Menzerath's law analysis in Brazilian Portuguese

Abstract

Under the perspective of Quantitative Linguistics, this paper revisits the Menzerath's Law, applying it to data from Brazilian Portuguese, using the following unities of analysis: words, syllables and phonemes. The data was extracted from the ABG Corpus. Statistical analyses are performed on the proposed models, corroborating the existence of a decay relationship between the mean length of words (in syllables) and the average length of syllables (in phonemes); what corroborates the Menzerath Law. It is noticed that better measures or variables not considered in the model might be used to improve it.