new translations

[kichenas at math.umn.edu]

This is further to two recent comments by Keijo Virtanen
and  Birgit Kellner on this subject.

First off, it seems many *would* feel unconfortable with a translation
which deliberately introduces new elements into a text. But then I suppose
that is also what is involved in giving an ``accurate, documented and
objectifiable'' translation.  I assumed that B. Kellner's point was about
what one wants to call `objectifiable criteria,' and not about their being
abandoned for the benefit of unsupported speculation. The issue of
defining these criteria is a secondary issue---first one has to agree, as
I think she did, that we do not want translation to be left in ``the realm
of ultimate arbitrariness and subjectivity.'' One could discuss in various
directions the role (and definition) of what was called ``textual
surface'' (seemingly trivial elements of syntax are often useful in order
to detect flaws in interpretations; at the same time, we certainly cannot
let the tree conceal the forest); this is certainly relevant and
interesting but is probably, as far as the issue of level of rigor is
concerned, a minor point. 

On the comment that (the term `scientific' or rigorous translation)
``suggests that there is a type of translation which is more mechanical
than another, less creative, and sort of shuts off the translator's own
ideas during a machine-like process ...'' it is interesting to note that
this is an extrapolation from the comment it is supposed to respond to,
and is based on the assumption that `scientific=mechanical.'

I suppose I already addressed the issue, and that the fact was recognized
by B. Kellner, but since it was brought forward again by K. Virtanen, it
may help to stress the quite obvious fact that contextual information is
fundamental in any attempt at understanding a text. These are elements
independent of the translator's fancy, which are implied by the text, and
without which the text would not exist... If nevertheless not enough
objectifiable arguments are available, which certainly happens, should we
not say so rather than favor speculation? The fact that we are able to say
so much about cultures very remote from us in time or space (remember
Sumer!) shows that there are some objective statements that can be said
about them.  It is desirable to separate them from others. This is of
course what the better editions accomplish. 

I was also puzzled by B. Kellner's comment on my third point, which was
clearly stated as related to another issue. To put it bluntly, it
addressed questions such as `why should be care about Dharmakiirti when we
have modern theories of perception'? Recall that the relevance of ancient
ideas to modern issues is settled: 

(1) by clarifying what these ideas were (that's where translation, and
other issues addressed in this list are helpful); 

(2) by `amending or updating' ancient conceptions to extract what is true
from them and giving correct justifications to replace flawed arguments,
*and* finally by explaining the origin of what we decide were `errors,'
provided of course the text is not considered to be beyond criticism for
some reason. 

This latter task can be addressed at various levels. `Updating' is made
necessary by the fact that correct conclusions can occasionally be derived
from incorrect premises or by incorrect arguments. 

It may be useful to stress that the establishment of rigorous translation
standards in translation into Western languages has a long history, going
back to editing problems for Western Classics. The application of these
principles to Indian languages is quite natural, and not new. 

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