chapter 7
Conclusion
7. 1 Towards a model of poetry translating
This Chapter discusses how Chapter 2’s provisional model of poetry translating
has been enriched and modified in the light of the studies in Chapters 3–6, and
assesses its potential as a foundation for future work. The chapter first revisits the
translator’s cognitive action of translating poetry; after examining the poetry
translator’s habitus as expert, as translator and as professional, it then summarizes
findings relating to poetry translating as teamwork; subsequently, it examines the
translator’s role in extended, second- and third-order networks. Finally, the up-
dated model is situated in a wider context by discussing its implications for poetry
translation research, for translation studies, and for what poetry translators do.
A graphic summary of the updated model is shown in Figure 53. There are
three key structural differences with Figure 4’s provisional framework. Firstly, tex-
tual networks vary in size – for example, from ‘translated Bosnian poetry in US
English’ to ‘US English literary production’. However, Chapter 3 gave no grounds
for keeping Systems and Polysystems as two distinct categories: hence these are
merged into the single feature (Poly)Systems. Secondly, Chapter 3 showed how
fields vary similarly in size – for example, from ‘translators of Bosnian poetry into
English’ to ‘literary translators into English’. To reflect this, Fields has been retitled
(Sub-)Fields. Thirdly, Chapter 3 also showed that textual (Poly)Systems are defined,
produced and consumed by Interest Networks of readers and other parties. This is
reflected by their overlap in Figure 53.
Beyond the first-order working team (dark grey shading), both Figures distin-
guish between more directly and more diffusely linked groupings of people or
texts: second-order (mid-grey) and third-order networks (light grey) respectively.
In practice, however, it is hard to fit into two categories the sheer variety of extended
groupings identified in the studies: translator career webs, interest networks, vir-
tual parties (Bosnian anti-nationalists, for instance), imagined communities and
sub-communities, and so on. Instead, such extended groupings are probably better
viewed in terms of two dimensions: how many actors each involves, and how many
links between actors rely on direct person-to-person or person-to-text contact.