The Whorf-Sapir Hypothesis and the Translation Muddle

Abstract

The history of translation is littered with normative, exhortatory, and judgemental propositions. 'Translators are the post-horses of civilization', said Pushkin. 'Poetry', said Robert Frost, 'is what gets left out in translation.' An old Italian proverb tells us that 'traduttore traditore', to translate is to betray, and the male chauvinist French speak of 'les belles infideles', faithless beauties; translations, it appears, are in Latin eyes like mistresses if they are beautiful, they are unfaithful, and vice versa. Comments of this kind have their place in the history of taste, but their descriptive and interpretative power is questionable. Moreover, they have tended to reinforce the common error at the heart of much translation theory which postulates an 'original text' or 'source text' as some ideal entity having an absolute, autonomous existence in its own right, against which any given translation can be measured by criteria of exactness or adequacy which are clearly, in themselves, exceptionally inexact and inadequate. There is no other branch of literary studies in which one right reading is hypothesized, or the reading process seen in terms of a simple cause-and-effect or linear model, which is what happens with alarming frequency in commentaries on translations. There are complicated historical reasons why Translation Studies have only just begun to dissociate themselves from the tyranny of this source text/target text model derived from Classical Studies and their afterlife in Modern Languages departments, a tyranny shaken many years ago in other branches of literary scholarship and literary criticism. One reason is the special type of 'inauthenticity' that is attributed to translations: being somehow designed to take the place of the original, translated texts thus acquire not only the distortions implicit in the value-charged idea of the 'original', they also compound such distortions in some kind of Oedipal model of successiveness, whereby the translated text acquires, by an act of usurpation, the bogus virtue of originality, while at the same time recognizing its own helpless subservience. On such grounds as these, we speak glibly and in quite inappropriate contexts of a 'good' translation, or sometimes even of an 'accurate' translation, or worse still (is it?) of a