by showing that the characteristics of the datas of the subjects called outliers are in line with what is described for the other subjects in the EF II corpus. We identified a subject who, despite differing from the general trends found, continues to mobilize hypotheses characteristic of the first years of schooling by making use, in some circumstances, of graphic resources that demarcate words as ways of highlighting privileged meanings of the text and linguistic information that permeate the observed regularities. In short, the unconventional word segmentation data provide general and particular clues regarding the notion of the written word mobilized by the writers, evidencing the marks of the transit of the writer through practices of language use, oral/spoken and literacy/written, in that are inserted (CORRÊA, 2004).
Descrição
Palavras-chave
Linguística, Ensino Fundamental, Segmentações, Palavra (Linguística), Escrita, Segmentation, Unconventional, Word, Prosody, Orthography, Orality, Literacy