Comprehensible descriptions

Mike Dallwitz miked at ENTO.CSIRO.AU
Wed Mar 22 16:30:20 CST 1995

                                                                 22 March 1995

> From: Curtis Clark <jcclark at CSUPOMONA.EDU>
> To: Taxacom
> Botanical Latin has a great advantage over English in that it is
> grammatically difficult to be ambiguous in the former. I remember reviewing
> a manuscript many years ago, and finding ... that it *sounded* okay, but it
> had one of those situations where a group of nouns and adjectives precede a
> noun, and it is impossible to tell what modifies what. The English seemed
> comprehensible, but wasn't. I would suggest that we create "botanical
> English", with a restricted vocabulary and precise grammar, to take the
> place of the old stalwart.

Natural-language descriptions generated by CONFOR from DELTA data are
comprehensible, unambiguous, and comparative, provided the DELTA data are well
constructed. Some people find them unacceptable because they place some
restrictions on the way that characters are combined into sentences, use the
semicolon as the normal delimiter between characters, may be a little stilted,
and may contain a few redundant words. We will improve these descriptions
somewhat in the new CONFOR, but, to a large extent, the above `undesirable'
attributes are the very ones that make the descriptions unambiguous.

I have always been surprised at the vehemence with which some people oppose
the use of the semicolon to separate characters (apparently solely on the
grounds that it's just not done that way). Against my better judgement, we
have provided the ability to use commas instead of semicolons, which greatly
increases the possibility of ambiguity, and can reduce the ability to see the
correspondence between different descriptions.

Mike Dallwitz                                  Internet md at
CSIRO Division of Entomology                   Fax +61 6 246 4000
GPO Box 1700, Canberra ACT 2601, Australia     Phone +61 6 246 4075

More information about the Taxacom mailing list