new translations

[kichenas]

Further to Gary Tartakov's comment on translations, it may be useful to
point out that in practice, the internal consistency of the texts usually
enables one eventually to figure out what the author really meant. It is
quite interesting to do this for poems of the Sangam, some of which are
rather short with little contextual information (from a colophon) but
which can be given a single correct interpretation. There are of course
many other examples. The translation itself further requires of course
using to the fullest the capacities of the target language, which leads to
issues already raised e.g. by B. Kellner. 

All things considered, there are only three possibilities: 
either one satisfactory interpretation has been achieved
or the text is deliberately ambiguous
or we must admit that we lack enough arguments to conclude what
the author originally meant. It seems desirable that a translation make clear
which of the above applies to it, and that a critical apparatus, if
provided, should explain why. That is what the better editions accomplish,
and there are many of those.

I don't think most of those who tried their hand at translating would
call the result `easily achieved,' but was it supposed to be? Contextual
problems always have to be coped with, even with technical literature---as
anyone who tried to read ancient mathematical treatises, or even
contemporary scientific works in a field other than his/her own, knows
well. 

Perhaps the term `scientific translation' should be replaced by `rigorous
translation' lest it may suggest a (pointless) opposition between
`scientific' and `literary' pursuits. 

                                Satyanad Kichenassamy
                                School of Mathematics  
                                University of Minnesota
                                E-mail: kichenas at math.umn.edu