Processos verbais em artigos científicos: uma análise com base na língua em uso

Abstract

This research aimed to develop a list of academic verbal lexemes, according to the vocabulary division proposed by Nation (2001), and qualitatively analyze the most frequent academic verbal processes, from a Usage-Based Linguistics perspective, especially Halliday and Matthiessen (2004). Thus, a corpus of research articles in all evaluation areas of CAPES was compiled in order to observe the existence of differences in lexical choices of each subcorpora and genre characteristics, as well as differences between uses of most frequent verbal processes. As a result, we presented two lists, one of 281 most frequent verbal lexemes in research articles, and one of 156 most common academic verbs, based on comparison with representative corpora of non-academic language. These lists confirm that some highly frequent words in the Portuguese, such as "olhar", "dizer" and "achar", had very few occurrences in our corpus, showing that academic vocabulary avoids high frequency words (LINDQVIST, 2014). As an alternative to the few tokens of "dizer", we analyzed qualitatively six most frequent verbal processes, of which half were prototypically verbal processes, "destacar", "sugerir" and "descrever" while the other part, "demonstrar", "determinar" and "estabelecer", were frequently associated with other types of processes, such as material processes. Despite this difference, all verbal processes were used in a similar manner over the analyzed passages, with predominance of tokens in simple present tense, which have the ability to bring an action from the past to the readers present, causing a timelessness effect (FUZER, 2012), or omitting the subject of a sentence, which tends to distance the author from its study. This same purpose has been achieved by using passive voice, in order to omit the authors participation or to highlight the message from verbal processes. Such mechanisms contribute to making the text more objective and its discourse impersonal. Even though Sayer is directly involved participant on verbal processes, it was often not explicit, and needed to be inferred from context. However, in all analyzed cases, there was a message conveyed either as a Verbiage or as projected sentence. Overall, the results allowed us to notice that there were differences between each discipline, especially in preferred academic words. Nevertheless, these characteristics did not interfere with our data quantitatively, being only observed when compared to preferences from each subcorpus ranking.