The uniqueness of metaphors appears to be the one thing translation theorists can agree upon, and it seems a bit conceited to maintain that translating a phenomenon held to be so exceptional represents no challenge at all, and can be done by a simple word-for-word rendition. Nevertheless, the view that metaphors are untranslatable also seems a bit too extremist, and my above-mentioned example (“and pigs/donkeys can fly”) appears to contradict this fact. I readily admit that the words “pig” and “donkey” have different connotations, but in this context, the meaning the metaphor conveys ought to be more or less thesame.

As Newmark notes: “I think the first purpose of metaphor is to describe something more comprehensively, economically and usually more forcefully than is possible in literal language (1998:111). Dagut seems to agree when he says: “The truth is that the resemblances underlying metaphor (…) are largely “created” by the observing and classifying mind of the speaker, and are therefore as infinite as they are unpredictable” (1976:27).

The functions of metaphor are basically twofold: one is its rhetorical function, while the other is its cognitive function (Newmark, 1995). They are pointed out in particular not because other types of language do not have these functions, but because in metaphor they perform differently.

On the one hand, metaphor, as a powerful rhetorical device, is employed to compare one concept to another with shared features or properties so as to appeal to the senses. In other words, metaphor, considered as a decorative addition to ordinary plain language, is used at certain times to achieve an aesthetic effect (Newmark, 1981b, 1995).

On the other hand, metaphor functions as a device of language formation whose purpose is to describe a concept, an action or an object more comprehensively and accurately than is possible by using literal or physical language. In this case, it is the denotation rather than the connotation of the metaphor that addresses the receptor, hence highlighting its cognitive function (Newmark, 1981b, 1995).

In a good metaphor, the two functions fuse like content and form. Yet the cognitive function is likely to dominate in a textbook, while the rhetorical function is often reinforced by sound-effect in an advertisement, popular journalism, an art-for-art’s sake work or a pop song (Newmark, 1995).

It seems apparent that the solution must lie somewhere in between the two opposed views, and Newmark’s diagram underpins this notion. In my opinion, language is so contextual that maintaining any view as to how something ought to be translated is futile. You must judge each utterance separately, and in its proper context before you can make a decision. Every use of language is unique.


Anderman, G. Lecture notes dated 01/05/02
Dagut, M. B. Can “Metaphor” Be Translated?, Babel, 1976
Lomheim, S. Omsetjingsteori, Universitetsforlaget, 1995
Newmark, P. Paragraphs on Translation, Multilingual Matters, 1993
Newmark, P. More Paragraphs on Translation, Multilingual Matters, 1998

Schaffner, C.(2004). Metaphor and translation: Some implications of a cognitive approach. Journal of Pragmatics 36, 1253-1269.

Snell-Hornby, M. (1988/1995). Translation studies: An integrated approach. Amsterdam: John Benjamins Publishing Company.

Tabakowska, Elzbieta (1993). Cognitive Linguistics and Poetics of Translation. Tübingen: Gunter Narr Verlag.

Van Den Broeck, Raymond (1981). “The Limits of Translatability Exemplified by Metaphor Translation. Poetics Today, 2: 4, 73-87.