01 April 2005

Metaphor and Translation

John Sallis, in the question period, pointed out that 'translate' has an etymology similar to 'metaphor', and he wondered what O'Rourke would say, then, about the relationship between translation and metaphor.

O'Rourke answered the question by saying that, in translating a metaphor from one language to another, one needs to find a similar metaphor, rather than carrying over the metaphor literally. He gave as an example a friend of his who translates for the EU. Once a German delegation was debating with an Irish delegation over some proposal which would apparently waste funds in an irresponsible way. The Germans expressed this in some manner appropriate to German (O'Rourke didn't say.) The translator ignored this and used, to represent the German position to the Irish, the familiar expression, "that's throwing around money like snuff at a wake"--which won the goodwill of the Irish.

But (besides the fact that that was a simile) I took it that this missed the point of Sallis' question. He was wondering whether, e.g. simply to translate 'rouge' as 'red' is to construct a metaphor--say, on the grounds that 'red' plays a role in the English language, similar to the role that 'rouge' plays in the French language. (An idea not unlike Sellars' view of translation.)

Is translation in this ordinary sense metaphor-construction? If not, why not?


Wes DeMarco said...

I took the thrust of Sallis' question to connect it not so much with Sellars as with Derrida. The issue on the table was the place of impropriety and the incidental, which according to Fran distinguish proper analogies from metaphors. Sallis used translation, I think, to highlight the question of whether translation can ever move entirely in the realm of proprieties, or whether it isn't always inevitably shot through with incidentals and improprieties. In print, Sallis typically answers 'yes,' and suggests that these improprieties and incidentals shatter any attempt at what we might like to call ‘literal’ translation. My view is that any simple dichotomy between literal and metaphorical (or proper/improper) will skew the answer. Moreover, any sense of 'metaphor' according to which all translation is metaphorical is helplessly broad. What is interesting to me is that the *same* cognitive operations used in the articulation and figuration of language are used in the refiguration of language according to any of the classical tropes. That fact has large implications for oru understanding of language and translation.

To translate 'rouge' as 'red' may involve a host of implications and associations lost or gained. In literary contexts particularly, it matters greatly whether 'red' is associated with danger, brotherhood, the blood of fallen comrades, the socialist international or what have you. The color yellow in corn-mother societies is often laden with the meaning of life, fertility, and the sacred. Talk of the ‘yellow brick road’ in these societies would be instantly associated with the sacred way of life, represented ritually by a ceremonial path strewn with corn. All this might get lost in translation, and a metaphor might be *needed* to translate the term in a literary context. Sometimes a metaphor better captures the literal sense than what we think of as the most literal term-term translation.

Even to say 'snow is white' may involve significant incidentals, since 'snow' and 'white' in English may be used to translate different terms from a source language. We have no single English words for Inuit terms that transliterate into English as snow-in-the-air and snow-on-the-ground. Whether that matters, and whether literal paraphrase can serve depends on the kind of language and its funtion. The Dani of New Guinea have a single term ('mola,' as I recall--the example is MacIntyre's) that covers that part of the visual spectrum we would call white-yellow-orange. Is the Dani sentence we could barbarously render as "snow is '*mola*'" true or false? 'Snow is orange' is false, but 'snow is white' is true. A number of languages embed in their color vocabulary the quality of sheen, so that dull-white is represented by a single lexical element different from shiny-white. There are lots of like examples, many of which are by now quite familiar to students of language.

One advantage of Aristotle's approach to metaphor--an advantage in itself and in relation to the translation issue--is that Aristotle does not contrast metaphorical with *literal.* Instread, he contrasts metaphorical use with ordinary use. (Ares' shield is proper to him and to call it a cup is a metaphor, but Ares does not literally have a shield.) The Latin rhetorical tradition follows Aristotle in this. Contemporary discussions would be improved if they were to follow this example. The translation issue looks quite different if we consider it in light of that contrast: the many metaphors we meet in translation need not automatically be contrasted with literal meaning or taken to undermine literal translation.