Re: [topicmaps-comment] multilingual thesaurus - language,scope,and topic naming constraint

[Bernard Vatant <bernard.vatant@mondeca.com>]

BV
> > So we have to go back to basics: one subject = one topic.
> > (DAN : økonomi), (DUT : economie), (ENG : economy), (FRE : économie), (GER :
Wirtschaft),
> > (SPA : economía) convey a priori six different concepts and views of the world, that
> > someone familiar with all those languages could certainly feel, even if the
differences
> > are subtle.

JA
> I think that I disagree with you (no intention to sound rude here).
> I am absolutely convinced that such a subject as the 'science of economics' is the
> same regardless of the language that it is expressed in.

And I am absolutely convinced of the contrary, and am not the only one :))
See e.g. John Sowa's "Knowledge Representation" chapter 7, pp 408 sqq.
This subject, like any other one, has been forged in some language/culture context, and
has not to be enforced as a relevant subject in any other languages/cultures willy-nilly.
That is linguistic imperialism that TM approach should try to fight back, and not
contribute to enforce. This goes with my absolute conviction that there are no subjects
preexisting to language, out there waiting to be named, but that subjects emerge from
linguistic activity, and therefore are essentially linked to some linguistic/cultural
context.

JA
> If a multilingual
> theaurus editor writes 'science of economics' down in three languages, I assume
> that he'd have a single subject in mind.

Yes, but the editor always stands from some default language viewpoint. The thesaurus is
multilingual, but the descriptors are always defined in a default language, generally
english.

JA
> Also, this does not solve the problem that economía (ESP) is a polysem, and any user
> of the thesaurus will still not know what economía she's looking at.

She will have a way to know, because she will see that "economía" is involved in two
different associations.

JA
> I really think that it's simply an error in the thesaurus and that it might be better
> to accept that rather than to build your TM design on it.

In this very case, it might be that it could be considered an "error", but I don't know
spanish enough to decide if "economía" has to be split into two subjects from a spanish
viewpoint. And correcting that kind of "errors" to align ontologies willy-nilly on the
default language descriptors is the pernicious and slippy slope towards linguistic
imperialism.

Bernard