Chapter 2. Poetry in a political preface
watchman simply represent anyone who tries to wipe out those who think differ-
ently? Or all of these? Moreover, readers’ world knowledge may differ from that of
the poet. Hence they may receive signals unintended by the poet, which are gener-
ated by the context of reading. Dizdar, for example wrote Kameni spavač in the
1960s-1970s. But someone reading the extract in 1994, for instance, might inter-
pret the watchman as foreshadowing Serbian ethnonationalist aggression in the
1992–1995 war (Hanneke Jones, personal communication). For reasons like these,
some scholars see poems as offering readers less a fixed meaning-structure, and
more a set of “interpretive potentials” – clues which different readers can interpret
in different and sometimes multiple ways (Kwan-Terry 1992: 213, after Isere
Stockwell 2002: 75–80).
When readers are translators, the knowledges they call on and build while
reading may also vary depending on their reading purpose. This purpose may be
to select which poems or even which poets to translate (Bishop 2000: 61). Just
before translating, the reading purpose may be to familiarize oneself with a source
poem, or identify its key features and problems (Peraldi 1978; Jones 1989; Flynn
2004: 276); and while producing a first translated version, to get source-text input.
As translating goes on, translators also read in order to check rough target-lan-
guage output, often in parallel with reading the source poem: thus, since I had al-
ready drafted the Dizdar poem, during E’s project I was reading in order to check
the version’s semantic fit with the source plus its effectiveness as English poetry.
When a translator rewrites a poem in another language, the microstructures
and/or macrostructures constructed while reading the source give input into the
translated version’s microstructures. The latter are also shaped by the translator’s
linguistic, genre and world knowledge relevant to the target language, culture and
audience, and by the translated version’s own developing macrostructure.
Here, Peraldi argues that there is no clear separation between source-poem
reading and target-version writing. He sees poetry translating as consisting of a
series of repeated “lectécritures” (1978: 109–110, after Kristeva): readings (lectures)
for the purpose of writing (écriture), combined with writing one’s readings. More-
over, the word “repeated” appears crucial. Translators’ reports indicate that multi-
ple drafting sessions (here labelled ‘drafts’ for short) which produce multiple trans-
lated versions are typical of poetry translating (Bly 1983; Born 1993; Shih 2006)^4.
Though, 16 years later, I cannot remember how far my reading and writing were
separate or combined, I still have some of the many versions of Stone Sleeper’s
poems I made over years of redrafting.- In this book, I distinguish between ‘draft’ as a working session, and ‘version’ as a provi-
sional or final target-language rendering of a source poem.