Acentuação gráfica no português brasileiro: desafios para a escrita infantil

Abstract

This dissertation investigates the graphic accentuation of words in Brazilian Portuguese in essays of children and pre-adolescents aged 6 to 12 years from the city of Belo Horizonte/MG, using the principles of Corpus Linguistics (BERBER SARDINHA, 2003; BERNARDINI, S; STEWART, D; ZANETTIN, F, 2003). Most Portuguese words are not accentuated graphically. There is preference for non-accentuation in twocases: 1) when the word is paroxytone and ends in a vowel: casa ['kasa] (house), 2) when the word is oxytone and ends in a consonant: melhor [meh] (better) or in a semivowel: papai [pa'pa] (father). On the other hand, when one word does not follow these preferences, the graphic accent occurs (cf. QUEDNAU e COLLISCHONN 2006). Thus, the graphic accent occurs when the tonicity of words deviates from the preferred pattern. The acute accent (´) denotes open stressed vowel, the circumflex (^) closed stressed vowel. Starting from the studies of Goncalves Viana in Ortografia Nacional from 1904 about orthography, the graphic accentuation in Portuguese is based on these notions. Thus, the graphic accentuation is applied to four principles: 1) to indicate that a syllable is stressed: e [' ] (is) e [i] (and); 2) to indicate the position of the stressed syllable: cará [ka'ra] (yam) cara ['kara] (face); 3) to indicate the opening or closing of vowel sound: cipo [si'p] (liana) capo [ka'po] (hood), leu [l] (aimlessly) leu ['le] (read); 4) to differentiate meanings in confusing words: por ['poh] (put) por ['poh] (by). These principles result in rules of graphic accentuation that are provided by spelling reforms and orthographic agreements that seek to organize, regulate and prescribe the uses of the spelling of Portuguese. This research analyzed a corpus of 62,673 words coming from essays of 7,892 students from schools located in Belo Horizonte/MG which were surveyed in the following relationships: graphically accented words versus correctness, omission, change of diacritical and syllable exchange. The results showed that: (1) most Portuguese words have no graphic accent. Among the graphically accented words (different types), most are proparoxytones, with oxytones coming second and paroxytones being fewer in number; (2) among all graphically accented words written by children in their compositions, more oxytones were attested, followed by proparoxytones and then by paraxytones, which were found in smaller numbers; (3) the rate of correct spelling in graphically accented words is relatively high in the following order: more oxytone words, followed by paraxytones and then proparoxytones at lower rates; 4) the main misspelling committed in words graphically marked for stress is the omission of the diacritic in the following order: more oxytone words, followed by paroxytones and then proparoxytones; (5) the use of a wrong diactritic and the use of graphic accent in the wrong syllable are errors that have low rates; (6) A small number of rules for graphic accentuation is enough for correctlymarking a large number of the words found in the analyzed corpus: a) three rules for oxytones; b) two rules for paroxytones and c) two rules for proparoxytones. These rules are responsible for the graphical marking of 90% of all the words in the corpus; (7) the new rules of Orthographic Agreement of 1990 represent only 0.1% of the entire corpus investigated.