Relações múltiplas entre oralidade e escrita: vogais médias e róticos

• Ano de defesa: 2023
• Autor(a) principal: Cecília Valle Souza Toledo
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; FALE - FACULDADE DE LETRAS; Programa de Pós-Graduação em Estudos Linguísticos; UFMG
• Palavras-chave em Português: Oralidade; Escrita; Esquemas múltiplos; Erros ortográficos; Variação; Língua portuguesa – Fonologia; Língua portuguesa – Ortografia e silabação; Língua portuguesa – Variação – Belo Horizonte (MG); Língua portuguesa – Vogais; Língua portuguesa – Português falado – Belo Horizonte (MG)
• Link de acesso: http://hdl.handle.net/1843/51074; https://orcid.org/0000-0001-7997-3064

This thesis aims to investigate the multiple links established between speech and spelling and between spelling and speech for the Brazilian Portuguese spoken in Belo Horizonte. Four linguistic categories in Brazilian Portuguese are assessed: (1) pretonic mid vowels, (2) final posttonic mid vowels, (3) final rhotics in verbs, (4) final rhotics in nouns. The analyzed linguistic categories are traditionally spelt with one of the following letters: <e, o, r>. Additionally, all categories are prone to variable phonological phenomena which may motivate spelling errors during the school years (ALVARENGA et al., 1989; NÓBREGA, 2013). The final pretonic and posttonic mid vowels may be pronounced as high vowels due to vowel raising and reduction (VIEGAS, 1995; DIAS; SEARA, 2013). These phonological phenomena tend to motivate spelling errors in which <e, o> are replaced by <i, u>. Ex: <*minino> ‘boy’, <*patu> ‘duck’. Final rhotics in verbs and nouns may be lost (OLIVEIRA, 1997; HUBACK, 2003). This phenomenon tends to trigger misspelled forms that lack a final <r>. Ex: <*cantá> ‘to sing’, <*celulá> ‘cellphone’. Considering the apparent similarities between categories (1) and (2) and (3) and (4), we seek to answer the following research question: do apparently similar linguistic categories manifest analogous or different behavior in speech and in spelling patterns of elementary and middle school students? Four hypotheses were tested: (1a) the replacement of <e, o> for <i, u> persists in spelling, at significantly different rates, across different phases of schooling, depending on the syllabic position: pretonic or final posttonic; (1b) the loss of letter <r> in spelling persists, at significantly different rates, across different phases of schooling, depending on word class: verb or noun; (2a) the increase in production of mid vowels in speech across the school years depends on the syllabic position: pretonic or final posttonic; (2b) the increase in the production of the final rhotic in speech across the school years depends on word class: verb or noun. We predicted differences in the children's speech and spelling depending on the linguistic categories. The hypotheses were tested on a set of spoken and written data from fifty children from the 1st, 3rd, 5th, 7th and 9th years of elementary and middle school classes in the city of Belo Horizonte, Minas Gerais, Brazil. Results confirmed the hypotheses that spelling errors related to posttonic and final pretonic vowels as well as to final rhotics in verbs and nouns tend to persist at different rates across different stages of schooling. The results evidence that learning to spell affects speech production. However, the effects of spelling in speech depend on the linguistic category. Results are discussed in light of Exemplar Theory (JOHNSON, 1997; BYBEE, 2001; PIERREHUMBERT, 2001; FOULKES; DOCHERTY, 2006; PORT, 2007) and Complex Dynamic Systems Theory (THELEN; SMITH, 1994, 2003; ELLIS; LARSEN -FREEMAN, 2009), which highlight that elementary and middle school aged children manage and reorganize multiple schemas during schooling. The term 'multiple schemas' is adopted to refer to the probabilistic patterns that emerge from multiple networks of phonetic, phonological, lexical, morphological, social connections etc. These relations can generate new grammatical patterns (LANGACKER, 2000; TOMASELLO, 2000; BYBEE, 2001; CRISTÓFARO-SILVA; GOMES, 2004). We argue that students access multiple schemas during schooling, which motivates the different spelling learning stages. We also argue that the learned orthographic knowledge motivates the reorganization of the language system. Such reorganization can be more or less evident, depending on the linguistic category. In summary, we intend to defend the thesis that letters are discrete symbols associated with multiple schemas, which emerge from the complex interaction between words and phonetic, morphological and orthographic patterns.