Abstract
The idea that text in a particular field of discourse is organized into lexical patterns, which can be visualized as networks of words that collocate with each other, was originally proposed by Phillips (1983). This idea has important theoretical implications for our understanding of the relationship between the lexis and the text and (ultimately) between the text and the discourse community/the mind of the speaker. Although the approaches to date have offered different possibilities for constructing collocation networks, we argue that they have not yet successfully operationalized some of the desired features of such networks. In this study, we revisit the concept of collocation networks and introduce GraphColl, a new tool developed by the authors that builds collocation networks from user-defined corpora. In a case study using data from McEnery’s (2006a) study of the Society for the Reformation of Manners Corpus (SRMC), we demonstrate that collocation networks provide important insights into meaning relationships in language.
Highlights
Because GraphColl does not limit the user’s choice of the association measure, in this study we explore the properties of five different statistics and their possible contributions to discourse analysis — including MI3 (which was McEnery’s (2006a) original preferred choice)
The replication of McEnery’s (2006a) study is presented using the MI2 statistic followed by new results using MI3, log-likelihood, Delta P and Cohen’s d
Two other groups of collocates can be observed: (i) collocates with general negative associations such as dismal, drinking, false, contemptuous, abominable, wantonness, lying and negligent; and (ii) descriptive collocates such as conversation, effects, land, examin’d, causes, essay, civility, engagement, caution and act. The former create an additional layer of general pejorative evaluations and associations, as concretely exemplified by the chapter title “The dismal Effects of prophane SWEARING” (Walker 1711). The latter set consists of other key terms contributing to the general shaping of the discourse around the nature of swearing, its causes and effects, and its legal consequences
Summary
Three criteria for identifying collocations have been proposed These are: (i) distance, (ii) frequency, and (iii) exclusivity. The distance specifies the span around a node word (the word we are interested in) where we look for collocates. This span is called the ‘collocation window’. Love is much more strongly and exclusively connected with the noun affair; when the word affair appears in text, there is a large probability that the preceding word is love. In addition to the three criteria discussed above, Gries (2013) points out three other criteria that should be considered: (iv) directionality, (v) dispersion and (vi) type-token distribution among collocates
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
