Abstract
The hallmark of Network Text Analysis (NTA) is the creation of semantic networks or concept maps from linguistic data. Its key insight—one borrowed from studies in Social Network Analysis—is that the position of concepts within such networks reveals vital information about the meaning of the text as a whole. A second hallmark of NTA is that the structure and size of a network are independent of the frequency of the words of which it is comprised. In this paper we demonstrate the application of NTA to Joseph Conrad’s Heart of Darkness (HoD). Using morphological and etymological relationships as the basis for the network, we first represent HoD as a network consisting of over 385 nodes. We then compare and contrast the results of our network analysis with those reported in a widely-cited corpus stylistic analysis of HoD. While our results are remarkably consistent with and complementary to results in that study, we also report results not identified in that analysis, results which could only have been identified using NTA.
Highlights
Despite intense and vocal criticism (Fish 1973, 1979; MacKay 1996, 1999), corpus-assisted evaluations of literary texts are becoming increasingly objective and systematic (e.g. Stubbs, 2001, 20005; O’Halloran, 2007; Fischer-Starcke 2009; Biber, 2011)
There is one widely-used and highly relevant set of methods that has been largely overlooked by corpus stylistics studies—Network Text Analysis (NTA)
When those things are working and function, they could be said to stand up. Broken, or useless, they could be said to have fallen into disrepair. None of these broad themes described in the last section are evident anywhere in Stubbs’ corpus stylistic analysis of Heart of Darkness
Summary
Despite intense and vocal criticism (Fish 1973, 1979; MacKay 1996, 1999), corpus-assisted evaluations of literary texts are becoming increasingly objective and systematic (e.g. Stubbs, 2001, 20005; O’Halloran, 2007; Fischer-Starcke 2009; Biber, 2011). In this paper we demonstrate the application of NTA to Joseph Conrad’s (1899/2010) Heart of Darkness (HoD). We compare and contrast the results of our network analysis with those reported in Conrad in the Computer (Stubbs 2005). In that paper Stubbs used traditional stylistics methods, i.e. the analysis word and phrase frequencies, collocation, distribution, etc. On the whole our results are remarkably consistent with and complementary to his Many of his high “keyness” words—the most statistically over-represented words within the text—are centrally positioned within our network representation of HoD. In the third section we describe our application of NTA to Conrad’s Heart of Darkness along with the results of that analysis. In the fourth section of the paper we compare and contrast our results with those of Stubbs (2005). With a discussion of the limitations of our analysis, as well as the opportunities and challenges associated with the application of NTA to literary texts
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
