Poetry Translating as Expert Action Processes, priorities and networks

Chapter 5. Five translators translate 

0

20

40

60

80

100

120

140

160

180

Minutes

Fleur Geoff Hugo Irene Francis

Figure 28. Toen wij: Whole-project translating time per translator

5.3.2 Drafts and versions

This implies that it is worth examining how translators distribute their effort across

drafts – firstly in terms of time, and then in terms of what they do in each draft.

5.3.2.1 Time management

Though the average time spent on each draft differs (Draft 1, 46m; Draft 2, 59m;

Draft 3, 30m), these differences are not significant because they are overshadowed

by large variations in how individual translators divide their time across drafts^7 :

see Figure 29. These fall into three rough patterns:

1. My Drafts 1 and 3 are relatively short, but my Draft 2 (102m) lasts over double
   the others’ average of 49m.


2. Fleur and Geoff spread their efforts relatively evenly across all three drafts.


3. Hugo and Irene put a lot of effort into Draft 1 but gradually tail off into a very
   brief Draft 3.

Statistically, Pattern 1 differs significantly from the other two in variance terms

(i.e. low-mid-high variation)^8. The difference between Patterns 2 and 3 could not

be statistically tested, however, and therefore remains tentative.

In the following two sub-sections, I first qualitatively explore how translators’

priorities and processes develop across a poem’s translating lifetime. This then al-

lows me to examine how far the three time-management patterns just identified

might reflect chance factors or differences in translator persona.

7. Repeated-measures ANOVA (tape-units per translator) F 2.5, p 0.15.


8. Levene’s F 2.7, p 0.11, not significant (Fleur, Hugo, Geoff, Irene); F 4.4, p 0.03 (all transla-
   tors), significant.