Poetry Translating as Expert Action
Bosnians, whatever their religion, are descendants of medieval heretics, united in
their resistance to persecution, are not present in the semantics of the extract.
Nevertheless, they are powerful cognitive implications and emotional reactions
that readers may derive from it.
As mentioned with Yù jiē yuàn, however, such intrinsic features and functions
are found in many poems, but none is unique to poetry. This implies that when we
recognize a text as a poem, we also rely on extrinsic form – that is, on features
around the poem’s text. Crucial here are genre-specific “framing signals” that, by
convention, tell the reader ‘this is poetry’ (Andrews 1991: 18; cf. Stockwell’s “e d g e-
work” – 2002: 80). Examples might be a book cover with the word Poems, or – very
typically, as in the Dizdar quote – short left-aligned lines on the page.
Genres, however, are almost certainly judged holistically, in terms of all three
aspects. Hence extrinsic form will offer audiences a social agreement that the text
they are reading or hearing is a poem. The more the text shows the intrinsic form
and functions expected within their literary community, the more readily audi-
ences will accept that agreement.2.4.2 Translations
Translations were defined earlier as having some “appropriate relation of relevant
similarity” to an other-language source (Chesterman 1997: 69). The examples
above show that what I felt to be appropriate with the Dizdar quote was, as far as
possible, to recreate in English the source poem’s semantic content, plus its intrin-
sic-poetic/stylistic form. Where both were not possible, I prioritized the latter.
Recreating, however, is not point-for-point remapping. Sometimes the source and
target poem use different effects, and sometimes the same effect in different places
(‘compensation strategies’: Harvey 2000). Thus I translated Crknut će taj pas pseći/
Od samih/Jada (‘Die will that dog canine/Of very/Miseries’: Lines 7–9), as This
cursed cur/Will be slain/With pain – transferring the pejorative nuance of Crknut
(‘die’) to ‘dog’ (cur), and adding cursed to convey the alliteration of pas pseći.
My attention to intrinsic form and style shows a wider relevant similarity: that
both source and target text should be poems. Though very common, this is not
universal: prose translations may serve as guides to source poems or, especially
with epic verse, to be read in their own right (as advocated by e.g. Dacier 1699/2006
and Goethe 1811–1814/2006; cf. Lefevere 1975, Boase-Beier 2009: 194). However,
from now on this book assumes that poetry translators aim to create a target text
that has the attributes of a poem.
Boase-Beier also points out (personal communication) that when translators
rewrite a poem, they are actually writing their experience and interpretation of it:
hence translations always contain something of the translator’s own subjectivity.