Hi Alison,

Thanks for your thoughtful reply.

Well, it's curious this idea that translating poetry 'can't be done' since I try to do
it all the time, and am always glad for translations, particularly from languages I
can't read in the original. So that seems a bit of a logical conundrum, no? and
perhaps beyond reason, since that would make all translators fools for
impossibility.

>If you read two or three or four versions of a poet's work, as I often do,
>then you might end up with a sense of what they hold in common, which
might
>be something like what the poet was doing.

I often do this too, but am just as interested in what they _don't_hold in
common, and perhaps more so, just as I am particularly interested where the
English translation (and I'll just assume English speakers here since most of us
are and much of what you say about reading from those assumptions is true,
and perhaps a given) becomes 'unbeautiful' or 'clumsy' or full of syntactical
interlocutions. It's out of the gap between that some sense of the other, the
original, arises like a shimmering phantom. So while I might note what severral
versions have in common, I am generally more questioned and interested in
being questioned by their differences.

I agree basically with your reading here

 >The kinds of things that particularly grated my ear were phrases like "the
>body that's loved" (V) as compared with "beloved body" (E) - they seem
>clumsy and unnecessary locutions which don't affect meaning at all but do
>affect my reading of the poem in English.

of the merits of Economu's translations, versus the Haviaras translation and that
should be (H) not (v), in the last stanza, but agree mostly in terms of efficacy in
English. While "beloved body" is a happier choice in English, just as perhaps the
ending trochee, as you noted earlier is, I wonder at the differences between the
two and at the Haviaras interlocutions if Haviaras isn't gesturing at something in
the original? that perhaps  the vague ambiguity and stealth, the ambivalent eye
of 'everyday morality' that found certain sexual behaviors allowable as long as
they weren't too open, too intense, too apparent, "unseemly" isn't closer to
Cavafy's being in Greek Alexandria, rather than Economu's version which
emphasizes a certain frisson between high moral value and the 'shameless', a
kind of polarity of Anglo/English culture and language. So the E version goes
over more strongly with us, but is it what Cavafy meant? Admittedly, I think I'm
assuming good will and a certain linguistic ability (as opposed to not knowing
the language or the stupid mistake or misreading) in translators, that there's
little to recommend translating poetry or being a fool for an impossibility that
renders one's work eventually  obsolete except persistent love of the original, a
sort of stupid stubborn devotion.

Also to answer your question
>I simply don't see how they
>reflect a "particular sensitivity to some element that exists in the
>original": it's the kind of English I am always paring out of my prose.

" the body that's loved" versus "the beloved body,"  there is more than a ghost of
a difference in the English, "the beloved body" takes on the suggestion of the
category, not a particular body that's loved, but just the beloved body, whereas
"the body that's loved" has at least the suggestion of a particular body. This is
very slight, but it does seem a difference that exists in English, "I thought of the
body that's loved" is different than "I thought of the beloved body," if it were
"the beloved's body" that would be more particular, but as it is, there is a slight
gap here between that makes me curious as to the original.

 I've often noticed in translating that these syntactical interlocutions
occur precisely when I am most aware of a sort of multivalent presence in the
text, an interweaving of subjectivity and culture that intersects with the
language, and yet my choice as a translator is to choose one word in one
language for one word in another and unable to find the word that can so
convey that multivalence which I sense there's a kind of knot of language that
occurs, sometimes merely explanatory. The sort of thing that's death to a
translation, in terms of working as a poem in English. And yet to the translator,
it's most interesting, since it's precisely there that one is being riddled by the
original. It's also those knots that are often most interesting to me as a poet in
reading other's work or my own.  I agree with much of what you say about
reading in English or as a poet with one's own sensitivities and practice. I
suspect part of the difference here is that generally I read translations with a
translator's head, which is to try and circumvent my own sensitivities and
practice and have a kind of identity void, full of various presences and
possibilities, a willingness to be haunted by whatever the phantom is, always a
most interesting encounter in translating in part because it can take me outside
of my own taste and practice and poetic thinking,

Best,

Rebecca
---- Original message ----
>Date: Mon, 17 Jan 2005 09:14:29 +1100
>From: Alison Croggon <[log in to unmask]>
>Subject: Re: Mark, two translations/same Cavafy poem
>To: [log in to unmask]
>
>Hi Rebecca
>
>The problem of translating poetry is that it can't be done.  How can you
>render a work that foregrounds the particular sounds and rhythms and
>semantic associations of one language into another?  But of course,
>translations happen all the time: and personally I'm grateful that they do.
>If you read two or three or four versions of a poet's work, as I often do,
>then you might end up with a sense of what they hold in common, which
might
>be something like what the poet was doing.  All the same, I am always going
>to prefer those translations that make the most beautiful poems in English.
>As a poet and reader, beauty matters to me; and I have a fairly broad idea
>of what I mean by that, which includes what is often called unbeautiful.  As
>a poet, and I don't see what can be done about that, I am English-centric;
>it's my language, it's my material, it's what I make poems out of; and my
>primary interest is always going to be, in the end, what disturbances
>translations might create in its fabric, how it might be torn open slightly
>or stretched or warped into some new possibility of expression.
>
>Arguments about translations are always going to be about subjectivities, as
>are all arguments about poetry.  I respond to the poems I respond to.  As
>with reason, or Rilke's ladders, such things are forever without ground.
>All the same, it seems a bit misleading to me, beyond fairly basic mistakes
>and obvious misinterpretations, to refer to a stable original as the
>authority to trump the argument.  The original is surely susceptible to all
>these interpretations; it's how each translator reads the poet.  In poetry,
>the aesthetic/stylistic choices are always going to be as crucial as any
>semantic decisions; I don't see how that can be avoided, or why it would be
>desirable.  The one thing you can't do is make exactly the same poem as what
>is translated; if that were so, all languages would be the same.  And they
>manifestly are not.  Given that, you end up with a bunch of different
>versions, each of which perhaps incline to a slightly differing aspect of
>the poem; and the rest is up to each individual reader.
>
>The kinds of things that particularly grated my ear were phrases like "the
>body that's loved" (V) as compared with "beloved body" (E) - they seem
>clumsy and unnecessary locutions which don't affect meaning at all but do
>affect my reading of the poem in English.  I simply don't see how they
>reflect a "particular sensitivity to some element that exists in the
>original": it's the kind of English I am always paring out of my prose.
>That said, it's not that I think Variasis' translations are without merit.
>I simply preferred the others, as is my right as a reader, and attempted to
>articulate why.
>
>Best
>
>A
>
>Alison Croggon
>
>Blog: http://theatrenotes.blogspot.com
>Editor, Masthead:  http://masthead.net.au
>Home page: http://alisoncroggon.com