Separating Nomenclature and Classification (use of rank the rank

[Jim Croft]

> During the preparation of the Prometheus taxonomic model (Pullan
> et al.,Taxon 49(1): 55-75 (2000)) we thrashed out these issues, and
> resulted in a model that truely separated nomenclature from
> classification. Thomas Lammers rightly comments that these are
> distinct and should be treated as such.

The situation is actually a little more complicated that this when
dealing with plant names in botanical databases, especially those that
have to be used for the real purposes we apply names to organisms, such
as managing bilogical collections, etc.

We need the names themselves and all their components (ie. nomenclature)
and publication details, whether the name is currently useable or what
is the useable equivalent (ie. taxonomy, synonymy), where the name fits
in the scheme of things (ie. classification, with alternatives), when
the taxon fits in the scheme of things (ie. phylogeny)

And for collections management, we need to know where we ended up
putting the bloody thing on the shelves (which may or may not
correspond to any of the above).

> We also arrived at the same decision as IPNI, in that names below
> the rank of species are, for nomenclatural purposes, trinomials,
> irrespective of how they were published. Susan Farmer gave more
> details on this, again confirmed by Thomas Lammers.

We too arrived at this decision for the databases at the Australian
National herbarium and Australian National Botanic Gardens, (and fed
this ideology into the IPNI database design process) relieved that
the International Code of Botanical of Botanical Nomenclature had at
last done something useful in an informatics sense and provided a
mechanism and justification to simplify data models, application
design, information management and the tasks to be done in the
herbarium.

> In Prometheus we link the infraspecific epithet to the specific
> epithet, which itself is linked to the generic name (these are termed
> 'name elements' in our model). This three-level hierarchy necessary
> for the construction (concatination) of the name, and is the only
> hierarchy present on the nomenclatural side.

Yep...  this seemed the way to go...

Impressed the the elegance of this solution, the wisdom of the crafters
of the Code, the imprimatur of the combined might of the international
botanical community, and the unquestionable authority of the published
Code, we removed all infraspecific ranks and data from the database and
presented this remarkable and succinct labour-saving device to the
users.

Well, rather than congratulating us and bestowing us with accolades and
unimaginable wealth, the users were outraged and demonstrated that they
were indeed motherless luddites with no appreciation for beauty or
truth.  These followers of Onan claimed that they could no longer find
genus X species Y subspecies A variety B subvariety C forma D subforma E
in the collection.  It was pointed out that under the Code that this was
the same as genus X species Y subforma E and could be filed as such and
found just as easily.  Rather than change our database, they should
rearrange their collection.

But they would not have it.  And we put a complete infraspecific
hierachy back in.  And now the database and herbarium are both a
complicated, inconsistent and partially populated mess, and botanists
continue to have a lot of meaningless trivia to talk about over morning
tea.  But apparently this is good, because real life is like that...

> Other links (to other
> infraspecific names, to infra generic names, or to names above
> genus) purely convey classification information. This data SHOULD
> NOT be included in the nomenclatural side (even ascribing a name
> to a family), but represented in the classification side of the
> database.

IPNI tries hard to avoid making statements of taxonomy and
classification (even though this is what most users want) because the
tasks involved would be completely overwhelming.  But simply assigning
genera to families so we can find them is taxonomic and classificatory
so it is difficult to be totally pure in this area.

> The exception to this is that there needs to be a
> nomenclatural link the highest rank (to the kingdom etc.) to
> indicate which code you should be using, ICBN, ICZN, etc. So to
> answer Curtis Clark, the various ranks of infraspecific 'names' are
> not hierarchic for nomenclatural purposes, but they should be used
> in a hierarchy when constructing a classification.

This always give me a headache and I am sure the problem is the binomial
system, but I am not sure why, nor what to do about it.  Coming down the
hierarchy, there seems to be an illogical paradigm shift just after you
hit the genus...

Ah, heck... it is just a name...

-- jim