Robert Frost’s often-quoted statement occurs as part of a discussion of‘‘Stopping by Woods on a Snowy Evening.’’ Frost says: ‘‘You’ve often heard me say - perhaps too often - that poetry is what is lost in translation. It is also what is lost in interpretation. That little poem means just what it says and it says what it means, nothing less but nothing more’’ (Untermeyer 1964: 18). Yet despite Frost’s assertion that meaning is straightforward, it is never a simple matter to determine ‘‘what a poem means.’’ Frost may claim that ‘‘Stopping By Woods on a Snowy Evening’’ in fact ‘‘means just what it says and... [says] what it means,’’ but this leaves unaddressed the question of how it means, of what might be the relationship between the poem’s form, sound, and vocabulary and its meaning. Surely Frost would agree that the poem’s formal elements are crucial, since ‘‘poetry is what is lost in translation’’; and yet even for native speakers of a poem’s language, the exact relationship between form and content is deeply problematic. The critics whose work Frost was deprecating were native speakers of his own language, reading a poem in that shared common language, within a few years of the poem’s creation. How much more impenetrable is the exact relationship between sound and sense when we are grappling with a poem written in another language, another era, and a radically different culture?
As Holmes says, poetic translations ‘‘differ from all other interpretative forms in that they also have the aim of being acts of poetry’’; he suggests the term metapoem ‘‘for this specific literary form, with its double purpose as metaliterature and as primary literature’’ (Holmes 1988: 24, building on Barthes’s terminology of‘‘metalanguage’’). Concerning the question of what form a metapoem should take, Holmes says that ‘‘no other problem of translation... has generated so much heat, and so little light, among the normative critics’’ (1988: 25). Different critics argue passionately for using equivalent verse-forms, for translating into native meters, and for translating into prose, and each of these options has arguments both for and against it.
Every translator of poetry must decide which of two goals is most important for the metapoem: to write poetry that succeeds as poetry in the target language, or to convey the tone, structure, and overall ‘‘feel’’ of the source text for those who cannot read the original language. Ideally, the translator would bridge this gap and perform both types of translation simultaneously; but while it is a rare translator who aims for one of these goals to the exclusion of the other, in practice it is seldom if ever possible to give equal prominence to both.
The question of which to privilege, the poetry of the target language or of the source text, affects every decision about vocabulary or form that the translator makes. Not only word choice but word placement, patterns, and sound are crucial in poetry, so much so that one could say that poetry consists in the deliberate arranging of words in certain patterns; the sound to a large extent makes the poetry. Since the sound of a poem depends not only on its words but also on its meter and structure, the translator has to decide from the outset whether and to what extent the original’s form is crucial. Should the translator attempt to reproduce the meters of the original? Should she attempt to mimic the source text’s sound, the interplay of alliteration, assonance, the use of particularly evocative words? Does a version that addresses only the lexical content of the original and gives no hint of its sound in fact translate the original at all? Consider, for example, Shakespeare’s line ‘‘O, she doth teach the torches to burn bright’’ (Romeo and Juliet I. v). If we paraphrase ‘‘O, she shows the lights how to glow brightly,’’ does the poetry remain? Or is it precisely the sound of the line that makes it beautiful, memorable, and moving - in short, that makes it poetry: the repetition of the sounds t, ch, b, and r in the words teach/torches/burn/bright, and in those same words the chiastic arrangement of long and short vowels with the short vowels followed by an r (teach, torch, burn, bright), the preponderance of monosyllables, the regular iambic pentameter? In this line, is the sound not far more important than the content; at the very least, is not the sound a primary means by which the sense is enhanced, and made to seem profound, to an extent that mere paraphrase of content cannot convey?
If sound trumps content, then we can understand the extreme view which claims that the only method for conveying poetry in translation is to ‘‘hear the sounds of the original while reading literal renditions’’ of the content (Baker 1998: 173, discussing Burnshaw 1960). On this view, the translation of poetry is impossible, and all that can honestly be offered a reader who does not know the source language is a paraphrase of the poem’s basic content into prose that aims for the nearest possible lexical fidelity, while the rest of the poem - sound, meter, word placement, and so on - should be left to ‘‘speak for itself.’’ Most translators would not agree with this assessment, which seems close to an argument of despair (and which, in fact, leaves unanswered many questions, not least among them what is meant by ‘‘lexical fidelity’’), but different translators and translation-study theorists disagree about the degree to which any other approach to translating poetry can be successful.4
All of these issues confront the translator of any poetry, even when the source is one modern language and the target another, closely related modern language. But these difficulties become even more formidable when the translator is working not only across languages but across many centuries as well; in Hardwick’s formulation, ‘‘translating words also involves translating or transplanting into the receiving culture the cultural framework within which an ancient text is embedded’’ (Hardwick 2000: 22). For the ‘‘cultural translation’’ of Catullus or of any Latin verse, issues of meter and of vocabulary are particularly important.
World History
