Theiling Online    Sitemap    Conlang Mailing List HQ   

Re: Relexes Pt. 1: Defence

From:Tim May <butsuri@...>
Date:Sunday, December 14, 2003, 17:44
Ray Brown wrote at 2003-12-14 15:53:51 (+0000)
 > On Saturday, December 13, 2003, at 10:23 PM, Tim May wrote:
 > [snip]
 > > In any case, grammar is only indirectly relevant to whether a
 > > conlang is a relex.  The real point is the lexicon - whether each
 > > dictionary entry covers the same semantic field as an equivalent
 > > in the natlang.
 >
 > No, the term _relexification_ is derived from sociolinguistics
 > where it was coined to refer a theory about the origins and
 > relationships between pidgins and creoles.  The theory is that the
 > range of English, French, Spanish, Dutch etc. pidgins is derived
 > from the first widely used pidgin, the Portuguese pidgin of 15th
 > cent. West Africa.  According to this theory, the _grammar_ of the
 > west African pidgin was retained while new _lexical items_ were
 > introduced from other European languages, i.e. the grammar stayed
 > the same but it was given a new (or revised) lexicon, hence
 > 'relexification'.  The supporters of the theory maintain it
 > provides a satisfactory explanation for the many grammatical
 > similarities noted between European-based pidgins & creoles.
 >

I'm aware of the origin of the term relexification, but I think
"relex" means something different in conlinguistic parlance than it
does in sociolinguistics.  However, on reflection, I'm not at all
confident that what I said above was correct even within the realm of
conlanging.  At least it should not have been expressed in such
black-and-white terms.  It is clear to me that the prototypical relex
conlang differs little from its parent in either lexical scope or
grammar.  I'm not sure that our terminology can make this distinction
clearly as yet.  Suggestions, then, would be valuable.  (Perhaps
such terms are already in existence unbeknownst to me?)

What I was thinking of when I wrote the above was a discussion of
which of these - grammar or the scope of lexical items - was the more
important in determining whether a conlang was or was not an
"encoding" of some other language, which I remember having read in the
archive.  I cannot find anything satisfying to quote, but I believe it
took place in this thread:
http://listserv.brown.edu/archives/cgi-bin/wa?A2=ind9811A&L=conlang&P=R6141

Reply

John Cowan <cowan@...>