[SEEK-Taxon] RE: concept types, relationship types

[Richard Pyle]

Hi Nico,

>    We're working towards a similary revised list for
> relationships. A first idea
> is to move away a little from the purely nomenclatural/mostly taxonomic
> distinction that has shaped the LC/TCS discussion. Instead we
> might talk about
> "Code-prescribed" relationships versus "concept relationships". The latter
> include  as subcategories vertical (within classification) and
> lateral (among
> classifications) as previously.

One thing I forgot to mention in my last email to you, that I've been
thinking a bit about since the LC/TCS discussions got started, is the
difference between these two statements:

"Jones designated Aus bus Smith as the type species for the new genus Xus
Jones."

  - and -

"Jones placed the species Xus bus (Smith) Jones SEC Jones within the parent
taxon Xus Jones SEC Jones."

There are two ways of looking at these two statements: In one view, the
first statement is simply a qualified version of the second statement, but
they are otherwise fundamentally the same.  In the other view, the first
statement is a strictly nomenclatural statement (and belongs in the LC
realm), and the second statement is a concept statement (and belongs in the
TCS realm).

It would seem redundant to record this information in two separate places
(once within LC, and once in TCS outside of LC) -- the assumption being that
one cannot designate a species as being the nomenclatural type of a genus,
without also simultaneously asserting that the genus is a parent taxon
having an "includes" relationship with the species concept.  There may be a
handful of oddball technical exceptions to this assumption, but I suspect
not enough of them to warrant dictating the design of LC & TCS.

So...when I say that my many thought-experiements concerning this
distinction consistently lead me to the conclusion that they are
fundamentally different pieces of information and that the first statement
should be accomodated by LC and the second statement simultaneously
accomodated by TCS -- I say so from a purely information-management
perspective.

I think this is one of the areas that needs further discussion.

>    With respect to the code-prescribed relationships, there may
> be at least
> three useful subcategories.
>
> (1) "objective" - judgments about the availability/validity of an
> old name (in
> relation to a new one) that follow necessarily just by applying
> code rules (pure
> literature study, no NEW judgments about taxonomic relationships;
> example: a
> grammatical error [masculine genus name/feminine species epithet] has been
> detected).
>
> (2) "subjective" - judgments that follow only because a new
> taxonomic assessment
> alters the availability/validity of an otherwise Code-compliant
> name (example: a
> species is transferred to another genus due to taxonomic
> reassessment, thus a
> new combination must be created; 2 types with different names are
> judged to
> belong to the same species - one name becomes a junior synonym).

As a zoologist, I have no problem whatsoever with relegating the latter of
these to the domain of TCS (non-LC).  However, I suspect that my botanical
name-nerd friends will disagree (as they should, given the way the botanical
Code is written).  Again, we might need to allow for seeming redundant
information structure (as discussed above), to separate the nomenclatural
act of a "New combination" event, from the concept assertion act of placing
a species taxon concept within a genus taxon concept.

> -- obviously, sometimes an objectively necessary change in naming
> is required
> because someone previously made a subjective taxonomic
> rearrangement without
> following through with all concomitant name changes. These kinds
> of potential
> downstream dependencies of taxonomy and nomenclature have
> affected the LC/TCS
> territory discussions. Some nomeclatural acts might ultimately
> end up in both
> categories ("normally objective, but in this case linked to a subjective
> analysis"), depending on how far one goes back in the timeline to
> correct a now
> invalid name. Most of the terms listed here
> http://bdei.cs.umb.edu/twiki/bin/view/UBIF/LinneanCoreNomenclaturalStatus
> pertain to the "objective" category.
>
> -- maybe we'll find a better couplet than objective/subjective at
> some point,
> but that is what the Codes use.

I think one of the things that needs to be nailed down once and for all is:
can an LC instance exist without the context of a TCS wrapper?  I would like
very much to say "no"; in which case some of this stuff with important
nomenclatural implications might satisfactorily be represented within the
TCS structure, outside of LC.  However, I may not be in the majority
thinking in this regard...

> (3) "other compliance-achieving acts" - sometimes there is NO new
> name needed
> but something like a neotypification (new type assignment) or
> other "repair
> acts" affect the status of a name.

As I think I said before, one of the only areas of real informational
overalp between LC and the broader (non-LC) parts of TCS is the question of
type specimen designation.  From a name-perspective, that's the only real
connection between "nomenclatural stuff" and the biological world.  I think
Gregor indicated that he had no problem with leaving type designations
outside of LC -- but I'm not sure if this question has been adequately
discussed.

>    We are sort of hoping that this terminology will help with the TCS/LC
> interaction. It is quite clear that a LOT of what the Codes of
> nomenclature
> prescribe ONLY follows if one agrees with a particular and
> subjective taxonomic
> arrangement. In those cases we are dealing with "Code territory"
> that is also
> traditionally "concept territory."

I'm not so sure I agree.  I really believe that the only real area of
conflict is with type-specimen designation.  To me, a "Concept" implies
involvement of more than one individual organism.  Code rules, to my
knowledge, never pertain to more than one individual organism (except in the
case of syntypes, or possibly neotypes if you acknowledge that an original
primary type existed).  But code rules never apply to a scope of individual
organisms that any of us would consider to be a "concept".

> So therefore we think that the
> conceptual/code-prescribed distinction is better than the
> conceptual/nomenclatural distinction.

Whenever I have used the term "nomenclatural", I have meant it in a strict
"code-prescribed" context.  Perhaps this is a point of miscommunication that
has led to much of the prior misunderstanding????

> Of course, Trevor & Jessie have already
> shown that all code-prescribed relationships and acts CAN in principle be
> represented using the concept approach.

Can you point me to this documentation?  I've been out ot the TCS/LC loop
since November, and I'm now trying to get caught up with the developments
since then.

>    Sorry bout the rambling (angel on the right). More later
> (devil on the left).

Please extend my compliments and encouragement to the devil on the left....

:-)

Aloha,
Rich

Richard L. Pyle, PhD
Ichthyology, Bishop Museum
1525 Bernice St., Honolulu, HI 96817
Ph: (808)848-4115, Fax: (808)847-8252
email: deepreef at bishopmuseum.org
http://www.bishopmuseum.org/bishop/HBS/pylerichard.html