[Taxacom] Read... and believe...

[Stephen Thorpe]

>And most taxonomists know with sufficient precision which of several published alternative concept 
>definitions for a species name most closely reflects the one they had in mind
Not convinced! If John Smith identifies N.Z. beetles, very few of which have been revised since the pioneering work of Thomas Broun, and I say "John, get your ass down here ... [I probably wouldn't say "ass"!], what did you have in mind when you called this Scraptogetus anthracinus???" Very likely, John won't have seen the type, and won't even have bothered to look at Broun's inadequate original description (and there are no redescriptions). More likely he would say something like "I compared it to specimens in such-and-such collection". Quite likely, that collection might have had summer students transferring things from old-style boxes to new cabinets, and making mistakes! Perhaps the specimens standing under that name (but without individual det. labels) were a mixture of several unrelated things. Perhaps someone will notice at some point and try to fix it up in whichever way they think it should be. The "Croftian pile" (not to be confused with the "Pylean pile"!) will change randomly over time, and never be recorded.

True story: an agency wanted an official "rubber stamped" identification from an official "expert". They sent someone to him with the specimen. It belonged to a subgroup of the wider group for which he is an "expert" - a subgroup for which he has no interest or knowledge. According to the messenger, the "expert" looked at it, said "what the f$%# am I supposed to do with this?", Googled up an image, said "looks like it", rubber stamped the identification, and got paid!

Take home message: maybe sometimes we can meaningfully specify a taxon concept with an identification, but reality is mostly far messier than that.

Stephen

PS: See attached (I don't think the attachment worked last time?)



________________________________________
From: Richard Pyle [deepreef at bishopmuseum.org]
Sent: Sunday, 6 September 2009 5:00 p.m.
To: Stephen Thorpe; 'Jim Croft'
Cc: TAXACOM at mailman.nhm.ku.edu
Subject: RE: [Taxacom] Read... and believe...

> What is true is that everybody who identifies any specimen
> does so SOMEHOW, with reference to SOMETHING, but whatever it
> is - call it "taxon concept", or whatever you like

Yes, please -- call it a taxon concept.  That's what people have called it
for a long time.  It's one of the few terms in our sphere that actually
means mostly the same thing to most people.  By "our sphere", I mean the
people who have been thinking about this stuff for a long time.  I started
in the early 1990's, and compared to a number of people on this list (some
of whom actually post, others lurk, others were part of founding the list in
the first place), I'm a newbie.

> - it very
> often isn't something that you can codify or even express in words.

No, but you can anchor it to a tangible entity -- that is, a publiushed
usage of the name.  And most taxonomists know with sufficient precision
which of several published alternative concept definitions for a species
name most closely reflects the one they had in mind.  So, when I identify a
specimen as "Centropyge fisheri sec. Randall 1995", at least I know how
Randall 1995 treated the other (potential) synonyms of this species, and
have a reasonably good representation of the scope of organisms that fall
within Randall's 1995 concept of C. fisheri.

> To take Richard's example of Centropyge, if John Smith
> identifies a specimen as C. fisheri in 2003, without any
> elaboration, the "taxon concept" is just "Centropyge fisheri
> det. J. Smith 2003",

Yep.  In fact, to accommodate these examples (which are the vast majority),
my data model allows specimen determinations to be an anchor-point for an
implied taxon concept.  We may no nothing about it from our database, but at
least it establishes a "thing" that can be mapped to other, more explicitly
published taxon concepts.  For example, if in 2009 I go back to John Smith
and say, "Yo, John [I probably wouldn't say 'Yo'] -- remember when you
identified that speciemn of C. fisheri? Well, can you remember whether you
were thinking of the species more the way Randall 1995 defined it, or more
the way Jones 1980 defined it?"  If he says "Sorry, bro [he probably
wouldn't say 'bro'] -- but I can't remember."; then at least I know there
was a concept in his mind, and the identification gets anchored to a record
for "C. fisheri sec. Smoth 2003".  But if I'm lucky, and John's memory is
better than mine, he might say "What?!? I'd never follow that
motherscratcher Jones! [I actually have no idea what "motherscratcher"
means, and I doubt John would say it]"  Then I can map "C. fisheri sec.
Smith 2003" as congruent with "C. fisheri sec. Randall 1995"  And because
Randall providced a much richer concept definition, I'll be able to map it
appropriately to all the other well-defined taxon concepts for that name.

> Maybe he has
> forgotten this, but the neurological connections in his brain
> which lead to identification remain intact!

No argument.  As I said, this is the source of the problem we're discussing.
And why I always plead (beg?) for people to take a moment and note in what
sense of the name they were identifying it -- just a simply "sensu Randall
1995" is probably all we would need to fill in the gaps.

Aloha,
Rich