On Mon, Jun 8, 2009 at 8:21 AM, MacLeod Dave<[log in to unmask]> wrote:

> You could also make the argument though that complete translations of
> large books carry much more weight when trying to impress people for
> the first time. Let's say Sambahsa had a complete translation of
> Dracula, the Bible, and Plato's Republic, compared to about fifty
> smaller pieces of some forty pages each. Which would be more
> impressive?

To me and probably other people knowledgeable about
languages and translation issues, for any given corpus size,
more varied = more impressive.  But you may be right
that a few long works are more impressive than a lot of short ones
to most people.

> people with. That also means though that choosing just the right novel
> is of crucial importance since it takes so much time.

It needs to be public domain, or a newish novel that's under a
Creative Commons license that allows translations; translating a
standardly copyrighted novel into a conlang on spec and then getting
permission to publish it after the fact can be a nightmare.  There's a
translation of the first Harry Potter book into Esperanto that's never
going to see the light of day because Ms. Rowling's agents or
publishers won't license it on terms that an Esperanto publisher can
afford, for instance.

And it needs to be a novel that the translator loves and won't mind
reading again many times.  Given those two criteria, I suspect adding
a third criterion (it has to be impressive to other people as well)
won't actually cut down the pool of candidates very much.

It seems there are at least two reasons to produce a translation: to
exercise the language itself or one's skills at writing in it, or to
produce something someone will want to read.   In the early stages of
conlang development, it may be best to focus on the first of those and
leave the other to take care of itself.  A lot of the early
translations into a new conlang will become archaic by the time the
language has a sizable speaker community (if it ever does) but that
doesn't mean that making them was a waste of time.

Tying into the other thread, where we were talking about consistent
semantic schemata for conlangs, I suspect that original writing in a
conlang is more likely than translation to help the conlang's
semantics and pragmatics develop in an internally consistent way
independent of natlang models; whereas translation is more likely to
uncover lexical and grammatical gaps where the conlang creator hasn't
thought of some issue yet until the challenge of rendering someone
else's thoughts and words into the conlang forces them to confront it.
  That might suggest that a new conlang should be exercised at first
by a certain quota of original writing, with the first translations
coming later.

Jim Henry