FREEDOM OF COMBINATION AND HETEROGENEITY: A CORPUS LINGUIST’S LOOK AT TWO SAUSSUREAN INSIGHTS

Tony Berber Sardinha

Authors

Tony Berber Sardinha PUC-SP

Keywords:

Saussure, corpus linguistics, collocation, freedom of combination, heterogeneity.

Abstract

This article offers a reexamination of two of Saussure’s insights from the point of view of corpus linguistics—namely, freedom of combination and heterogeneity in language in use. Regarding the first insight, an analysis of word combinations in a corpus of newspaper texts written in Brazilian Portuguese was carried out to determine how many of these combinations were actual collocations—that is, were used frequently enough in a very large reference corpus (the Brazilian corpus) to warrant statistical significance. The results suggested that most word combinations are not free; rather, they follow previously established preferences among speakers. Regarding the second notion, that of heterogeneity, the collocations in the newspaper texts were tracked as they were deployed one after the other along each text, and this flow was visually depicted. The inspection of the charts revealed unique patterns of the distribution of collocation, thereby suggesting that the evidence supports the view of heterogeneity. A cluster analysis was later conducted on the amount of collocations in each text, revealing three basic collocation bands onto which all the texts can be fitted. This was interpreted as suggesting that heterogeneity, despite being present and noticeable, is constrained rather than limitless. The article concludes that the methods and techniques afforded by present-day corpus linguistics can shed light onto Saussure’s many valuable insights.

---

Original in English

Author Biography

Tony Berber Sardinha, PUC-SP

Professor associado no Programa de Linguística e Linguística Aplicada na Pontifícia Universidade Católica de São Paulo (PUCSP). Entre suas publicações, que incluem inúmeros artigos e livros sobre tópicos como linguística de corpus, metáfora e ensino de línguas, podem ser citados “Linguística de Corpus” (2004), “Metáfora” (2007), “Caminhos da Linguística de Corpus” (2012; com Tania Shepherd e Marcia Veirano), “Tecnologias e mídias no ensino-aprendizagem de inglês” (2013; com Tania Shepherd, Denise Delegá, e Telma São Bento Ferreira), “Working with Portuguese Corpora” (2014, com Telma São Bento Ferreira), “Multidimensional Analysis 25 years on” (2014, com Marcia Veirano), e “Metaphor in Specialist Discourse” (com Berenik e Hermmann). Tony Berber Sardinha está no Comitê Consultivo de vários periódicos sobre linguística e linguística aplicada, tais como The International Journal of Corpus Linguistics, Corpora, Metaphor and the Social World, Journal of Research Design and Statistics in Linguistics and Communication Science, DELTA, Revista Brasileira de Linguística Aplicada, Fórum Linguístico, Revista de Estudos da Linguagem, e O ESPecialist, além de séries de livros como Language, Cognition, and Communication, e Espaços da Linguística de Corpus. Entre seus projetos de pesquisa estão ‘Dimensões de Variação de Registro do Português Brasileiro’ (Fapesp; CNPq), ‘Linguística de Corpus: Padronização e Variação’ (CNPq), ‘Dicionário de Colocações do Português Brasileiro’ (CNPq), e ‘Dimensions of register variation on American television’.