Kwame Anthony Appiah is apparently a philosopher and so his reflections on "Thick Translation" spring from some fundamental theories. He begins by looking at some proverbs from the Twi or Akan language (there's evidently an iceberg of reasons why both names have been used for the language).
In their literal translations, these proverbs don't seem to mean much ("The drongo says: if he had known that the palm nuts were going to ripen, then he would not have married the raffia palm with a twisted leg"), but Appiah then goes on to consider their "literal intention". He makes a working assumption that the Sapir-Whorf hypothesis, however true or not, doesn't mean translation is impossible. Twi has no word for "neutral boson", but then neither did English until recently. YOu can always build a concept by the use of existing words within a language.
So to Searle's notion of direct and indirect speech acts (which is different from the notion of literal and non-literal). Direct speech acts occur when the main point of the utterance is accounted for by the literal
intentions. Actually, I think he goes wrong here, since he then says "sometimes indirect communication proceeds by way of literal intentions and sometimes it doesn't." But it looks like the concept in fact means that all speech acts have literal intentions; the degree of directness depends on the closeness of the utterance to that intention. For example, if I say "This is a very busy road" to a traffic surveyor, it's a direct utterance. If I say it to a child, it's indirect, because my literal intention, the meaning of my speech, is "Hold my hand". I think the term "literal intention" is the problem here, but I guess we're stuck with it.
In the rest of the essay, Appiah argues that it always possible to reconstruct the intention of the original text, if necessary with the use of "thick" contextualisation. This brings up something that's been nagging at me for some time: does a translator have to consider the intention of the source text, in a way that a modern reader would regard as inappropriate? Appiah tackles this question, but again, I think he muddles the meaning of "literal intention". I also suspect he's not adequately distinguishing between "meaning" and "intention". It all becomes a bit of a mess.
And then this:
But for literary translation our object is not to produce a text that reproduces the literal intentions of the author - not even the one's [sic] she is cancelling - but to produce something that shares the central literary properties of the object-text; and, as is obvious, these are very much under-determined by its literal meaning, even in the cases where it has one.
So many begged questions there. But the essay closes better with the view that there is a variety of valid approaches to translating (as there are to reading).
I feel impertinent in suggesting that Appiah has misled himself in his argument, but that seems to be the case and it's infected the whole piece, so I can't rely on the conclusions.
But to return to the question of the author's intention. I'm beginning to think the translator must try to understand this, in order to reproduce it. From the examples of the proverb, it's clear that you need to know that the utterance isn't intended to be an Attenborough-ish comment on the behaviour of the drongo. If you were translating, say, Siegfried Sassoon's war poetry you'd need to know the ironic intent, and it would help to know something of his personal story. That, modern theory would say, isn't something you
need to know in advance of reading the poems.
I took a long time over that last paragraph. Something's wrong here, and I'll come back to this again.