deepreef at BISHOPMUSEUM.ORG
Sun Feb 27 21:44:35 CST 2005
> > I think that the tricky part is that often there is an imperfect
> > correspondence between latin names and vernacular names
> Imperfect is such an emotive and disparaging descriptor...
I prefer to characterize it as "wry understatement"...
> there is
> nothing intrinsically imperfect about a vernacular concept or a
> vernacular name - they are real...
Indeed! But it is the correspondence between latin names and vernacular
names (not the names themselves) that is imperfect. I was commenting on the
task of establishing correspondence between the two kinds of names.
> but there often is in the way they
> are used or interpreted, arbitrary invention of something that does not
> need to exist being the ultimate evil...
One could say the same thing about many Linnean-type names as well.
[For the record: with very rare and justifiable exceptions, I steadfastly
refuse to invent new vernacular names.]
> > Clearly there needs to be a many-to-many relationship -- and
> > perhaps one that accomodates "in part" correspondence.
> That's the right answer - there is nothing wrong or imperfect with a
> many-to-many relationship...
I don't believe I ever intended to suggest that there was something wrong or
imperfect with a many-to-many relationship. Personally, I love 'em and
incorporate 'em all the time (in my databases, that is -- I wouldn't want my
wife the get the wrong idea...)
But I used the word "imperfect" with the intended meaning of "not perfect";
the presumption being that a "perfect correspondence" between latin names
and vernacular names would mean that each vernacular name represented a
concept that exactly matched a corresponding concept represented by a
Linnean-type latin name. But alas, there is no such "perfect
correspondence"; and hence an "imperfect correspondence".
> it just 'is' and has to be accommodated,
> referenced, contextualized... that may not be what people want, but hey,
> life was not meant to be easy... :)
That's certainly what I want -- as verified by the fact that my database
design accomodates a many-to-many correspondence between latin names and
taxon concepts, and further supports a many-to-many correspondence between
taxon concepts and venracular names, and therefore a many-to-many-to-many
correspondence between latin names and vernacular names. (It's actually more
convoluted than that, but I'll save it for another thread....)
More information about the Taxacom