Sapir-Whorf Hypothesis

Hi all,

I've been lurking here for a while now, with a view to finding out more
about conlangs (not sure why, I'm just interested in
languages).

Anyway, I read some literature on the Sapir-Whorf hypothesis that the
language in which a person thinks structures their thoughts.
An interesting proposal, obviously leading to such languages as loglan,
lojban and laadan, to name a few.

However, while I discussed the idea  with my father (who is a native
speaker of German and speaks both German and English fluently), we came up
with a new hypothesis:

Let us assume that a person is fluent in two (or more) languages. For the
purposes
of this argument, let these languages be English and Japanese, both from
two separate language families (however they could be two languages from
the same family). The one language will never quite translate directly
into the other. However, our hypothetical person can think equally well in
either one.
Suppose they have, unconsciously, developed a para-language of "symbols"
which they use to translate between the two verbal languages. In essence,
the person would be able to think in symbols, as well as (or instead of)
words.
If this were correct, then just learning several languages would
facilitate thinking that was as fast (or faster) than if one were to
be a native speaker of lojban?

I'd love to hear your comments on this one.

Walter Mack
mack@uq.net.au