Re: A fairy tale

[CJ FINE <cbmvax!uunet!BRADFORD.AC.UK!C.J.Fine>]

Continuing the exchange with Ivan:

> >  > >
> >  > >  lu le nolraixli nega'u le dembi li'u
> >  >
> >  > {ga'u}, `up from <direction modal>'?  Does this work for static `on'?
> >
> >  I don't know how else to do it!
>
> How about {gapru} ({le nolraixli poi gapru le dembi})?

I wasn't sure at the time, but glancing at the tense paper (s. 8) it
seems to me that the 'up from' in the definition of ga'u has misled us
into thinking of movement - unless you use "mo'i", it means "the
direction from the reference to the modificand is upward".


>   If this ambiguity is regular, that is, if it is not by
> chance that English uses the same word, then we're in trouble, because
> the disjunctive concept won't be easy to put into Lojban.

But it is rarely necessary to do so. When translating into a foreign
language, you sometimes cannot (or have to work hard to) get a precise
rendering - especially if the word in question has idiomatic uses or a
number of overlapping meanings. Different strategies will be appropriate
in different contexts. Sometimes a precise and lengthy rendering, even
translating the uncertainty as an alternation; sometimes a choice,
arbitrary or not, among the possible meanings; and sometimes a
relatively vague term which preserves as much of the meaning as the
translator thinks relevant or useful.

I remember an example Whorf quoted from Pawnee, a sentence he translated
as "I clean it out with a ram-rod" (referring to a gun). His analysis of
the Pawnee was, if memory serves,
        "I interior;moving-place;with-a-tool;it"

A comprehensible English translation of this is anyway holistic, but
both inserts and omits detail.

If others produce Lojban that renders the original with great semantic
precision, I am quite content - though I reserve the right to criticise
if I think the result does not convey the intended meaning, or indeed if
I think it is inelegant. But equally I wish to translate according to my
own canons, which tend towards
        1) preservation of the structure of the original
        2) exploration and exploitation of the resources of Lojban
        3) willingness to discard semantic precision when I judge it
                inessential and awkward to render.
Of course I sometimes vary these: my Quine translation works hard to
preserve semantic precision, because that is the nature of the text. And
I welcome criticism (this discussion is not a dispute - I am glad that
our successive comments have helped me to articulate these personal canons).
>
> >          The precision you suggest is not necessary here. Even if the
> >                  prince is in fact the son of a king (as it later appears
> >                  he is) that does not mean that you have to say so.
>
> I fail to see why you would want not to, given that it takes no extra
> labour; that is, why you want to keep the imprecision of the original.

I refer to the above. In general, I regard a longer lujvo as "extra
labour".

>
> >  Good thought about the pronoun. I hate "goi" - it always seems
> >  intrusive, particularly in a translation (as opposed to an original
> >  composition).
>
> Not in my eyes.  I love {goi}.  Three short syllables - {goi qo'w} -
> save you a lot of repetition afterwards, particularly in original
> composition (as opposed to a translation; but still you'll find that
> many referential noun phrases have no particularly literary value and
> will profit from being replaced by a pronoun).

I agree retrospectively - anaphora is wonderful, and I use "ri" and "ra"
a lot. Indeed, I use "ko'a" etc. a lot - but the initial assignment I
always find intrusive, particular when there are other relatives, and
you need "zi'e" as well.

                        kolin