[Taxacom] Towards a consensus higher classification oforganisms (was: List of Orders of the world), misspellings, etc...

Richard Pyle deepreef at bishopmuseum.org
Fri Jun 13 22:14:13 CDT 2008

So let me get this striaght.  

For years I have been a subscriber of Taxacom, and for most of those same
years I have seen (and actively participated in) what amounts to the same
(more or less) debates and conversations regarding the distinctions between
taxon names and taxon concepts, the historical problems with both, and
various envisioned solutions to those problems.

And after all of those words, and all of those KB's (MB's) of bandwidth....I
see not one, but *TWO* posts that beautifully capture and articulate these
issues?  In a relatively fresh way, no less?

I'm speechless. Kudos to both DR and JC!

I so desperately want to keep my fingers to myself and off my keyboard and
just revel in the Zen of it all....but I feel a need to comment on one
component common to both posts, which is the word "lists".

Jim focused mostly on "lists" in the sense of the (misleadingly?) named
"Authority" lists -- which I interpret as lists of names that are implied to
represent some sort of "correct" taxonomic concept.  I think Dave was also
talking about "lists" ("indexes") at other, more fundamental stages (i.e.,
lexical indexes, and names index; before we even get to lists intended to
represent concepts).

I think in terms of three classes of "lists" (Dave and I and others have
discussed this before). 

At the most fundamental level is the list of what Dave (and I and others)
refer to as "NameStrings".  The "Strings" bit is computer-geek-ese in
reference to the notion of a "string" of textual characters.  This list has
very little to do with biology.  In fact, the only relevance to biology at
all has to do with sorting out what text-strings get included in the list
("Pomacanthidae"), and which do not ("Automobile").  Alas, this isn't always
obvious (Dave has used the example "Virginia", but I prefer "Phallus"). In
any case, the issues relating to this kind of list involve scope of content
(vernaculars, or not...), formatting (with or without authorships and/or
years, in the case of scientific names), and character encoding (UTF-8).

There is no "authority" here at all -- this is the very rawest of pure
facts. The only wiggle-room for interpretation is which encoded character to
use for certain non-standard glyphs when translating from ink-on-paper to
UTF-8 characters -- surely not an issue that taxonomists have much to say
about (unless they also happen to be computer geeks).

The second class of "lists" in this domain is what most of us think of as
"Taxonomic Names".  These are also not the sorts of things of which Jim
wrote (at least not by my read of his post).  These are "mostly facts", in
the sense that we have codes of nomenclature, which have reasonably
objective rules for how to define and characterize the items in this list,
as well as most of their attributes.  There are some fundamental issues of
non-parity between the various Codes of Nomenclature, and a few grey areas
(think ambiregnal), and certainly some room for interpretation (witness the
recent conversation on the ICZN List regarding a certain publication in
PLoS).  But the truth is, most (nearly all?) of this is also outside the
domain of biology -- more the fodder for name-nerds and Code-warriors (most
of whom happen to be biologists as well).  The only real explicit connection
between this class of list and biology is what Jim says here:

> In taxonomy, the only thing with any real authority is the 
> combination of the type specimen and the protologue 

But Jim later says:

> This is one of the reasons why I worry about lists of names 
> being given authority, and processes like registration that 
> give authority - it discourages or removes the ability to 
> challenge and thus make progress.  

...and this is where we may be somewhat out of phase.  When I think of
"registration", I think of ZooBank.  And in the world of ZooBank, we are
very-much within the realm of the second class of lists as described above.
It does not "[discourage] or remove the ability to challenge and...make
progress", because in this class of "lists", the only "challenges" involve
such pedantry as the precise definitions of certain words (like
"obtainable").  If taxon concepts were being "registered" in some sort of
authoritarian way, then I would be in 100% agreement with Jim on this point.
But in the realm of taxon names (as semi-facts, governed by well-established
Codes), I strongly believe that such a registry (a list) only *enhances* the
ability to challenge and make progress about what I think Jim was really
talking about.

Which brings me to the third class of "lists" in this context -- which is
described variously as "Checklists" or "Taxonomic Authority Files", or (as I
call them) "Meta-Authorities".  These are the lists that bridge the "name"
(class 2) to the circumscription of real-world organisms represented by
those names: that is, the taxon concepts (class 3).

So for this class of "lists", I think I am quite comfortable agreeing with
everything Jim says.  But to return back to Dave's post, we need to remember
that the currency of exchange when discussing taxon concepts (class 3) is
built upon an infrastructure of "names" as governed by the Codes (class 2);
and that those lists of "names" are ultimately derrived from a vast sea of
text-strings (class 1) that exist in some documented form or another
(publications, specimen labels, databases, etc.)

Each class of lists needs to be addressed before we can approach taxonomic
Utopia, and frankly, the sorts of problems and solutions for each class are
quite different from each other. For this reason, I think we need to be much
more careful about understanding which class of list we are talking about
when we have these discussions. A taxonomist may look at a list from class 1
and proclaim "Complete Rubbish!", without fully understanding the
purpose/function and context of the list.  Conversely, some lists (e.g., in
Class 1) are mistakenly portrayed as serving the function of a different
class of list (in which case the proclamations of the aforementioned
taxonomist are justified).

> Our big challenge as taxonomists and biodiversity 
> informaticians is to remove the tacit implication and 
> assumption from what we do and to be more explicit (and dare 
> I say, scientific) in our communication.  At the moment the 
> name implies *a* concept - we need more information to get to 
> *the* concept or tease out the alternative concepts.  We need 
> to get the stage where a particular concept can be 
> communicated unambiguously, or at least where the ambiguity 
> is explicit and unambiguously visible.  In the meantime we 
> will have continue making do with our lists of  'near enough 
> is good enough' scientific names.
> They sort of work, much of the time, but as you indicated, we 
> must be able to do better.

I agree -- but this is not an easy thing to implement (as you well know).
There are some shortcuts, but they are imperfect.  And the hardest part, of
course, is getting the majority of active taxonomists to play along (again,
as you well know).

> See?  All that without any coffee at all...  :)

Dude -- I don't even drink coffee.


Richard L. Pyle, PhD
Database Coordinator for Natural Sciences
  and Associate Zoologist in Ichthyology
Department of Natural Sciences, Bishop Museum
1525 Bernice St., Honolulu, HI 96817
Ph: (808)848-4115, Fax: (808)847-8252
email: deepreef at bishopmuseum.org

More information about the Taxacom mailing list